Racial/Ethnic Differences in Traumatic Brain Injury: Pathophysiology,Outcomes,and Future Directions

Pathophysiology of TBI

TBI occurs when there is sufficient external force that the homeostatic mechanisms of the brain are disrupted allowing for immediate and delayed destruction of both brain tissue and neuronal networks. The immediate phase of injury, called the primary injury, occurs at the exact moment of the insult. ^{1,2,21 -23} Depending on the extent of the precipitating event, the damage may be confined to physical destruction of brain tissue. The delayed phase, or secondary injury, involves a complex cascade of mechanisms and reactions that occur in the hours, days, or weeks following the initial insult. ^{1,2,21 -23} Secondary injuries are much more complex than primary injuries and have a more diverse array of outcomes. While specific mechanisms, molecules, and pathways continue to be investigated and revised, there does appear to be a general consensus regarding the major factors and mechanisms involved in the pathogenesis of TBI.

Physical stretching of the neuronal membranes occurs following the initial insult, which allows for an influx of excitatory ions such as sodium (Na⁺) and calcium (Ca²⁺). Pathologic accumulation of intracellular Ca²⁺ induces the release of glutamate with subsequent autocrine activation of NMDA receptors on the neuron, ²¹ which allows for further influx of Ca²⁺ into the neuron. The elevated intracellular Ca²⁺ concentration also activates various proteases and endonucleases, which promote apoptosis of the neuron. Elevated levels of intracellular Ca²⁺ activate the Ca²⁺-dependent permeability transition ion pore located on the inner mitochondrial membrane. ^22,24 Prolonged activation of the permeability transition pore disrupts normal mitochondrial activity, including ATP production, and allows for the release of pro-apoptotic molecules into the cytosolic space, triggering the intrinsic pathway of apoptosis. ^21,22,24,25 Excess glutamate within the synaptic cleft has also been linked to overstimulation of surrounding neurons via NMDA and AMPA receptors. ^{22,26 –29} Glial cells such as astrocytes, which are normally responsible for maintaining homeostasis in the brain, also play a major role in the pathogenesis of TBI. Reactive gliosis is a process in which astrocytes hypertrophy and release cytokines and chemokines, promoting scar formation. Therefore, the excessive release and binding of glutamate to both NMDA and AMPA receptors has an excitotoxic effect on these cells resulting in cell death and necrosis. ^{2,21,22,26 –32}

The blood–brain barrier (BBB) has one of the most, if not the most, powerful influences on the brain micro-environment and contributes significantly to healthy neuronal function. ^2,33,34 The BBB is a heavily regulated complex of cells and cerebral capillaries that function to prevent toxins, metabolic waste, and even inflammatory cells from entering the brain parenchyma. ^2,21,22 Fundamentally, the cerebral capillaries are composed of endothelial cells connected to one another via tight junctions in a continuous, non-fenestrated arrangement. ^2,35 The endothelial cells possess a number of features that allow them to metabolize potentially harmful substances and prevent harmful substances from being transported across the basement membrane. ^2,33 The basement membrane of these cells is attached to pericytes, which are modified vascular smooth muscle cells responsible for regulating endothelial cell proliferation and differentiation, immune surveillance, and management of endothelial cell permeability. ^2,36,37 Attached to the pericytes are the foot processes of astrocytes, which are specialized glial cells that support both neurons and endothelial cells by maintaining important repair processes (i.e., biochemical, metabolic, nutritional, ionic, and repair). ^{2,33,37 -39}

Although mostly impermeable, when operating effectively, the BBB serves as a physical, metabolic, and transport barrier. ^2,33,34 However, when disrupted, any one of the components of the BBB can become compromised, allowing for the free passage of molecules, such as pro-inflammatory cytokines, into the micro-environment of the brain. Disruption of the BBB is extraordinarily multi-faceted and can be associated with several mechanisms. Initial injury can cause direct damage to cerebral vasculature, stimulating the coagulation cascade and inflammatory processes. ^2,40 Damage to pericytes results in increased vascular permeability and allows for circulating immunoglobulins to migrate into the brain parenchyma, perpetuating ongoing inflammation. ^2,41,42 As previously described, brain injury also results in excessive release of glutamate and the dysregulation of Ca²⁺ and Na⁺. ^2,43,44 The combination of neuronal death, reactive gliosis, and pericyte damage results in ischemia, alterations in ATP production, and release of reactive oxygen species (ROS) and nitric oxide (NO)—NO has been shown to negatively affect endothelial tight junctions, ultimately leading to proteolytic cleavage. ² Essentially, the breakdown of the BBB occurs following conditions, including the activation of proteases, membrane lipid peroxidation, mitochondrial damage and dysfunction, and cleavage of the extracellular matrix. ^2,42

The purpose of this study was to review the current literature on TBI and determine whether there is evidence of racial/ethnic differences and/or disparities in TBI in the U.S. Special consideration was given to mechanism of injury, injury severity, treatment/rehabilitation, and outcomes. Importantly, this paper also examined the literature for potential evidence of underlying racial/ethnic biological differences in TBI pathophysiology. Lastly, our final goal was to identify future directions in research regarding the impact of race/ethnicity on TBI.

Racial/ethnic differences in TBI mechanism of injury

Six studies were identified that found significant racial/ethnic differences in TBI mechanism of injury. ^{4,45 -49} Specifically, B/AA are more likely to incur a TBI from third party assault/violence. In one sample of 803 patients with TBI, 64% of those with violence-related injuries were B/AA while 68% of those with nonviolence-related injuries were NHW. The majority of the violence-related injuries (56%) were blunt assault with 23% being penetrating injuries. ⁴⁹ Hanks and colleagues ⁴⁷ performed a logistic regression analysis on TBI injury etiology (violent vs. nonviolent) and additional covariates, including race, and found that minorities were 3.1 times more likely to have a violent TBI compared with Whites; B/AA comprised 86% of the minority group. Burnett and colleagues ⁴⁸ found that MVC's were the predominate mechanism of injury for both minorities and non-minorities; however, minorities were more likely to develop a TBI from violence and auto-versus-pedestrian collisions. Arango-Lasprilla and colleagues ⁴⁶ performed a retrospective study of 4929 patients with moderate to severe TBI and found that B/AA were four times more likely to obtain a TBI from violence compared with NHWs. More recently, Bertisch and colleagues ⁴⁵ examined a sample of 399 individuals with either assault or self-inflicted firearm-related TBI and found that victims of assault-related TBI were more likely to be B/AA.

Daugherty and colleagues ⁴ analyzed trends in TBI-related deaths by race/ethnicity and sex between 2000 and 2017. Injuries were categorized by intent and mechanism. From 2006 to 2017, Daugherty and colleagues ⁴ observed a 32% increase of intentional self-harm among NHWs, while homicide was the most common cause of TBI-related death among B/AAs. The most common cause of TBI-related death among AI/AN, who consistently had the highest number and rate of TBI-related death of any racial or ethnic group, from 2000 to 2017 was unintentional motor vehicle injuries. The major causes of TBI-related death among Hispanics from 2000 to 2008 was unintentional motor vehicle crashes. However, from 2012 to 2017, unintentional falls had become the primary mechanism for TBI-related deaths in the Hispanic population. ⁴

Racial/ethnic differences in TBI injury severity

Ten studies examined potential differences by race/ethnicity in TBI injury severity with nine studies showing that minorities had either similar or less severe injuries compared with NHW (Table 1). ^{46,47,49 –54} Rosenthal and colleagues ⁵⁰ found that minorities (B/AA, Hispanics) had slightly lower post-traumatic amnesia and higher Glasgow Coma Scale (GCS) scores compared with NHW indicating less severe injuries; these findings did not reach statistical significance. Similarly, Hart and colleagues ⁵¹ also observed lower post-traumatic amnesia scores among B/AA compared with NHW, although this finding did not reach statistical significance either. Burnett and colleagues ⁴⁸ observed less severe injuries among minorities compared with NHW on admission to acute inpatient rehabilitation; similar to the previous two studies, this result did not achieve statistical significance. Harrison-Felix and colleagues ⁴⁹ collected three separate GCS scores as measures for TBI injury severity: On admission to the emergency department, the highest GCS score within 24 h of injury, and the lowest GCS score within 24 h of injury. All three measurements were found to be higher among patients in the violence-related injury group, comprised predominantly of B/AA, compared with the non-violence related injury group, comprised predominantly of NHW. ⁴⁹ The GCS scores taken on admission to the emergency department and the highest GCS within 24 h of injury achieved statistical significance. ⁴⁹ Neither Staudenmayer and colleagues ⁵² nor Arango-Lasprilla and colleallgues ⁴⁶ found any significant differences by race/ethnicity in TBI injury severity.

In contrast to the prior studies, two studies were identified that found evidence of racial/ethnic differences in TBI injury severity. Dismuke and colleagues ⁵⁵ examined a national sample of 132,995 veterans with combat- and non-combat-associated TBI for racial/ethnic differences. Findings revealed that there was a greater percentage of moderate to severe TBI in B/AA and Hispanics compared with NHW. B/AA and Hispanics also were found to have had higher odds of moderate to severe TBI compared with NHW. Richie and colleagues ¹⁷ found that minorities had significantly more moderate and severe TBI compared with NHW.

Racial/ethnic differences in TBI treatment

Three studies were found that revealed no significant racial/ethnic differences in the acute care phase of TBI rehabilitation. ^{56 -58} In a study on the impact of race/ethnicity on emergency care, Wall and colleagues ⁵⁶ measured the probability of receiving a head CT for evaluation of a blunt head injury. While minority patients (B/AA, Hispanic, Asian/Pacific Islander) had a probability of receiving a head CT 0.84 times as high as NHWs, this result was not statistically significant. During secondary analyses, Hispanic and Asian patients were found to be less likely to receive a head CT compared with NHWs, although the researchers cited a lack of power in their study to detect differences in racial/ethnic subgroups. Shafi and Gentilello ⁵⁷ performed a multi-center study encompassing 406 hospitals to investigate the impact of patient ethnicity on initial ED assessment. There were no significant differences between NHW and minorities (B/AA and Hispanics) in the intensity of ED assessment, ED monitoring, treatment, or emergency department disposition. ⁵⁷

Sharma and colleagues ⁵⁸ analyzed rates of neuromonitoring and neurointerventions in a sample of 1064 patients from level 1 and 2 trauma centers with severe TBI from blunt trauma. Results showed that there were no significant differences in rates of neuromonitoring, neurosurgical interventions, or mortality between NHW and non-White patients (American Indian, Asian, B/AA, Other). The non-White group was then disaggregated and additional analyses comparing each individual racial group were performed. No significant differences were observed in rates of neuromonitoring, neurosurgical intervention, or mortality. It was concluded that while racial and ethnic disparities do exist and impart a powerful influence over functional outcome, those differences were not due to disparities in the acute care phase of TBI management. ⁵⁸

One study was identified that did reveal significant differences between races/ethnicities in the acute care phase of TBI management (Table 1). Bazarian and colleagues ¹⁰ examined the role of race (B/AA, Native Hawaiian/Other Pacific Islander, AI/AN) and ethnicity (Hispanic or Latino, not Hispanic or Latino, blank) in emergency care/treatment of minorities with mild TBI. The results indicated that Hispanics waited a significantly longer time to see a physician and had a higher likelihood of leaving before being seen compared with non-Hispanics. The authors suggested that these longer wait times were due to the need to find an interpreter. B/AAs were significantly more likely to receive care from an ED resident and significantly less likely to be cared for by an ED physician compared with NHWs. Bazarian and colleagues ¹⁰ made a number of suggestions that could potentially explain this finding, including possible “cherry-picking” by predominantly NHW ED residents of cases involving B/AAs that appeared easier and subject to less scrutiny. B/AAs were also significantly less likely to be returned to a referring physician. There were no significant differences observed among such variables as rate of CT scanning, hospital admission, or receiving analgesics.

Twelve studies were found in the literature that showed minorities have poorer quality and/or quantity of rehabilitative services (Table 1). ^{10,11,14,16,20,48,52,59 –63} In one sample of 2020 patients with moderate to severe TBI enrolled in a Traumatic Brain Injury Model Systems rehabilitation program, Burnett and colleagues ⁴⁸ examined both the quantity and intensity of four modes of therapy: Physical, occupational, speech, and neuropsychological. Analysis revealed that minorities received significantly less time in all four therapy services and significantly less intense occupational, speech, and vocational therapy. More recently, Meagher and colleagues ¹⁴ performed a retrospective analysis on adults with moderate to severe TBI to assess for racial and ethnic disparities in discharge to rehabilitation. In contrast to prior studies, Meagher and colleagues ¹⁴ defined discharge services along a spectrum (i.e., home with no services, home health with outpatient or in-home services, skilled nursing facility, and inpatient rehabilitation). The results revealed that B/AAs and Hispanics were significantly less likely to be discharged to the highest level of rehabilitation (inpatient rehabilitation) compared with NHWs. The authors noted that this was the first study to show that B/AAs and Hispanics are less likely to be discharged to a “concordant level of rehabilitation” than matched NHWs. ¹⁴ This was also the first study to show the persistence of racial/ethnic disparities in older adults with TBI and uniform coverage—that is, the researchers were able to account for the effect of insurance status on rehabilitation discharge. ¹⁴

Chang and colleagues ⁵⁹ performed a retrospective analysis on a sample of 9240 patients age 65 and older to assess the association between race/ethnicity and discharge destination. B/AA were significantly less likely to be discharged to an assisted living facility compared with NHW. Further, older Hispanics and B/AA were significantly more likely to be discharged home compared with NHW. These findings are in contrast to those of Vadlamani and colleagues, ⁶⁴ who found that B/AA were more likely to be discharged to inpatient rehabilitation compared with NHW. Fuentes and colleagues ⁶³ performed a secondary analysis of an existing data set of 111 patients admitted to an urban public hospital for acute TBI rehabilitation. Results showed that NHW were admitted to acute TBI rehabilitation significantly faster than minorities.

Three studies were identified that contradicted the findings of the previous studies (Table 1). ^53,64,65 One study on the effect of race/ethnicity and English-language proficiency on receiving rehabilitation services, de la Plata and colleagues ⁶⁵ found no significant impact of race/ethnicity. However, the researchers did observe that race/ethnicity was a predictor of worse functional outcome.

Racial/ethnic disparities in TBI outcomes

Twenty-six studies found evidence of significant racial/ethnic differences in TBI outcome (Table 1). Minorities were commonly found to score worse on at least one measure of TBI functional outcome compared with NHW. ^{46,47,52,54,66 –71} Functional outcome is a broad term that encompasses multiple dimensions of patient recovery. However, it can be thought of as the ability of the patient to perform various motor, cognitive, and social activities independently following a traumatic injury. For example, Hanks and colleagues ⁴⁷ measured long-term functional outcomes of minorities (B/AA, Hispanic, Asian/Pacific Islander, Other) upon discharge from an acute rehabilitation center 1 year and 2 years following a violent TBI. Compared with NHWs, minorities had significantly worse scores on measures of functional outcomes such as the Functional Independence Measure (FIM), Disability Rating Scale (DRS), and the Community Integration Questionnaire (CIQ). The FIM is an 18-item ordinal scale that measures independent performance on activities related to daily living, including self-care, locomotion, communication, social cognition. ^67,72 The FIM is one of the most widely accepted instruments for measuring patient functional outcomes in a rehabilitative setting. ^67,72,73 Community integration can be broadly defined as the return to typical family and community activities. ⁷² The CIQ is a 15-item questionnaire that measures a patient's ability to reintegrate in the home setting, in the social setting, and in productive activities following a TBI. ^67,72 Patients are scored in each category primarily based on the frequency with which they perform each activity or role.

Lastly, the DRS measures a patient's level of impairment (eye opening, communication ability, motor response), disability (feeding, toileting, and grooming), and handicap (level of functioning and employability) over the course of their recovery in a rehabilitation setting. ⁷² Similar to the CIQ, the scores on the DRS also range from 0 (no disability) to 29 (extreme vegetative state). Hanks and colleagues ⁴⁷ found that at discharge from an acute rehabilitation center, at 1-year post-injury, and at 2 years post-injury, minorities has less functional independence, higher rates of disability, and poorer community integration particularly among productivity subscale.

Staudenmayer and colleagues ⁵² examined a sample of 211 patients in acute care hospitalization at two different trauma centers in order to determine specific functional domains in which minorities (B/AA, Asian, Hispanic) experience worse outcomes compared with NHWs. Here, the researchers utilized the Functional Status Examination, which measures patient activity and participation in 10 domains, including personal care, ambulation, work or school activity, home management, leisure and recreation, travel, social integration, standard of living, financial independence, and executive functioning. ⁷² Patients are scored in each category on an ordinal scale of 0 (no change from pre-injury) to 3 (completely dependent/does not participate at all). Although minorities were just as likely to be discharged from acute care rehabilitation as NHWs, they were more likely to score in the worse percentile of all 10 domains. Minorities scored significantly worse among measures of standard of living, leisure activities, and return to work or school. ⁵² Shafi and colleagues ⁶⁹ made a similar observation in their study on whether there are racial/ethnic disparities in trauma care particularly with rehabilitation services and long-term functional outcome. The authors observed that, minorities were significantly more likely to have moderate to severe disability despite similar access to rehabilitation services as NHWs.

In a thoroughly executed review on the racial and ethnic disparities in functional, psychosocial, and neurobehavioral outcomes following TBI, Arango-Lasprilla and Kreuzter61 identified nine studies between 1996 and 2007 that focused primarily on disparities in functional outcomes. Seven of these studies ^{46,47,52,66 -68,74} observed worse functional outcomes in minorities compared with NHWs. Six of these seven studies met the criteria for inclusion in the current paper. One such study was performed by Arango-Lasprilla and colleagues ⁶⁷ and explored the relationship between Hispanic ethnicity and functional outcome following rehabilitation for moderate to severe TBI. Utilizing the FIM, DRS, CIQ, and the Extended Glasgow Outcome Scale (GOS-E), measurements were obtained upon admission, at discharge, and at the 1-year follow-up mark from inpatient rehabilitation. While there were no significant differences between Hispanics and NHWs on admission and discharge, at the 1-year follow up, Hispanics were significantly more likely to score in the worst percentile on the FIM, DRS, CIQ, and the GOS-E, illustrating lower cognitive and physical functioning and community integration. ⁶⁷ Arango-Lasprilla and colleagues ⁴⁶ performed a similar study using the same measurements (FIM, DRS, CIQ, and GOS-E), this time investigating the differences in functional outcome between minorities (B/AA and Hispanic) and non-minorities (NHW) upon admission, discharge, and 1 year following inpatient rehabilitation. Arango-Lasprilla and colleagues ⁴⁶ found that at 1-year follow up minorities were significantly more likely to fall in the worst quartile on all four measures compared with NHWs despite both groups being clinically similar on admission and discharge.

Warren and Garcia ⁷¹ analyzed the data of 41,847 patients with a diagnosis of TBI admitted to inpatient rehabilitation facilities in the U.S. between 2005 and 2016. The results of the secondary analysis revealed that, on discharge, B/AA race was significantly associated with a 1-2-point lower score on motor, cognitive, and total FIM scores compared with NHW. On admission, however, B/AA, Hispanics, and Asians, were significantly associated with a 1-3 point lower motor, cognitive, and total FIM score compared with NHW. In a different sample of 7953 patients with severe TBI between 2000 and 2016, Asians and Hispanics had lower functional outcomes at 1-year post-injury compared with NHW. ⁵⁴ Further, although Asians had the lowest injury severity on admission to inpatient rehabilitation, these patients had less improvement compared with that made by Hispanics and NHW at 1-year follow-up. ⁵⁴

In contrast to the previous studies, Burnett and colleagues, ⁷⁵ using the FIM and DRS, did not find any significant differences in disability or functional independence between minorities and NHWs. A similar finding was observed in an earlier study by Rosenthal and colleagues, ⁵⁰ who looked at the impact of race/ethnicity on both functional outcome and community integration. Using the FIM, DRS, and CIQ, Rosenthal and colleagues ⁵⁰ did not find any significant differences between minorities (B/AA, Hispanic, Asian) and NHWs in functional independence (FIM) or disability level (DRS). However, minorities were observed to perform significantly worse on the productivity and social integration subscales of the CIQ as well as the total CIQ score. The productivity subscale measures the extent to which a person engages in daily employment, volunteering, or school activities. The social integration scale focuses on the extent to which the individual is engaged in social behaviors within a community setting. This latter finding that minorities have significantly worse community integration following TBI has been observed in a host of additional studies (Table 1). ^{46,47,50,51,60,61,67,70,76}

In their review, Arango-Lasprilla and Kreutzer ⁶¹ were able to identify eight articles ^{46,47,50,51,66,67,70,76} that measured community integration among racial/ethnic minorities following TBI, all of which reported worse outcomes for minorities. In addition to objective measures of community integration, there were also findings that minority status predicted dissatisfaction with community, civic, life, and leisure. ⁷⁶ One study by Sander and colleagues ⁷⁰ examined the impact of race/ethnicity and income on community integration 6 months following mild-to-severe TBI. The results showed that B/AAs had lower total CIQ scores compared with NHWs, and both B/AAs and Hispanics had lower scores compared with NHWs on the home integration scale of the CIQ. ⁷⁰ Lastly, B/AAs scored lower than both NHWs and Hispanics on the CIQ productivity scale.

In additional to poorer functional outcomes, minorities have been shown to have increased mortality risk (Table 1). ^{17 –20,77,78} A recent retrospective study performed by Richie and colleagues ¹⁷ examined the relationship between race and in-hospital mortality of 6352 patients enrolled in the Trauma One Database at Oregon State University Hospital. The researchers observed a statistically significant greater in-hospital mortality among minority patients (B/AA, AI/AN, Asian, Hispanic/Latino, Pacific Islander/Native Hawaiian, and Other) compared with NHW patients. One of the reasons posited for this finding was the number of minority patients that presented with severe injuries; contrary to prior studies, there was a significant difference in severity between minorities and NHWs with minorities having more severe injuries. In a sample of 17,957 severe TBI patients obtained from the Los Angeles County Trauma System, Berry and colleagues ¹⁹ found that Asians had a significantly higher mortality risk when compared with NHWs. Further, Asians with insurance had a significantly higher risk of mortality compared with insured B/AA and insured Hispanics. Of note, B/AAs and Hispanics were not found to have increased mortality risk compared with NHWs. In a study analyzing 56,482 moderate-to-severe TBI patients from the National Trauma Data Bank found that both B/AAs and Asians had significantly increased risk of death compared with NHWs. ²⁰ Lastly, AI/AN have been shown to have the highest rates and numbers of TBI-related death compared with any other racial/ethnic group (Table 1). ^{4,9,79 –86}