Skip to main content
NCBI home page
Neurology logoLink to Neurology
. 2022 Aug 2;99(5):e521–e535. doi: 10.1212/WNL.0000000000200602

Sensory Phenotypes for Balance Dysfunction After Mild Traumatic Brain Injury

Peter C Fino 1,✉, Leland E Dibble 1, Elisabeth A Wilde 1, Nora F Fino 1,✉, Paula Johnson 1, Melissa M Cortez 1, Colby R Hansen 1, Susanne M van der Veen 1, Karen M Skop 1, J Kent Werner 1, David F Tate 1, Harvey S Levin 1, Mary Jo V Pugh 1, William C Walker 1
PMCID: PMC9421603  PMID: 35577572

Abstract

Background and Objectives

Recent team-based models of care use symptom subtypes to guide treatments of individuals with chronic effects of mild traumatic brain injury (mTBI). However, these subtypes, or phenotypes, may be too broad, particularly for balance (e.g., vestibular subtype). To gain insight into mTBI-related imbalance, we (1) explored whether a dominant sensory phenotype (e.g., vestibular impaired) exists in the chronic mTBI population, (2) determined the clinical characteristics, symptomatic clusters, functional measures, and injury mechanisms that associate with sensory phenotypes for balance control in this population, and (3) compared the presentations of sensory phenotypes between individuals with and without previous mTBI.

Methods

A secondary analysis was conducted on the Long-Term Impact of Military-Relevant Brain Injury Consortium—Chronic Effects of Neurotrauma Consortium. Sensory ratios were calculated from the sensory organization test, and individuals were categorized into 1 of the 8 possible sensory phenotypes. Demographic, clinical, and injury characteristics were compared across phenotypes. Symptoms, cognition, and physical function were compared across phenotypes, groups, and their interaction.

Results

Data from 758 Service Members and Veterans with mTBI and 172 individuals with no lifetime history of mTBI were included. Abnormal visual, vestibular, and proprioception ratios were observed in 29%, 36%, and 38% of people with mTBI, respectively, with 32% exhibiting more than 1 abnormal sensory ratio. Within the mTBI group, global outcomes (p < 0.001), self-reported symptom severity (p < 0.027), and nearly all physical and cognitive functioning tests (p < 0.027) differed across sensory phenotypes. Individuals with mTBI generally reported worse symptoms than their non-mTBI counterparts within the same phenotype (p = 0.026), but participants with mTBI in the vestibular-deficient phenotype reported lower symptom burdens than their non-mTBI counterparts (e.g., mean [SD] Dizziness Handicap Inventory = 4.9 [8.1] for mTBI vs 12.8 [12.4] for non-mTBI, group × phenotype interaction p < 0.001). Physical and cognitive functioning did not differ between the groups after accounting for phenotype.

Discussion

Individuals with mTBI exhibit a variety of chronic balance deficits involving heterogeneous sensory integration problems. While imbalance when relying on vestibular information is common, it is inaccurate to label all mTBI-related balance dysfunction under the vestibular umbrella. Future work should consider specific classification of balance deficits, including specific sensory phenotypes for balance control.

While commonly considered a transient injury, mild traumatic brain injuries (mTBIs) can have long-lasting effects; more than 53% of individuals who present to an emergency department after their mTBI report functional limitations at 1 year postinjury.1 In military populations, persisting symptoms are similarly evident, with high rates of persistent post-traumatic stress, somatic and affective symptoms, imbalance, and cognitive difficulties.2,3 Care for these persisting symptoms can exceed $60 billion annually,4 with untreated or undertreated problems leading to lifelong mTBI comorbidities.5

Despite recognition of a common constellation of symptoms reported by many individuals with TBI, heterogeneity in the mechanism, pathophysiology, anatomic focus, and severity of injury may complicate the diagnosis, prognosis, and treatment. Recent consensus and position statements acknowledge the heterogeneity of mTBI and encourage clinicians and researchers to consider the unique confluence of symptoms in each individual.6,7 Toward this goal, models of care—based on symptom subtypes or clusters—have been introduced to guide targeted treatment strategies.8,9 However, even subtypes of mTBI can overgeneralize. For instance, individuals with balance problems are often triaged into the vestibular subtype despite potential heterogeneous origins of balance problems and the potential contribution of nonvestibular systems.8,10

Standing balance, defined as the ability to control the motion of the body's center of mass within the base of support, is accomplished through the integration of visual, vestibular, and proprioceptive information and the selection of appropriate motor outputs to control postural sway.11 This sensory integration and motor transformation process flexibly adjust the weighting of sensory information based on intrinsic and extrinsic factors. For example, patients with vestibular loss become more reliant on proprioceptive and visual sensory information for balance control.12 Further loss of vision, such as standing in a dark room, elicits a dominant role of proprioceptive information for balance control.

Persistent balance deficits in individuals with mTBI, particularly military Service Members, exhibit marked heterogeneity despite previous reports implicating inappropriate visual or vestibular processing.13-15 Not all individuals who report balance problems after mTBI have abnormal vestibular processing nor do all with abnormal vestibular function for balance control have abnormal visual function.16 In addition, subsets of those with either type of abnormalities may not exactly overlap. Such heterogeneity in people with mTBI suggests that the combinations of normal and abnormal sensory processing for balance control constitute unique sensory phenotypes (e.g., vestibular dominant; proprioceptive impaired; visual dominant; vestibular impaired; etc.) for balance control similar to neurobehavioral phenotypes.17 Furthermore, these sensory phenotypes for balance control may have clinical significance. Rather than a single mechanism underlying all balance deficits, it is possible that unique presentations of balance dysfunction originate from distinct, injury-specific neuroanatomical structures or networks, which are uniquely expressed symptomatically and have distinct clinical needs. Unfortunately, an uncommonly large sample size with quantitative posturography is required to characterize sensory phenotypes for balance deficits after mTBI. While previous investigations examined sensory ratios in a moderately sized sample,15,16 a larger sample is necessary to determine nonoverlapping, multisystemic sensory phenotypes.

To gain insight into sensory phenotypes contributing to imbalance after mTBI, our objectives were to (1) explore whether a dominant sensory phenotype exists in the chronic mTBI population, (2) determine the clinical characteristics, symptomatic clusters, functional measures, and injury mechanisms that associate with sensory phenotypes for balance control in this population, and (3) compare the presentations of sensory phenotypes between individuals with previous mTBI with those without (non-mTBI controls). To accomplish these objectives, we leveraged the Long-Term Impact of Military-Relevant Brain Injury Consortium—Chronic Effects of Neurotrauma Consortium (LIMBIC-CENC)18 to examine sensory phenotypes for balance control in current and former Service Members with chronic mTBI. We expected that individuals with mTBI would not exhibit uniform sensory deficits/preferences for balance control—thereby providing further evidence of the heterogeneity of balance deficits and the lack of a dominant sensory phenotype for balance control after mTBI. Furthermore, we expected that different phenotypes would associate with different clinical characteristics, symptom clusters, functional measures, and mechanistic presentations in individuals with mTBI and individuals with mTBI would exhibit more severe symptomatic and functional deficits compared with individuals without mTBI in the same phenotype.

Methods

Standard Protocol, Approvals, Registrations, and Patient Consents

All study procedures and analyses were approved by institutional review boards at each study center in accordance with federal regulations. All participants provided informed written consent before participating.

Participants

This study is a secondary analysis of data from the LIMBIC-CENC Prospective Longitudinal Study (PLS) of the late neurologic effects of mTBI among previously combat-deployed military Service Members and Veterans. More information on the background, breadth, and objectives of the overarching study are provided elsewhere. The methods pertaining to this study were previously reported in an interim analysis.19 In brief, participants were recruited from post-9/11 era Service Members and Veterans who were deployed in a combat zone. In 2020, eligibility was widened to all combat eras. Their lifetime mTBI histories varied from none (18%) to multiple mTBIs. Data used in the analyses were collected during baseline visits for participants who were enrolled between 2013 and October 2020 at 8 PLS enrollment sites across the United States. Participants were recruited through mailing campaigns and through advertisements, flyers, outreach events, and clinician referrals. Exclusion criteria for the larger study were as follows: (1) a history of moderate or severe TBI as defined by an initial Glasgow coma scale <13, a coma duration >0.5 hours, or post-traumatic amnesia longer than 24 hours, (2) a history of major neurologic disorder (e.g., stroke and spinal cord injury), and (3) a history of major psychiatric disorder (e.g., schizophrenia), with major defined as resulting in a significant decrement in functional status or loss of independent living capacity. Of note, common mental health comorbidities such as post-traumatic stress disorder (PTSD) and depression were not exclusions.

An additional inclusion criterion for this analysis was PLS enrollment at one of the sites with capability to conduct computerized posturography testing (see further below). Exclusion criteria specific to this analysis were as follows: (1) errors during the posturography testing resulting in unreliable data, (2) blindness or severe vision problems, or (3) pregnancy. A large proportion of the sample reported health problems that required special health equipment (e.g., hearing aids, canes, and walkers). Because of the size of this group (261 individuals), we performed a sensitivity analysis excluding them. The overall results were consistent regardless of their inclusion (see eTables 1–5 and eFigures 1–4, links.lww.com/WNL/C37). Therefore, the reported use of specialized health equipment was not an exclusion criterion.

Participants were stratified by those with a positive history of mTBI (1+ lifetime mTBIs) and those without a history of mTBI (non-mTBI, 0 lifetime mTBIs). The LIMBIC-CENC PLS study uses a standardized structured interview that first screens participants for all lifetime potential concussive event (PCE) exposure using a modified Ohio State University TBI Identification instrument. Next, each PCE is further evaluated using the Virginia Commonwealth University Retrospective Concussion Diagnostic Interview to arrive at a provisional determination of the presence or absence of TBI through an embedded algorithm using the fully structured items that are based on the DoD/VA common definition of mTBI.20 The algorithm diagnosis is vetted against the free-text open-ended content and any acute medical record documentation. A central diagnosis committee is the final arbitrator for any questionable diagnoses.

Performance and Symptom Validity

As part of the larger study, all participants were screened with a performance validity measure to detect suboptimal effort on cognitive testing, the Medical Symptom Validity Test (MSVT).21 The Neurobehavioral Symptom Inventory (NSI) Validity-10 index and the mild brain injury atypical symptom (mBIAS) measures were used to detect symptom overreporting.22 As noted in the consort diagram, individuals with scores greater than the standard cutoffs for this population on symptom reporting (i.e., mBIAS and NSI Validity-10 index) were excluded from analysis. However, given the focus on balance rather than cognition, we elected not to exclude participants from this analysis based on their performance on the MSVT but do include data on the failure rates using standard cutoffs in this population in the summary tables.

Sensory Phenotype Determination

Balance control was quantified using the sensory organization test (SOT) protocol on the NeuroCom Smart Balance Master (NeuroCom; Natus Medical Inc., San Carlos, CA).23 The SOT represents a specific protocol of computerized dynamic posturography that quantifies balance under 6 different conditions with varying sensory information: (1) eyes open with a fixed surface and visual surroundings; (2) eyes closed with a fixed surface; (3) eyes open with a fixed surface and sway referenced visual surroundings; (4) eyes open with a sway referenced surface and fixed visual surrounding; (5) eyes closed with a sway referenced surface, and (6) eyes open with a sway referenced surface and visual surroundings. For each trial, participants stood on the force platform and attempted to maintain their balance for 20 seconds. Each condition was completed 3 times. An equilibrium score ranging from 0 (falling/losing balance) to 100 (little to no sway) was calculated from the anteroposterior postural sway during each trial. An overall composite equilibrium score was calculated as the weighted average of all conditions, each of which was an average of the 3 trials within the given condition.

To identify sensory phenotypes, visual, vestibular, and proprioception sensory ratios were calculated from the individual condition SOT equilibrium scores.23 Sensory ratios closer to 0 indicated a worse ability for the central nervous system to appropriately interpret and use that sensory system for balance, while ratios closer to 1 indicated a better ability. For each sensory ratio, an abnormal sensory ratio was defined as the one that fell below the bottom quartile of non-mTBI participants' ratios; an abnormal sensory ratio does not necessarily mean there is peripheral dysfunction within that sensory system. Rather, an abnormal sensory ratio means there is some difficulty—whether in peripheral sensation, sensorimotor integration, or motor transformation and output—when required to use that sensory system for balance control. Eight nonoverlapping sensory phenotypes were then defined using the normal/abnormal classification (2 states and 3 ratios yield 8 possible phenotypes: 23 = 8).

Demographic and Clinical Characteristics

Standard demographic information such as age, sex, height, weight, and education were collected as part of the standard protocol. Clinical outcomes included smoking status, high cholesterol, diagnosis of diabetes, hazardous alcohol use (Alcohol Use Disorders Identification Test [AUDIT]),24 drug abuse (Drug Abuse Screening Test),25 and the Glasgow Outcome Scale—Extended (GOS-E).26

Self-Reported Symptoms

Self-reported symptoms were recorded using the Dizziness Handicap Inventory (DHI)27; NSI28; the Pittsburgh Sleep Quality Index29; Post-traumatic Stress Disorder Checklist (PCL-5)30; Patient Health Questionnaire (PHQ-9)31; Traumatic Brain Injury—Quality of Life (TBI-QOL)32; Headache Impact Test (HIT-6)33; and Tinnitus Functional Impact (TFI).34 Where possible, total symptom scores and symptom subscores were assessed (e.g., NSI vestibular subscore, TBI-QOL pain subscore).

Physical and Cognitive Function

Functional characteristics were assessed using standard mobility and cognitive assessments. Physical mobility function was assessed using the composite score on the SOT and the 4-m walk test. Cognitive function was assessed using the Trail Making Test (TMT-A and TMT-B),35 the NIH Toolbox,36 and the Wechsler Adult Intelligence Scale—Fourth Edition.37 All subscales of the cognitive tests were recorded, and age-adjusted and scaled scores were used in this analysis.

TBI History and Injury Mechanisms

TBI history and injury mechanisms were assessed using validated structured interview, as described earlier. Specifically, the curated data included the number of blast and blunt-force TBIs, the number of combat TBIs, the number of TBIs before deployment and the number after, the number of TBIs with post-traumatic amnesia, the time since the first TBI, and time since most recent TBI, and finally, the number of lifetime TBIs.

Statistical Analysis

Frequency distributions of both non-mTBI and mTBI individuals were generated across the 8 possible phenotypes. Within the mTBI group, the Kruskal-Wallis tests (or the χ2 tests, as appropriate, in the case of categorical data) compared each clinical and mechanistic outcome across phenotypes. To determine how symptomatic and functional outcomes differed by phenotype and between mTBI and non-mTBI participants, generalized linear models were fit for each outcome with the main effects of phenotype, mTBI/non-mTBI group, and the group × phenotype interaction. An F-test for fixed effects determined whether outcomes differed between phenotypes (main effect of phenotype), between mTBI and non-mTBI groups (main effect of group), or whether the effects varied depending on the phenotype and group (interaction effect). Models were adjusted for age, height, weight, and the frequency of alcohol use. A significance level of 0.05 was used throughout, and a Benjamini-Hochberg false discovery rate correction was applied to all the Kruskal-Wallis, χ2, and F-test p values to adjust for multiple comparisons across outcomes (total comparisons = 149).38

Data Availability

Data from the CENC study are available by download from the Federal Interagency Traumatic Brain Injury Registry (fitbir.nih.gov/content/access-data) for those who complete data access requirements or upon special request directly from the LIMBIC-CENC data core (limbic-cenc.org/index.php/for-scientists-and-clinicians/data-requests-public/).

Results

A total of 758 people with mTBI and 172 with no lifetime history of mTBI were included in this study (Figure 1). Overall, the non-mTBI group had a larger proportion of women (29.1%) compared with the mTBI group (10.4%), but both groups comprised individuals of similar age (40.6 years for mTBI; 40.1 years non-mTBI).

Figure 1. Consort Diagram Depicting Participant Selection and Exclusion for This Study.

Figure 1

Open in a new tab

Demographic information for the mTBI and non-mTBI groups are also presented as mean (SD), except for sex, which is presented as n (%). A large proportion of the sample reported regular use of special health equipment, including the use of hearing aids, walkers, and canes. Results excluding that group are presented in the online supplement. The overall results and interpretations remained consistent whether including or excluding this group, so the full sample was retained in the main results section. mTBI = mild traumatic brain injury; SOT = Sensory Organization Test.

Sensory Phenotype Determination

The distributions of sensory ratio scores for both mTBI and non-mTBI groups are given in Figure 2, B and C. The normal/abnormal cutoffs were 0.774, 0.622, and 0.954 for the visual, vestibular, and proprioception ratios, respectively. Within the mTBI group, 29% of individuals had abnormal visual ratios, 36% had abnormal vestibular ratios, and 38% had abnormal proprioception ratios. Only 41.7% of the mTBI group and 50.0% of the non-mTBI group exhibited normal sensory ratios across all 3 sensory systems (Figure 1C). The mTBI group had a higher proportion of individuals with abnormal SOT composite scores (<75) compared with the non-mTBI group (45.1% vs 35.4%). While sensory phenotypes varied among individuals with abnormal SOT composite scores, the plurality of such participants with mTBI had abnormal sensory ratios for all 3 sensory systems (27.5%). Of these participants with mTBI and abnormal SOT composite scores, 67.8% had abnormal vestibular ratios, 59.4% had abnormal visual ratios, and 48.2% had abnormal proprioception ratios.

Figure 2. Descriptions and Distributions of Sensory Phenotypes for Balance.

Figure 2

Open in a new tab

(A) Description of the 8 possible sensory phenotypes for balance control based on the sensory ratios from the SOT. (B) Distributions of the visual, vestibular, and proprioception ratios for the non-mTBI (blue) and mTBI (orange) groups. The dashed gray line indicates the threshold used to identify abnormal sensory ratios (25th percentile of the non-mTBI group). (C) Distributions of participants across phenotypes within the non-mTBI (blue) and mTBI (orange) groups. Both top and bottom panels illustrate the distribution across phenotypes for all individuals in the non-mTBI (blue) and mTBI (orange) groups. The bottom panel is restricted to only individuals in each group who had balance problems based on an abnormal CS on the SOT (CS < 75). CS = composite score; mTBI = mild traumatic brain injury; SOT = Sensory Organization Test.

Demographic and Clinical Characteristics

Sensory phenotypes within the mTBI group did not differ by sex, education, combat resiliency (DRRI2 Combat Total), number of deployments, or time since deployment (select outcomes reported in Table 1, see eTable, links.lww.com/WNL/C37). Similarly, sensory phenotypes for balance did not have different proportions of smokers and individuals with high cholesterol, diabetes, or drug abuse. However, hazardous alcohol use (AUDIT), the frequency of alcohol use (AUDIT), overall outcome (GOS-E score), PTSD symptom severity (PCL-5), and validity of medical symptoms (MSVT) differed by phenotype; height and weight also differed across sensory phenotypes (see Table 1 and eT).

Table 1.

Descriptive Statistics of Selected Demographic, Clinical Characteristics, Injury Mechanisms, and TBI History for Each Sensory Phenotype Within the mTBI Group

graphic file with name WNL-2022-200653t1.jpg

Open in a new tab

TBI History and Injury Mechanisms

Individuals with mTBI across different phenotypes reported different numbers of lifetime TBIs, combat TBIs, mixed blunt-blast TBIs, and short time since last TBI. Individuals with more than 1 abnormal sensory system generally reported more lifetime TBIs and shorter times since TBI compared with the All-Intact or single-deficiency phenotypes. Sensory phenotypes did not report different number of TBIs with post-traumatic amnesia or time since the first TBI (Table 1).

Self-Reported Symptoms

All symptom burdens differed by phenotype (Table 2) within the mTBI group. Generally, the all-impaired phenotype reported more severe symptom burdens than the all-intact, visual-impaired, vestibular-impaired, and proprioception-impaired phenotypes across all symptom inventories. While not uniform across all symptom inventories, a general trend was observed where phenotypes with at least 2 abnormal sensory ratios (visual dominant, vestibular dominant, proprioception dominant, all impaired) reported more severe symptoms than phenotypes with 1 or fewer abnormal sensory ratios. Generally, the mTBI group reported more severe dizziness (DHI), neurobehavioral (NSI), somatic (NSI Somatosensory), cognitive (NSI Cognitive and TBIQOL Cognition), vestibular (NSI Vestibular), ability to participate (TBIQOL Social), headache (HIT-6), and pain (TBIQOL Pain) symptoms compared with the non-mTBI group. However, there was a significant interaction for dizziness (DHI), cognitive (NSI Cognitive), depression (PHQ-9), ability to participate, executive function, anger, anxiety, and emotion (TBIQOL), where participants with mTBI in the vestibular-deficient phenotype reported better (i.e., less severe) symptoms compared with non-mTBI participants in the same vestibular-deficient phenotype (Figure 3).

Table 2.

Descriptive Statistics of Self-Reported Symptoms for Each Sensory Phenotype Within the mTBI Group

graphic file with name WNL-2022-200653t2.jpg

Open in a new tab

Figure 3. Average Z Scores for Symptomatic Outcomes in Both the mTBI (Colors) and Non-mTBI Control (Black) Groups Within Each Phenotype.

Figure 3

Open in a new tab

The thick solid lines indicate the mean z score for each group within each phenotype; dashed lines represent ±1 SD around the mean z score. As the directionality of some outcomes varied (e.g., higher symptoms is often worse, but higher cognitive scores are better), all z scores were transformed to align with a negative z score indicating worse performance/higher burden. Across many outcomes and phenotypes, the mTBI group exhibited more severe symptoms. However, outcomes that are bolded and italicized had a significant group × phenotype interaction, whereby non-mTBI controls in the vestibular-deficient phenotype had more severe symptoms compared with the mTBI vestibular-deficient phenotype. DHI = Dizziness Handicap Inventory; HIT-6 = Headache Impact Test; mTBI = mild traumatic brain injury; NSI = Neurobehavioral Symptom Inventory; PCL-5 = Post-traumatic Stress Disorder Checklist—DSM 5 Version; PHQ-9 = Patient Health Questionnaire—9 item; PSQI = Pittsburgh Sleep Quality Inventory; TBIQOL = TBI Quality of Life; TFI = Tinnitus Functional Impact.

Physical and Cognitive Function

Physical and cognitive functioning for the mTBI participants in each sensory phenotype are summarized in Table 3. Similar to the self-reported symptoms, physical and cognitive functioning worsened as the number of abnormal sensory ratios increased. However, the all-deficient phenotype did not always exhibit the worst functioning; the visual-deficient phenotype had the poorest neurocognitive performance compared with other phenotypes. When adjusting for phenotype, there were no main effect differences in physical or cognitive functioning between the mTBI and non-mTBI groups. However, there was a group × phenotype interaction for the SOT composite score (lower scores in the mTBI all-impaired phenotype compared with those in the non-mTBI all-impaired phenotype) and SOT vestibular ratio (lower ratios in the mTBI visual-dominant phenotype compared with those in the non-mTBI visual-dominant phenotype) (Figure 4).

Table 3.

Descriptive Statistics of Physical and Cognitive Function for Each Sensory Phenotype Within the mTBI Group

graphic file with name WNL-2022-200653t3.jpg

Open in a new tab

Figure 4. Average Z Scores for Symptomatic Outcomes in Both the mTBI (Colors) and Non-mTBI Control (Black) Groups Within Each Phenotype.

Figure 4

Open in a new tab

The thick solid lines indicate the mean z-score for each group within each phenotype; dashed lines represent ±1 SD around the mean z score. As the directionality of some outcomes varied (e.g., higher symptoms are often worse, but higher cognitive scores are better), all z scores were transformed to align with a negative z score indicating worse performance/higher burden. After adjusting for phenotype, there were no main effects of group between the mTBI and non-mTBI control groups. There were 2 significant group × phenotype interaction (SOT composite equilibrium score and SOT vestibular ratio). mTBI = mild traumatic brain injury; SOT = Sensory Organization Test; SS = scaled score; WAIS-IV = Wechsler Adult Intelligence Scale—Fourth Edition.

Discussion

The results of this secondary analysis of Veterans and Service Members with and without mTBI indicate the chronic effects of mTBI on balance are heterogeneous. Specifically, current and former Service Members with mTBI exhibited various combinations of abnormal balance control ranging from single-system deficiencies (e.g., proprioception deficient) to multisystem deficiencies (e.g., all deficient). Thus, while previous studies have presented the proportion of individuals with abnormal sensory ratios on single systems,15,16 the present results suggest more nuanced phenotypes for balance control exist within Service Members and Veterans with mTBI. These abnormal sensory ratios for each phenotype do not necessarily mean there is peripheral sensory dysfunction, but instead that individuals have difficulty—whether in sensation, sensorimotor integration, or motor transformation and execution—when required to use that sensory system for balance control. Variations in global outcomes, indicated by the GOS-E, suggest participants in different sensory phenotypes have different levels of disability after their injury. Furthermore, participants within each phenotype exhibited unique symptomatic complaints and physical and cognitive functioning. Taken as a whole, these results suggest clinical care may benefit from a more nuanced and individualized understanding of balance control after mTBI. With more precise phenotyping of balance control problems, the next challenge will be to develop and test rehabilitation protocols that specifically address the deficient balance control systems. Specific protocols tailored to specific sensory phenotypes—including multimodal protocols targeting the mixed, multisystem sensory phenotypes—may lead to precision rehabilitation and achieve better outcomes for patients.

The most striking result of this analysis is the vestibular-deficient phenotype. Within team-based models of clinical care for mTBI, balance problems are classified under a vestibular subtype and typically lead to vestibular rehabilitation-based treatments.6,8 However, the results in this analysis suggest this label is an oversimplification and potentially therapeutically misleading; individuals with poor balance, as indicated by an abnormal SOT composite score (Figure 3, C and D), had a variety of abnormal sensory ratios. While more than 68% of participants with mTBI and abnormal SOT composite scores had abnormal vestibular ratios, most of the participants had at least 1 other abnormal sensory ratio. While the participants included in this analysis were Veterans and Service Members who were, on average, 8 years removed from their most recent mTBI, these findings agree with recent findings in university students with more recent mTBI, where combined visual and vestibular processing abnormalities were most common in individuals with mTBI.13 Furthermore, a recent systematic review found a disproportionate number of case studies, relatively few empirical studies, and no longitudinal studies of the vestibular consequences of TBI.39 The subtype label of vestibular, therefore, seems to be incomplete and potentially inaccurate. Individuals with mTBI who exhibited the vestibular-deficient phenotype in this study reported fewer and less severe symptoms of dizziness, cognition, participation, executive function, anger, and anxiety than non-mTBI participants within the vestibular-deficient phenotype. Thus, individuals who have trouble using vestibular information for balance control may be less likely to be classified under a vestibular subtype using symptomology alone. These results challenge the traditional understanding of balance and dizziness after mTBI; common models assuming vestibular-centric views of dizziness do not hold true across the heterogeneous mTBI population.40 Rather, growing evidence suggests that mTBI affects balance through changes in central sensorimotor integration and motor transformation, rather than impaired peripheral vestibular function.16,39 It is worth noting that despite the potential misnomer of the vestibular subtype, many of the components of a vestibular rehabilitation program for mTBI embed sensory integration and neuromuscular control that reflect the more generalized dysfunction within the central nervous system.41 Nevertheless, the vestibular subtype seems to be a misnomer, and it may be better named the sensorimotor subtype.

While problems integrating vestibular and visual information are still considered primary drivers of imbalance after mTBI,13,15 the proprioception-deficient phenotype was the most common single phenotype, besides the all-intact phenotype, in individuals with mTBI. Results in civilians with chronic complaints of imbalance also suggest abnormal control of balance using proprioception is common in individuals with mTBI; more individuals with mTBI had abnormal proprioception ratios than abnormal visual or vestibular ratios.16 Speculatively, an abnormal proprioception ratio may be related to high levels of pain in Veterans and Service Members with mTBI42,43 based on associations between pain and impaired proprioception in other populations44 and interim results relating pain to balance performance in this cohort.19 Yet, the clinical effect of abnormal proprioceptive control of balance after mTBI remains unclear. In this study, the proprioception-deficient phenotype had, on average, better balance (i.e., higher SOT composite scores) than all other phenotypes with at least 1 abnormal sensory ratio. Participants in this study were several years post-mTBI and may have adapted to impaired proprioception by relying on visual and vestibular information to control balance. Despite the relatively high SOT composite scores, the prevalence of proprioceptive-related balance impairments in the mTBI group deserves additional clinical scrutiny because of the relationship between proprioception and musculoskeletal injuries45,46 and the high rate of musculoskeletal injuries after mTBI/concussion in both athletes47 and military Service Members.48 Future research should more thoroughly explore proprioceptive function and the integration of proprioceptive signals for balance control in individuals after mTBI.

Physical functioning, cognitive functioning, and many self-reported symptoms were more strongly associated with sensory phenotypes than with mTBI history. After adjusting for phenotypes, there were few differences between mTBI and non-mTBI groups, and there were no differences in cognitive or physical functioning based on mTBI history after adjusting for phenotype. Conversely, every symptom and all assessments of physical and cognitive function, except for 2 (NIH Toolbox List Sorting and Picture Sequence), differed by phenotype. Both sensory phenotype and mTBI history should be considered when assessing one's symptoms. However, while some symptoms may still provide additional information, one's sensory phenotype for balance control may be more informative and clinically relevant to physical and cognitive function than his or her mTBI history, particularly in a chronic population. Future work may consider exploring structural and functional neuroimaging markers across phenotypes or including phenotypes as a covariate to control for heterogeneity in cohorts with mTBI.

Previous work postulated that military Service Members subject to blast-induced and noise-induced trauma may be at an increased risk of peripheral vestibular dysfunction that would manifest in deficient use of vestibular information for balance control.49 Yet, empirical evidence of this relationship is still lacking.16 In this sample of military Service Members and Veterans, the vestibular-dominant phenotype reported the most blast TBIs. Whether blast-related TBIs have different long-term consequences for the sensorimotor control of balance remains unclear.

It is worth noting that 35% of non-mTBI participants had relatively low SOT composite scores (less than 75). Based on the manufacturer's stated normative data with mean (SD) composite scores equal to 79.8 (5.7), we would expect only 20% of normal individuals to have composite scores less than 75. The higher proportion in our sample may be due to comorbidities, including chronic pain, PTSD, and sleep apnea in Veterans and Service Members,43 which previous preliminary analyses have linked to lower SOT composite scores in Veterans and Service Members19 and the general population.

Several limitations should be acknowledged. First, despite the large sample size in this study, several phenotypes still contained only a handful of non-mTBI individuals. As expected with any phenotyping study, these small non-mTBI subgroups limited the ability to make statistical inference across some of the rarer phenotypes, particularly the visual-dominant and vestibular-dominant phenotypes. Similarly, these results are based on a specific population of military Service members and Veterans with and without a previous history of mTBI. It is unknown whether these phenotypes are present with similar frequency or presentation in a civilian population or in a population with acute mTBI. Finally, the thresholds used to determine the normal/abnormal cutoff for each sensory ratio were based on the non-mTBI Veteran and Service Member population, which may not be generalizable to a civilian population. The sensory ratios can fluctuate depending on the specific population. While the thresholds used here were based on this specific population, future work may consider general population-based thresholds for future application of sensory phenotypes.

The results presented in this study clearly illustrate that Veterans and Service Members with mTBI exhibit a variety of balance deficits that involve heterogeneous problems maintaining balance when relying on different sensory information. While the use of vestibular information for balance control is commonly impaired, it is inaccurate to label all mTBI-related balance dysfunction under the vestibular umbrella. Such inaccurate labeling may discourage clinicians from exploring the nuances of each individual's sensory phenotype after mTBI. Furthermore, different sensory phenotypes exhibit important differences in symptomology and functional outcomes and highlight a key opportunity to perform more individually targeted therapies. We think it is important, therefore, to accurately classify patients with persisting symptoms from mTBI into more discrete categories of balance dysfunction, and propose using the sensory phenotype classifications as used here. Future work should use more specific classification of balance deficits, including identifying specific sensory phenotypes for balance control, rather than the current vestibular terminology. At a minimum, the vestibular subtype should be renamed the sensorimotor subtype.

Future work is also needed to test the effectiveness of more tailored rehabilitation programs using this refined classification system of balance dysfunction. Presumably, the understanding and use of more accurate and specific classifications will encourage clinical practice patterns that move toward clinical examinations and treatments tailored to individual sensorimotor deficits, hopefully translating to improved health and outcomes in this important population. Finally, future investigations exploring more clinically accessible technology, such as portable inertial sensors and the clinical test for sensory interaction and balance14,50 to generate the multisystem sensory ratios proposed in this study (rather than relying on expensive stationary equipment and the SOT), could further advance these methods toward more widespread clinical application.

Acknowledgment

The authors thank the Service Members and Veterans who participated in this study. They also acknowledge the efforts of the LIMBIC-CENC Prospective Longitudinal Study Site PIs and staff and other members of the consortium who contributed to data collection.

Glossary

AUDIT

Alcohol Use Disorders Identification Test

DHI

Dizziness Handicap Inventory

GOS-E

Glasgow Outcome Scale—Extended

HIT

Headache Impact Test

LIMBIC-CENC

Long-Term Impact of Military-Relevant Brain Injury Consortium—Chronic Effects of Neurotrauma Consortium

mBIAS

Mild Brain Injury Atypical Symptoms

mTBI

mild traumatic brain injury

MSVT

Medical Symptom Validity Test

NSI

Neurobehavioral Symptom Inventory

PCE

potential concussive event

PCL

Post-traumatic Stress Disorder Checklist

PHQ

Patient Health Questionnaire

PLS

Prospective Longitudinal Study

PTSD

post-traumatic stress disorder

SOT

Sensory Organization Test

TBI-QOL

Traumatic Brain Injury—Quality of Life

TMT

Trail Making Test

Appendix. Authors

Appendix.

Open in a new tab

Footnotes

CME Course: NPub.org/cmelist

Study Funding

This work was supported by the Assistant Secretary of Defense for Health Affairs endorsed by the Department of Defense, through the Psychological Health/Traumatic Brain Injury Research Program Long-Term Impact of Military-Relevant Brain Injury Consortium (LIMBIC) Award/W81XWH-18-PH/TBIRP-LIMBIC under Awards No. W81XWH1920067 and W81XWH-13-2-0095 and by the US Department of Veterans Affairs Awards No. I01 CX002097, I01 CX002096, I01 HX003155, I01 RX003444, I01 RX003443, I01 RX003442, I01 CX001135, I01 CX001246, I01 RX001774, I01 RX 001135, I01 RX 002076, I01 RX 001880, I01 RX 002172, I01 RX 002173, I01 RX 002171, I01 RX 002174, and I01 RX 002170. The US Army Medical Research Acquisition Activity, 839 Chandler Street, Fort Detrick MD 21702-5014 is the awarding and administering acquisition office. Additional support was provided by the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the NIH under Award Number K12HD073945 and R21HD100897 and VA Health Services Research and Development Service (RCS 17–297, Award No. IK6HX002608).

Disclosure

The authors report no disclosures relevant to the manuscript. Go to Neurology.org/N for full disclosures.

References

  • 1.Nelson LD, Temkin NR, Dikmen S, et al. Recovery after mild traumatic brain injury in patients presenting to US level I trauma centers: a transforming research and clinical knowledge in traumatic brain injury (TRACK-TBI) study. JAMA Neurol. 2019;76(9):1049-1059. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Kontos AP, Kotwal RS, Elbin RJ, et al. Residual effects of combat-related mild traumatic brain injury. J Neurotrauma. 2013;30(8):680-686. [DOI] [PubMed] [Google Scholar]
  • 3.Phipps H, Mondello S, Wilson A, et al. Characteristics and impact of U.S. military blast-related mild traumatic brain injury: a systematic review. Front Neurol. 2020;11:559318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Helmick KM, Spells CA, Malik SZ, Davies CA, Marion DW, Hinds SR. Traumatic brain injury in the US military: epidemiology and key clinical and research programs. Brain Imaging Behav. 2015;9(3):358-366. [DOI] [PubMed] [Google Scholar]
  • 5.Wilson L, Stewart W, Dams-O'Connor K, et al. The chronic and evolving neurological consequences of traumatic brain injury. Lancet Neurol. 2017;16(10):813-825. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Harmon KG, Clugston JR, Dec K, et al. American Medical Society for Sports Medicine position statement on concussion in sport. Br J Sports Med. 2019;53(4):213-225. [DOI] [PubMed] [Google Scholar]
  • 7.Quatman-Yates CC, Hunter-Giordano A, Shimamura KK, et al. Physical therapy evaluation and treatment after concussion/mild traumatic brain injury. J Orthop Sports Phys Ther. 2020;50(4):CPG1-CPG73. [DOI] [PubMed] [Google Scholar]
  • 8.Kontos AP, Collins MW, Holland CL, et al. Preliminary evidence for improvement in symptoms, cognitive, vestibular, and oculomotor outcomes following targeted intervention with chronic mTBI patients. Mil Med. 2018;183(suppl 1):333-338. [DOI] [PubMed] [Google Scholar]
  • 9.Moller MC, Lexell J, Wilbe Ramsay K. Effectiveness of specialized rehabilitation after mild traumatic brain injury: a systematic review and meta-analysis. J Rehabil Med. 2021;53(2):jrm00149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Maruta J, Lumba-Brown A, Ghajar J. Concussion subtype identification with the rivermead post-concussion symptoms questionnaire. Front Neurol. 2018;9:1034. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Peterka RJ. Sensorimotor integration in human postural control. J Neurophysiol. 2002;88(3):1097-1118. [DOI] [PubMed] [Google Scholar]
  • 12.Peterka RJ, Statler KD, Wrisley DM, Horak FB. Postural compensation for unilateral vestibular loss. Front Neurol. 2011;2:57. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Caccese JB, Santos FV, Yamaguchi FK, Buckley TA, Jeka JJ. Persistent visual and vestibular impairments for postural control following concussion: a cross-sectional study in university students. Sports Med. 2021;51(10):2209-2220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Fino PC, Raffegeau TE, Parrington L, Peterka RJ, King LA. Head stabilization during standing in people with persisting symptoms after mild traumatic brain injury. J Biomech. 2020;112:110045. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Haran FJ, Slaboda JC, King LA, Wright WG, Houlihan D, Norris JN. Sensitivity of the balance error scoring system and the sensory organization test in the combat environment. J Neurotrauma. 2016;33(7):705-711. [DOI] [PubMed] [Google Scholar]
  • 16.Campbell KR, Parrington L, Peterka RJ, et al. Exploring persistent complaints of imbalance after mTBI: oculomotor, peripheral vestibular and central sensory integration function. J Vestib Res. 2021;31(6):519-530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Brett BL, Kramer MD, Whyte J, et al. Latent profile analysis of neuropsychiatric symptoms and cognitive function of adults 2 weeks after traumatic brain injury: findings from the TRACK-TBI study. JAMA Netw Open. 2021;4(3):e213467. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Walker WC, Carne W, Franke LM, et al. The Chronic Effects of Neurotrauma Consortium (CENC) multi-centre observational study: Description of study and characteristics of early participants. Brain Inj. 2016;30(12):1469-1480. [DOI] [PubMed] [Google Scholar]
  • 19.Walker WC, Nowak KJ, Kenney K, et al. Is balance performance reduced after mild traumatic brain injury?: interim analysis from chronic effects of neurotrauma consortium (CENC) multi-centre study. Brain Inj. 2018;32(10):1156-1168. [DOI] [PubMed] [Google Scholar]
  • 20.Walker WC, Cifu DX, Hudak AM, Goldberg G, Kunz RD, Sima AP. Structured interview for mild traumatic brain injury after military blast: inter-rater agreement and development of diagnostic algorithm. J Neurotrauma. 2015;32(7):464-473. [DOI] [PubMed] [Google Scholar]
  • 21.Green P. Manual for the Medical Symptom Validity Test . Green's Publishing; 2004. [Google Scholar]
  • 22.Cooper DB, Nelson L, Armistead-Jehle P, Bowles AO. Utility of the mild brain injury atypical symptoms scale as a screening measure for symptom over-reporting in operation enduring freedom/operation iraqi freedom service members with post-concussive complaints. Arch Clin Neuropsychol. 2011;26(8):718-727. [DOI] [PubMed] [Google Scholar]
  • 23.Nashner LM. Computerized dynamic posturography. In: Jacobson GP, Newman CW, Kartush JM, eds. Handbook of Balance Function Testing. Mosby Year Book; 1993:208-307. [Google Scholar]
  • 24.Saunders JB, Aasland OG, Babor TF, Delafuente JR, Grant M. Development of the alcohol-use disorders identification test (audit)—who collaborative project on early detection of persons with harmful alcohol-consumption .2. Addiction. 1993;88(6):791-804. [DOI] [PubMed] [Google Scholar]
  • 25.Skinner HA. The drug abuse screening test. Addict Behav. 1982;7(4):363-371. [DOI] [PubMed] [Google Scholar]
  • 26.Wilson JT, Pettigrew LE, Teasdale GM. Structured interviews for the Glasgow Outcome Scale and the extended Glasgow Outcome Scale: guidelines for their use. J Neurotrauma. 1998;15(8):573-585. [DOI] [PubMed] [Google Scholar]
  • 27.Jacobson GP, Newman CW. The development of the Dizziness Handicap Inventory. Arch Otolaryngol Head Neck Surg. 1990;116(4):424-427. [DOI] [PubMed] [Google Scholar]
  • 28.Cicerone KD, Kalmar K. Persistent postconcussion syndrome: the structure of subjective complaints after mild traumatic brain injury. J Head Trauma Rehabil. 1995;10(3):1-17. [Google Scholar]
  • 29.Carpenter JS, Andrykowski MA. Psychometric evaluation of the Pittsburgh Sleep Quality Index. J Psychosom Res. 1998;45(1):5-13. [DOI] [PubMed] [Google Scholar]
  • 30.Blevins CA, Weathers FW, Davis MT, Witte TK, Domino JL. The posttraumatic stress disorder checklist for DSM-5 (PCL-5): development and initial psychometric evaluation. J Trauma Stress. 2015;28(6):489-498. [DOI] [PubMed] [Google Scholar]
  • 31.Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med. 2001;16(9):606-613. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Tulsky DS, Kisala PA, Victorson D, et al. TBI-QOL: development and calibration of item banks to measure patient reported outcomes following traumatic brain injury. J Head Trauma Rehabil. 2016;31(1):40-51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Kosinski M, Bayliss MS, Bjorner JB, et al. A six-item short-form survey for measuring headache impact: the HIT-6. Qual Life Res. 2003;12(8):963-974. [DOI] [PubMed] [Google Scholar]
  • 34.Meikle MB, Henry JA, Griest SE, et al. The tinnitus functional index: development of a new clinical measure for chronic, intrusive tinnitus. Ear Hear. 2012;33(2):153-176. [DOI] [PubMed] [Google Scholar]
  • 35.Reitan RM. The relation of the trail making test to organic brain damage. J Consult Psychol. 1955;19(5):393-394. [DOI] [PubMed] [Google Scholar]
  • 36.Gershon RC, Wagster MV, Hendrie HC, Fox NA, Cook KF, Nowinski CJ. NIH toolbox for assessment of neurological and behavioral function. Neurology. 2013;80(11 suppl 3):S2-S6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Wechsler D. Wechsler Adult Intelligence Scale. Psychological Corp.; 1955. [Google Scholar]
  • 38.Benjamini Y, Hochberg Y. Controlling the false discovery rate—a practical and powerful approach to multiple testing. J R Stat Soc B. 1995;57(1):289-300. [Google Scholar]
  • 39.Sarkic B, Douglas JM, Simpson A, et al. Frequency of peripheral vestibular pathology following traumatic brain injury: a systematic review of literature. Int J Audiol. 2021;60(7):479-494. [DOI] [PubMed] [Google Scholar]
  • 40.Reneker JC, Cheruvu VK, Yang J, James MA, Cook CE. Physical examination of dizziness in athletes after a concussion: a descriptive study. Musculoskelet Sci Pract. 2018;34:8-13. [DOI] [PubMed] [Google Scholar]
  • 41.Alsalaheen BA, Mucha A, Morris LO, et al. Vestibular rehabilitation for dizziness and balance disorders after concussion. J Neurol Phys Ther. 2010;34(2):87-93. [DOI] [PubMed] [Google Scholar]
  • 42.Grandhi R, Tavakoli S, Ortega C, Simmonds MJ. A review of chronic pain and cognitive, mood, and motor dysfunction following mild traumatic brain injury: complex, comorbid, and/or overlapping conditions? Brain Sci. 2017;7(12):160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Pugh MJ, Finley EP, Copeland LA, et al. Complex comorbidity clusters in OEF/OIF veterans: the polytrauma clinical triad and beyond. Med Care. 2014;52(2):172-181. [DOI] [PubMed] [Google Scholar]
  • 44.Lee AS, Cholewicki J, Reeves NP, Zazulak BT, Mysliwiec LW. Comparison of trunk proprioception between patients with low back pain and healthy controls. Arch Phys Med Rehabil. 2010;91(9):1327-1331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Nagai T, Schilaty ND, Strauss JD, Crowley EM, Hewett TE. Analysis of lower extremity proprioception for anterior cruciate ligament injury prevention: current opinion. Sports Med. 2018;48(6):1303-1309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Chaput M, Onate JA, Simon JE, et al. Visual cognition associated with knee proprioception, time to stability, and sensory integration neural activity after ACL reconstruction. J Orthop Res. 2022;40(1):95-104. [DOI] [PubMed] [Google Scholar]
  • 47.McPherson AL, Nagai T, Webster KE, Hewett TE. Musculoskeletal injury risk after sport-related concussion: a systematic review and meta-analysis. Am J Sports Med. 2019;47(7):1754-1762. [DOI] [PubMed] [Google Scholar]
  • 48.Reneker JC, Babl R, Flowers MM. History of concussion and risk of subsequent injury in athletes and service members: a systematic review and meta-analysis. Musculoskelet Sci Pract. 2019;42:173-185. [DOI] [PubMed] [Google Scholar]
  • 49.Akin FW, Murnane OD, Hall CD, Riska KM. Vestibular consequences of mild traumatic brain injury and blast exposure: a review. Brain Inj. 2017;31(9):1188-1194. [DOI] [PubMed] [Google Scholar]
  • 50.Gera G, Chesnutt J, Mancini M, Horak FB, King LA. Inertial sensor-based assessment of central sensory integration for balance after mild traumatic brain injury. Mil Med. 2018;183(suppl 1):327-332. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data from the CENC study are available by download from the Federal Interagency Traumatic Brain Injury Registry (fitbir.nih.gov/content/access-data) for those who complete data access requirements or upon special request directly from the LIMBIC-CENC data core (limbic-cenc.org/index.php/for-scientists-and-clinicians/data-requests-public/).

Articles from Neurology are provided here courtesy of American Academy of Neurology

RESOURCES