Skip to main content
  • Study protocol
  • Open access
  • Published:

Assessment and rehabilitation of central sensory impairments for balance in mTBI using auditory biofeedback: a randomized clinical trial

Abstract

Background

Complaints of imbalance are common non-resolving signs in individuals with post-concussive syndrome. Yet, there is no consensus rehabilitation for non-resolving balance complaints following mild traumatic brain injury (mTBI). The heterogeneity of balance deficits and varied rates of recovery suggest varied etiologies and a need for interventions that address the underlying causes of poor balance function. Our central hypothesis is that most chronic balance deficits after mTBI result from impairments in central sensorimotor integration that may be helped by rehabilitation. Two studies are described to 1) characterize balance deficits in people with mTBI who have chronic, non-resolving balance deficits compared to healthy control subjects, and 2) determine the efficacy of an augmented vestibular rehabilitation program using auditory biofeedback to improve central sensorimotor integration, static and dynamic balance, and functional activity in patients with chronic mTBI.

Methods

Two studies are described. Study 1 is a cross-sectional study to take place jointly at Oregon Health and Science University and the VA Portland Health Care System. The study participants will be individuals with non-resolving complaints of balance following mTBI and age- and gender-matched controls who meet all inclusion criteria. The primary outcome will be measures of central sensorimotor integration derived from a novel central sensorimotor integration test. Study 2 is a randomized controlled intervention to take place at Oregon Health & Science University. In this study, participants from Study 1 with mTBI and abnormal central sensorimotor integration will be randomized into two rehabilitation interventions. The interventions will be 6 weeks of vestibular rehabilitation 1) with or 2) without the use of an auditory biofeedback device. The primary outcome measure is the daily activity of the participants measured using an inertial sensor.

Discussion

The results of these two studies will improve our understanding of the nature of balance deficits in people with mTBI by providing quantitative metrics of central sensorimotor integration, balance, and vestibular and ocular motor function. Study 2 will examine the potential for augmented rehabilitation interventions to improve central sensorimotor integration.

Trial registration

This trial is registered at clinicaltrials.gov (NCT02748109)

Peer Review reports

Background

Between 1.6 and 3.8 million sports-related mild traumatic brain injuries (mTBI) occur annually in the United States [1] and up to 15% of military combat veterans sustain mTBI [2]. The estimated annual cost of TBI, including lost productivity, exceeds $60 billion [3]. While self-reported symptoms of mTBI typically recover within three months [4], between 11 and 64% of individuals, depending on diagnostic criteria, develop post-concussive syndrome with persistent, non-resolving symptoms [5]. Despite the high incidence and cost to society, studies investigating the treatment of chronic mTBI symptoms have been scarce. In particular, complaints of imbalance are common following mTBI [6, 7] and are significant contributors to anxiety and difficulty returning to work [8]. Yet, there is currently no consensus, evidence-based rehabilitation procedure for balance complaints following mTBI [9].

The difficulty establishing consensus rehabilitation protocols is likely due to the complex nature of balance deficits in individuals with mTBI that can include neurological damage to the brainstem [10, 11], thalamus [12], and cerebellum [13], and / or injury to peripheral vestibular organs [1417]. Sensory information from the vestibular, visual, and somatosensory systems are integrated in the central nervous system to control static and dynamic balance (Fig. 1). In patients with mTBI, vestibular function can be impaired by damage to the peripheral vestibular organs. The semicircular canals can be impaired in up to two-thirds of individuals with mTBI [15, 17], and the otolith organs are also susceptible to damage from mTBI [14, 16]. In addition to disrupting vestibular function, mTBI has also been linked with central deficits affecting ocular motor function including extended saccadic latency [18] and poor visual tracking accuracy during smooth pursuit tasks [19, 20]. In one study, 60% of individuals with mTBI had errors during pursuit whereas no controls exhibited similar errors [21].

Fig. 1
figure 1

Sensory systems contributing to static and dynamic balance

A proportion of individuals with mTBI who do not have prominent or measurable sensory deficits may still have balance problems attributable to abnormal central control of balance. Central sensorimotor integration is a critical aspect of balance control that may underlie dysfunction in static and dynamic balance. Proper central sensorimotor integration relies on the central nervous system to 1) select the optimal combination of sensory sources for balance and 2) reweight the sensory contributions as sensory conditions change [22, 23]. When peripheral vestibular, visual, and somatosensory systems are intact, difficulty performing static or dynamic balance tasks under challenging surface and/or visual conditions may be due to centrally-mediated suboptimal weighting and/or reweighting of vestibular, visual, and somatosensory information or due to abnormal sensory-to-motor transformations that inappropriately scale the magnitude of corrective motor actions [2224]. For example, standing or walking on compliant foam or irregular surfaces requires the nervous system to rely more on vestibular and visual inputs while down-weighting the normal dependence on somatosensory information [22]. This reweighting is also important for individuals with sensory impairment, as altered patterns of weighting are necessary to compensate for the sensory dysfunction. Some patterns of compensation may be more functional than others, suggesting an important role of rehabilitation in helping subjects develop effective compensation strategies to improve recovery of central sensorimotor integration and balance [24]. However, central sensorimotor integration has not been systematically characterized in patients with mTBI, regardless of sensory loss.

The lack of knowledge about central sensorimotor integration following mTBI may be in part due to the current clinical tools used to assess balance, gait, and sensorimotor integration. The Balance Error Scoring System (BESS) is the most frequently administered clinical balance assessment for patients with suspected mTBI. Yet, the BESS suffers from a high degree of subjectivity, provides only limited, low-resolution information about the balance control system [25], and is not sensitive beyond the acute stages of mTBI [26]. Balance measures using force platforms and sophisticated post-processing and analysis techniques have shown greater sensitivity in detecting long-lasting postural deficits [2731]. Similar to static balance, measures of dynamic balance, such as trunk motion during gait [3235] have shown persistent impairment post-concussion. Sensitive, objective measures of balance can be obtained using wearable sensors [3639] and are well-suited for clinical and rehabilitation environments [40, 41].

For instance, wearable sensors enable biofeedback based on quantitative measurements of balance and gait. Specifically, auditory biofeedback (ABF), which communicates sway through changes in the pitch or location of a tone, has shown promise in aiding balance and retraining central sensory integration mechanisms in populations with balance impairments [42, 43]. Quantitative measures of sway and body position are communicated to the patient via audition, in real-time, to augment existing sensory information and provide accurate information to the central nervous system about body sway that may help recalibrate distorted sensory integration. Compared to visual biofeedback paradigms, ABF is particularly well suited for populations with mTBI as the auditory cue does not interfere with other postural sensory systems, such as vision.

The heterogeneity of balance deficits and varied rates of recovery after mTBI suggest varied etiologies and a need for interventions that address the underlying causes of poor balance function. Over-simplified clinical balance measures often do not account for the complex integration of sensory and motor systems. Therefore, the first goal of this study is to characterize balance deficits in people with mTBI who have chronic, non-resolving balance deficits compared to healthy control subjects without a history of mTBI. We hypothesize that a) objective measures of central sensorimotor integration, static and dynamic balance will better distinguish people with mTBI from control subjects than clinical measures, b) a subset of people with mTBI will have abnormal central sensorimotor integration test measures, even without peripheral vestibular or ocular motor deficits, and c) the relationship between poorer static/dynamic balance performance and mTBI is regulated/mediated by central sensorimotor integration. The second goal is to determine if augmenting a traditional balance rehabilitation program with a wearable-sensor based ABF system improves central sensory integration in patients with chronic mTBI compared to the standard balance rehabilitation program. Our central hypothesis is that chronic balance deficits after mTBI result from impairments in central sensorimotor integration that may be helped by rehabilitation that challenges balance while simultaneously providing feedback on balance performance. We hypothesize that a) central sensorimotor integration scores will improve with rehabilitation and ABF will increase the improvement of central sensorimotor integration scores beyond the standard of care, b) intervention with ABF will improve objective measures of balance and c) people with central sensorimotor integration impairment will show sustained improvement in central sensorimotor integration scores and balance after rehabilitation with ABF.

Methods / Design

This study has two parts: Study 1) A cross-sectional study to identify and characterize balance control strategies after mTBI compared to healthy controls, and Study 2) An interventional randomized pilot study using a novel ABF rehabilitation technique to improve central sensorimotor integration after mTBI. All subjects will complete baseline clinical sensory, ocular motor, vestibular, neurocognitive, static and dynamic balance, and central sensorimotor integration testing. Following baseline testing, a subset of mTBI participants will be randomized into two treatment groups for 6 weeks of rehabilitation. Within 1 week after completing the rehabilitation program, subjects will repeat the balance and central sensorimotor integration testing to examine the short-term effects of rehabilitation. Subjects will complete the balance and central sensorimotor integration testing a third time, 6 weeks after completion of the rehabilitation, to assess retention. This trial is registered at clinicaltrials.gov (NCT02748109).

Participants

The study will include 65 mTBI participants and 65 control subjects for the cross-sectional arm of the study. All subjects will be between the ages of 18 and 60. Participants in the mTBI group will have a diagnosed mTBI or concussion and chronic, non-resolving complaints of balance problems determined by a score >0 on the Dizziness Handicap Inventory (DHI) at the time of testing and > 3 months post-injury. Control subjects will be gender and age matched to the mTBI group. For the second arm of the study, a subset of 40 mTBI subjects with abnormal central sensorimotor integration with or without peripheral vestibular or ocular motor deficits will be randomized into two rehabilitation groups, experimental treatment and standard of care, consisting of two rehabilitation sessions per week for 6 weeks. All subjects will be tested at either Oregon Health & Science University or the VA Portland Health Care System. All rehabilitation will take place at Oregon Health & Science University. Informed written consent will be obtained from all participants.

Inclusion and exclusion criteria

Inclusion criteria for mTBI recruitment are 1) a diagnosis of mTBI based upon VHA/DoD criteria with persisting symptoms >3 months post injury [44], 2) between 21–50 years old, 3) a score >0 on the DHI, 4) minimal cognitive impairment; a score between 0 and 8 on the Short Blessed test for cognitive function [45], and 5) may or may not have had a loss of consciousness (LOC) with their initial injury. Inclusion criteria for controls are between 21–50 years old and no history of mTBI or brain injury. Exclusion criteria are: 1) have had or currently have any other injury, medical, substance or neurological illness that could potentially explain balance deficits (e.g., CNS disease, stroke, moderate TBI, lower extremity amputation) 2) meet criteria for moderate to severe substance use disorder within the past month, as defined by DSM-V, 3) display behavior that would significantly interfere with validity of data collection or safety during study, 4) be in significant pain during the evaluation (5/10 by patient subjective report), 5) be a pregnant female (balance considerations), 6) have past history of peripheral vestibular pathology or ocular motor deficits, 7) have significant hearing loss that would interfere with Study 1 no worse than 60 dB HL (PTA 0.5–3 kHz) and Study 2; hearing loss no worse than 30 dB HL (PTA 0.5–3 kHz), in better ear, with the difference in ears being less than 15 dB PTA, or 8) be unable to abstain for 24 h in advance of testing in the use medications that might impair their balance.

Study 1: Characterization of balance deficits

Assessment procedures

All people who are eligible per phone screening will come into the clinic for the informed consent process. An investigator will verbally explain the consent form, allow the person ample time to read through the consent form and then will acknowledge consent by signing the form. All subjects will first read and sign consent forms. For study 1, 65 mTBI participants and 65 control subjects will be assessed on a battery of tests designed to measure vestibular function, sensorimotor integration, static and dynamic balance, cognition, reaction time, clinical sensory loss, and self-reported symptoms. All protocols have been approved by the OHSU Institutional Review Board.

Participants will complete a novel central sensorimotor integration test (CSMI test) [22] on a modified research NeuroCom platform using custom-designed, low-amplitude (2° and 4° peak-to-peak) pseudorandom stimuli that apply continuously repeated cycles of wide bandwidth surface-tilt and/or visual-tilt stimuli with individual tests lasting ~5 min. The participants’ anteroposterior (AP) body displacement at shoulder and hip levels, and surface and visual surround rotation angles will be recorded. Subjects will perform 8 trials (2 amplitudes x 4 test conditions) in one session with trials presented in a randomized order and interspersed with rest breaks (Table 1). Anthropometric measures will be obtained for the purpose of estimating body mechanics needed for later analysis (moment of inertia, mass, height of center of mass, CoM) [22].

Table 1 Aim 1 CSMI test conditions and sensory weight comparisons

Primary outcome measure

The primary outcome measures of central sensorimotor integration are the parameters derived from a model-based interpretation of CoM sway evoked by pseudorandom stimuli (Table 2). The procedure for calculating these parameters is to (1) calculate CoM angular displacement from experimental measures of hip and shoulder displacement on each test trial, (2) use Fourier methods to calculate a frequency response function, FRF (e.g. Fig. 2d) [46], and (3) adjust parameters of a model of the balance control system represented by the Fig. 2a block diagram using a constrained optimization algorithm (Matlab “fmincon” function; The Mathworks, Inc., Natick, MA) until the model-derived FRF optimally matches the experimental FRF (e.g. model-derived FRFs in Fig. 2d) [22, 47]. This analysis procedure will be applied to each of the 8 individual tests performed by each subject. The full set of model parameters are defined in Table 2.

Table 2 Balance control model parameters
Fig. 2
figure 2

CSMI methods with example data and analysis. (a) A feedback control model forms the basis for identifying model parameters (Table 2) that account for experimentally evoked body-in-space (BS) sway, representing angular tilt of the body center-of-mass (CoM), evoked by support surface (SS) and/or visual surround (VS) rotations. (b) An example of one cycle of a pseudorandom surface-tilt stimulus that evoked the CoM body sway (averaged across 5 stimulus cycles) shown in (c) for a Control subject and TBI subject. (d) Frequency domain analysis of stimulus/response data in (b) and (c) yields frequency response functions (FRFs) expressed as gain (ratio of CoM response amplitude to stimulus amplitude) and phase (timing of response relative to stimulus) measures as a function of frequency components in the pseudorandom stimulus. Parameters of the model in (a) are calculated by a fit procedure that finds parameters that optimally account for the FRF gain and phase data

The parameters include sensory weights that represent the relative contributions of proprioceptive, visual, and vestibular sensory information to balance control such that in each of the 4 test conditions the sum of weights equals 1. These weights are known to vary with test conditions that alter the availability (e.g. eyes open or closed) or reliability (e.g. varying stimulus amplitudes) of sensory information and are affected by sensory abnormalities (22, 24). For condition 1 tests (surface tilt stimulus with eyes closed) the model fit to the experimental FRF provides a measure of the proprioceptive contribution (Wprop). With eyes closed, the visual contribution (Wvis) is zero and therefore the vestibular contribution is Wvest = 1 – Wprop. Condition 2 tests also use a surface tilt stimulus but eyes are open with both visual and vestibular cues providing accurate orientation information. The model fit again provides an estimate of Wprop but now 1 – Wprop = Wvis + Wvest. Condition 3 tests present a visual tilt stimulus where the model fit provides a measure of Wvis and 1 – Wvis = Wprop + Wvest. Finally condition 4 tests simultaneously present the same surface and visual-tilt stimulus. In this condition the model fit provides a measure of Wprop + Wvis with Wvest = 1 – (Wprop + Wvis).

The neural controller parameters Kp and Kd (Table 2) represent the transformation from an internal sensory error signal (i.e. deviation from a desired body orientation) to a corrective ankle torque the moves the body toward a body orientation that reduces the sensory error. The neural controller parameters (Kp and Kd) are known to scale with subject anthropometrics [22] and, therefore, will be normalized prior to performing across subject comparisons. The balance model also assumes that a sense of effort contributes to the sensory error signal. This effort sense is represented by a torque feedback pathway with parameter Kt that is a scaling factor applied to the mathematical integration of the corrective ankle torque signal. This feedback pathway accounts for the low frequency characteristics of the FRFs that show gain declines and phase advances at frequencies below about 0.1 Hz (Fig. 2d). The time delay parameter td represents all delays in the system including sensory transduction, neural transmission, central processing, and muscle activation. The time delay parameter accounts for a substantial portion of the increasing FRF phase lag with increasing frequency.

Sample size

Power analysis for Study 1 was derived from a pilot dataset of 12 control subjects who participated in CSMI tests with the goal of estimating the sample size necessary for a normative reference group. This was defined as the number of control subjects necessary to see a coefficient of variation (CV) of less than 10%, a common threshold for measurement precision, in the CMSI outcome measures in order to develop a robust reference control group with a reasonable and acceptable amount of variability (49). Estimation was done with bootstrapping techniques using the stiffness control parameter (Kp) of the CMSI test of the pilot cohort as a sample distribution. For a considered sample size (ranging from 10 to 100 subjects), 1000 random samples with replacement were taken from the cohort’s Kp dataset to simulate a null distribution at that sample size and estimate the corresponding CV. Based on these permuted samples, the CV of the Kp parameter is expected to be less than 10% with 65 subjects, giving an adequately representative reference group. This same sample size of 65 will allow for a CV of ~6.5% in other CSMI measures such as visual sensory weighting. Therefore, a cohort of 65 mTBI subjects should be more than sufficient to properly identify any underlying classification differences arising from central sensorimotor integration abnormalities.

Secondary outcome measures

Secondary outcome measures of symptomology, neurocognition, static and dynamic balance, and central sensorimotor integration are provided in Table 3.

Table 3 List of all outcome measures by domain

Static and dynamic balance testing

Subjects will be instrumented with body-worn inertial sensors during the performance of static and dynamic balance tests. Five synchronized wireless Opal inertial sensors (APDM, Inc., Portland, OR) will be affixed to the participants’ head, sternum, lumbar, left and right feet using elastic straps. Data will be collected at 128 Hz and transferred to a laptop for automatic generation of gait and balance measures by Mobility Lab software [48] and additional analysis of the raw time-series data. Specific tests and instrumented measures of static and dynamic balance are presented in Table 3.

Clinical testing for classification and covariates

To classify mTBI participants as with or without vestibular dysfunction, standard vestibular testing will be performed as described in Table 4. Clinical tests to examine potential covariates of sensory loss will also be included at the initial visit in Study 1.

Table 4 List of clinical sensory and vestibular tests for classification and covariates / comorbidities of primary and secondary outcomes

Statistical analysis

To determine if objective/novel measures of balance will better distinguish people with mTBI from control subjects than clinical measures, we will use logistic regression classification models. Improvements in classification will be determined by calculating the area under the curve (AUC) of the receiver-operator characteristic and then comparing AUC for each traditional clinical measure with the novel objective measures. Significance will be assessed using bootstrapping to determine if the observed differences in AUC are greater than would be expected by chance at a level of 0.05.

To determine if a subset of people with mTBI will have abnormal CSMI test measures, even without peripheral vestibular or ocular motor deficits, we will use descriptive statistics to characterize the distribution of CSMI test measures and vestibular/ocular motor scores. We will define CSMI test measures and vestibular/ocular motor scores as abnormal if the scores are greater than 2 standard deviations from the mean in controls. We will then classify mTBI subjects into one of 4 CSMI-vestibular/ocular motor categories 1) abnormal CSMI –abnormal vestibular/ocular motor scores, 2) abnormal CSMI –normal vestibular/ocular motor scores, 3) normal CSMI –abnormal vestibular/ocular motor scores, 4) normal CSMI –normal vestibular/ocular motor scores.

To determine if the observed relationship between poorer static/dynamic balance performance and mTBI is regulated/mediated by central sensorimotor integration, we will use mediation analysis. Mediation analysis [49] will allow for the partitioning of the association between mTBI and chronic imbalance into two parts, the direct relationship of mTBI on balance impairment as well as the causally mediated effect of mTBI on central sensorimotor integration, which, in turn is having the actual effect on imbalance. These varying generative mechanisms of balance deficits will be assessed by utilizing linear models relating mTBI to CSMI test measures alongside models of mTBI and CSMI test measures jointly affecting imbalance using the apportionment methods described by Baron and Kenny [49].

Study 2: Rehabilitation of balance deficit

Assessment procedures

Subjects with mTBI who participate in Study 1 will be screened for Study 2. An investigator will verbally explain the consent form and time commitments involved with the rehabilitation, allow the person ample time to read through the consent form and then will acknowledge consent by signing the form. For Study 2, 40 mTBI participants will participate in a 6-week rehabilitation program randomized to either 1) a standard vestibular focused rehabilitation program or 2) the same standard vestibular focused rehabilitation program with the addition of audio biofeedback. Post-rehabilitation assessments will be conducted one week after the conclusion of the program. Retention will be assessed 6 weeks following the post-rehabilitation session. A flowchart illustrating the study design is depicted in Fig. 3.

Fig. 3
figure 3

Flowchart illustrating the study design. Study 1 is a cross-sectional design. Study 2 recruits subjects from Study 1 for a randomized intervention

Intervention

Participants will be randomized into either 1) a standard of care vestibular rehabilitation program or 2) a standard or care vestibular rehabilitation plus ABF training program. Rehabilitation will begin approximately 7-10 days after pretest assessment. Both groups will receive the rehabilitation that includes progressive (3 levels of difficulty) exercises that target gaze stabilization, vestibular stimulation, balance and proprioceptive retraining [5052] (Table 5). Subjects randomized to the ABF training, will undergo the same vestibular rehabilitation program but will wear an ABF device while performing the rehabilitation exercises. All researchers performing post-rehabilitation testing will remain blinded to the randomized group assignment throughout the duration of the study.

Table 5 Rehabilitation protocol

Auditory biofeedback

ABF will be supplied by a lumbar-mounted smartphone (mHealth Technologies s.r.l., Bologna, Italy). The ABF system detects AP and mediolateral (ML) linear accelerations near the body’s CoM. The participant wears headphones and the ML inclination with respect to gravity is encoded as a sound in either the right or left ear while the AP tilt is encoded as changes in pitch as the person leans forward or backward. When the body is in perfect equilibrium, the system is quiet. A calibration test is used to determine optimal balance parameters such that the auditory feedback increases only when the person exceeds their baseline sway by 1° or the tilt of the trunk during gait exceeds an established threshold [42, 53]. Both AP and ML feedback will be supplied to the participant during static balance exercises. During dynamic balance exercises, only ML feedback will be provided. During balance exercises, participants will be instructed to utilize the auditory sounds provided by the ABF system to maintain balance and keep the system quiet (i.e., minimal deviations from the stable calibration).

Primary and secondary outcome measures

The primary outcome measure is the change between pre-rehabilitation, post-rehabilitation, and 6-week follow-up testing of the model-based parameters of the novel CSMI test (Table 2). Secondary outcome measures are the changes between pre-rehabilitation, post-rehabilitation, and 6-week follow up testing of the clinical and objective balance measures described in Study 1 and in Table 3. Additional secondary outcome measures from the at-home inertial sensors will include activity rate and average turning speed and number of turns.

Sample size

We powered the intervention study to detect improvement in the CSMI test deficit we found in our pilot subjects. Conservatively, 20 subjects will be recruited for both the experimental treatment and standard of care groups to provide results that are capable of powering a more in-depth future clinical trial investigating rehabilitation for balance deficits in mTBI patients. The CSMI test deficits observed in 3 mTBI pilot subjects and 12 healthy controls were used to power the study. Specifically, the CSMI parameter Kp under open eye surface tilting (t = -2.02, p = 0.065) and visual sensory weight Wvis under fixed surface visual tilting (t = 2.07, p = 0.060) were used to calculate sample size based on independent sample t-tests. Rehabilitation-based differences within the TBI cohort are not expected to be as extensive as those identified between mTBI subjects and controls. However, an effect size at 50% of the magnitude of the difference observed between mTBI and controls in the preliminary data (Coehn’s d = 0.748) would still yield 72% power with α = 0.10 in a repeated measures framework in as little as 10 subjects per group. Conservatively, we will recruit 20 subjects each for the experimental treatment and standard of care groups to provide results that are capable of powering a more in-depth future study investigating rehabilitation for balance deficits in mTBI patients.

Statistical analysis

To determine if ABF intervention improves central sensorimotor integration, we will build linear regression models to simultaneously evaluate the change in CSMI measures reflecting central sensorimotor integration and the effect of ABF compared to standard of care. Our outcome will be the change in pre- versus post-rehabilitation program CSMI measures; using the change (difference) in CSMI measures will control for the correlation of the repeated pre- and post- measures within an individual as well as maximizing the number of covariates we can potentially add into the regression models. Covariates of interest are vestibular/ocular motor scores, PTSD, Veteran versus civilian status, age, and gender. In addition, we will determine whether rehabilitation improved the static and dynamic balance metrics that were measured at the pre- and post-rehabilitation testing visits.

To determine if people with CSMI impairment show sustained improvement in static and dynamic balance after rehabilitation, even after ABF is discontinued, we will use linear mixed effects regression modeling which will include a third time point of interest (6 weeks after the rehabilitation intervention has stopped). Time will be treated as a categorical variable in order to evaluate changes in CSMI scores and balance during specific time periods, in this case, between the cessation of the intervention and the 6 weeks post-intervention follow-up. We will explicitly test the significance of a group by time interaction to determine if participants in the ABF rehabilitation group showed sustained improvements in outcomes beyond that experienced by participants with the standard of care. If applicable, we will then stratify the analysis by those with abnormal/normal measures of vestibular/ocular motor function to assess the relationship between vestibular/ocular motor function deficits and sustained improvement in CSMI score and balance.

Discussion

The goal of this study is to determine whether persistent balance and gait deficits following mTBI are attributable to improper central sensorimotor integration and to examine the potential for augmented rehabilitation to improve daily activity levels. In particular, this study is unique in its investigation of central sensorimotor integration in people with persistent balance complaints following mTBI in that confounding covariates such as vestibular disruption, neurocognition, and physical activity will be objectively quantified. Additionally, if we find that the quantitative balance, gait, or novel CSMI measures have a significantly greater AUC compared to the clinical measures (BESS, gait speed, SOT) at detecting patients with chronic mTBI, this will indicate the necessity for objective quantification of gait and balance deficits for mTBI assessment.

If a portion of subjects with chronic balance problems after mTBI have abnormal CSMI test measures but normal clinical vestibular testing results, this suggests that sensory integration processes in the brain can be impaired following mTBI, even without peripheral sensory deficits, and that abnormal central sensorimotor integration plays a key role in persistent abnormal balance in subjects with normal vestibular function. If a portion of mTBI subjects have abnormal CSMI test measures and abnormal clinical vestibular function, the key mediator for balance problems is less clear and may be due to the peripheral vestibular dysfunction, abnormal central sensorimotor integration, or inadequate adaptation of central sensorimotor integration to account for abnormal sensory information. If we find that abnormal central sensorimotor integration mediates abnormal static or dynamic balance, rehabilitation will need to be focused on improving central sensorimotor integration to improve balance. If we find that CSMI scores improve with rehabilitation, we will conclude that central sensorimotor integration strategies are flexible and can improve with training. Further, if we find that central sensorimotor integration and gait and balance jointly improve with rehabilitation, we will have further evidence that central sensorimotor integration plays a key role in the mediation of gait and balance following mTBI. If we find a greater improvement in the ABF group, we will conclude that ABF may facilitate improved rehabilitation. Finally, if we find that people with central sensorimotor integration impairment maintain improvements after stopping the use of ABF device, we will conclude that a recalibration of sensory integration and/or sensorimotor transformation for balance has occurred, rather than sensory substitution that requires active ABF. If we find that sensory strategies can change with practice and biofeedback, we will conclude that this targeted approach to balance rehabilitation should be explored further in a larger, randomized trial. It is possible that different subclasses based on CSMI measures respond differently to rehabilitation or ABF-augmented rehabilitation. In such a case, the results of this study will provide valuable information for powering an in-depth future study carefully examining the relationship between CSMI subclasses and the response to rehabilitation with and without ABF. The results of this study will further our understanding of the mechanisms underlying posture and gait dysfunction in individuals with chronic mTBI and inform future studies about the efficacy of ABF augmented rehabilitation programs.

Abbreviations

ABF:

Auditory biofeedback

ANAM:

Automated neuropsychological assessment metrics

AP:

Anteroposterior

AUC:

Area under the receiver-operator characteristic curve

BDI-II:

Becks depression inventory

BESS:

Balance error scoring system

CoM:

Center of mass

CSMI test:

Central sensorimotor integration test

CTSIB:

Clinical test of sensory organization and balance

CV:

Coefficient of variation

DHI:

Dizziness handicap inventory

FRF:

Frequency response function

LOC:

Loss of consciousness

ML:

Mediolateral

mTBI:

Mild traumatic brain injury

NSI:

Neurobehavioral symptom inventory

PLI:

Pain location inventory

PTSD:

Post-Traumatic stress disorder

RMS:

Root-mean-square

SCAT3:

Sport concussion assessment Tool 3

SF-36:

Short form 36

SOT:

Sensory organization Test

TBI:

Traumatic brain injury.

References

  1. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of traumatic brain injury: a brief overview. J Head Trauma Rehabil. 2006;21(5):375–8.

    Article  PubMed  Google Scholar 

  2. Hoge CW, McGurk D, Thomas JL, Cox AL, Engel CC, Castro CA. Mild traumatic brain injury in U.S. Soldiers returning from Iraq. N Engl J Med. 2008;358(5):453–63.

    Article  CAS  PubMed  Google Scholar 

  3. Coronado VG, McGuire LC, Sarmiento K, Bell J, Lionbarger MR, Jones CD, Geller AI, Khoury N, Xu L. Trends in Traumatic Brain Injury in the U.S. and the public health response: 1995-2009. J Saf Res. 2012;43(4):299–307.

    Article  Google Scholar 

  4. Boake C, McCauley SR, Levin HS, Pedroza C, Contant CF, Song JX, Brown SA, Goodman H, Brundage SI, Diaz-Marchan PJ. Diagnostic criteria for postconcussional syndrome after mild to moderate traumatic brain injury. J Neuropsychiatry Clin Neurosci. 2005;17(3):350–6.

    Article  PubMed  Google Scholar 

  5. Ingebrigtsen T, Waterloo K, Marup-Jensen S, Attner E, Romner B. Quantification of post-concussion symptoms 3 months after minor head injury in 100 consecutive patients. J Neurol. 1998;245(9):609–12.

    Article  CAS  PubMed  Google Scholar 

  6. Fife TD, Kalra D. Persistent vertigo and dizziness after mild traumatic brain injury. Ann N Y Acad Sci. 2015;1343:97–105.

    Article  PubMed  Google Scholar 

  7. Guskiewicz KM, Weaver NL, Padua DA, Garrett WE. Epidemiology of concussion in collegiate and high school football players. Am J Sports Med. 2000;28(5):643–50.

    Article  CAS  PubMed  Google Scholar 

  8. Daneshvar DH, Riley DO, Nowinski CJ, McKee AC, Stern RA, Cantu RC. Long-term consequences: effects on normal development profile after concussion. Phys Med Rehabil Clin N Am. 2011;22(4):683–700.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Giza CC, Kutcher JS, Ashwal S, Barth J, Getchius TS, Gioia GA, Gronseth GS, Guskiewicz K, Mandel S, Manley G. Summary of evidence-based guideline update: Evaluation and management of concussion in sports Report of the Guideline Development Subcommittee of the American Academy of Neurology. Neurology. 2013;80(24):2250–7.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Montgomery E, Fenton G, McClelland R, MacFlynn G, Rutherford W. The psychobiology of minor head injury. Psychol Med. 1991;21(02):375–84.

    Article  CAS  PubMed  Google Scholar 

  11. McGowan JC, Yang JH, Plotkin RC, Grossman RI, Umile EM, Cecil KM, Bagley LJ. Magnetization transfer imaging in the detection of injury associated with mild head trauma. Am J Neuroradiol. 2000;21(5):875–80.

    CAS  PubMed  Google Scholar 

  12. Grossman EJ, Ge Y, Jensen JH, Babb JS, Miles L, Reaume J, Silver JM, Grossman RI, Inglese M. Thalamus and cognitive impairment in mild traumatic brain injury: a diffusional kurtosis imaging study. J Neurotrauma. 2012;29(13):2318–27.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Peskind ER, Petrie EC, Cross DJ, Pagulayan K, McCraw K, Hoff D, Hart K, Yu C-E, Raskind MA, Cook DG, et al. Cerebrocerebellar hypometabolism associated with repetitive blast exposure mild traumatic brain injury in 12 Iraq war Veterans with persistent post-concussive symptoms. NeuroImage. 2011;54, Supplement 1:S76–82.

    Article  Google Scholar 

  14. Basta D, Todt I, Scherer H, Clarke A, Ernst A. Postural control in otolith disorders. Hum Mov Sci. 2005;24(2):268–79.

    Article  PubMed  Google Scholar 

  15. Davies RA, Luxon LM. Dizziness following head injury: A neuro-otological study. J Neurol. 1995;242(4):222–30.

    Article  CAS  PubMed  Google Scholar 

  16. Ernst A: Sensory disorders of the auditory and vestibular systems following blunt head trauma. In: IUTAM Symposium on Impact Biomechanics: From Fundamental Insights to Applications: 2005. The Netherlands: Springer. 2005:213–20.

  17. Toglia JU, Rosenberg PE, Ronis ML. Posttraumatic dizziness: Vestibular, audiologic, and medicolegal aspects. Arch Otolaryngol. 1970;92(5):485–92.

    Article  CAS  PubMed  Google Scholar 

  18. Drew AS, Langan J, Halterman C, Osternig LR, Chou L-S, van Donkelaar P. Attentional disengagement dysfunction following mTBI assessed with the gap saccade task. Neurosci Lett. 2007;417(1):61–5.

    Article  CAS  PubMed  Google Scholar 

  19. Suh M, Basu S, Kolster R, Sarkar R, McCandliss B, Ghajar J, Cognitive, Consortium NR. Increased oculomotor deficits during target blanking as an indicator of mild traumatic brain injury. Neurosci Lett. 2006;410(3):203–7.

    Article  CAS  PubMed  Google Scholar 

  20. Suh M, Kolster R, Sarkar R, McCandliss B, Ghajar J, Cognitive, Consortium NR. Deficits in predictive smooth pursuit after mild traumatic brain injury. Neurosci Lett. 2006;401(1):108–13.

    Article  CAS  PubMed  Google Scholar 

  21. Capó-Aponte JE, Urosevich TG, Temme LA, Tarbett AK, Sanghera NK. Visual dysfunctions and symptoms during the subacute stage of blast-induced mild traumatic brain injury. Mil Med. 2012;177(7):804–13.

    Article  PubMed  Google Scholar 

  22. Peterka RJ. Sensorimotor Integration in Human Postural Control. J Neurophysiol. 2002;88(3):1097–118.

    CAS  PubMed  Google Scholar 

  23. Peterka RJ, Loughlin PJ. Dynamic Regulation of Sensorimotor Integration in Human Postural Control. J Neurophysiol. 2004;91(1):410–23.

    Article  PubMed  Google Scholar 

  24. Peterka R, Statler K, Wrisley D, Horak F. Postural compensation for unilateral vestibular loss. Front Neurol. 2011;2:57.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Finnoff JT, Peterson VJ, Hollman JH, Smith J. Intrarater and interrater reliability of the Balance Error Scoring System (BESS). PM&R. 2009;1(1):50–4.

    Article  Google Scholar 

  26. McCrea M, Barr WB, Guskiewicz K, Randolph C, Marshall SW, Cantu R, Onate JA, Kelly JP. Standard regression-based methods for measuring recovery after sport-related concussion. J Int Neuropsychol Soc. 2005;11(1):58–69.

    Article  PubMed  Google Scholar 

  27. Basford JR, Chou L-S, Kaufman KR, Brey RH, Walker A, Malec JF, Moessner AM, Brown AW. An assessment of gait and balance deficits after traumatic brain injury. Arch Phys Med Rehabil. 2003;84(3):343–9.

    Article  PubMed  Google Scholar 

  28. Cavanaugh J, Guskiewicz K, Giuliani C, Marshall S, Mercer V, Stergiou N. Detecting altered postural control after cerebral concussion in athletes with normal postural stability. Br J Sports Med. 2005;39(11):805–11.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. De Beaumont L, Mongeon D, Tremblay S, Messier J, Prince F, Leclerc S, Lassonde M, Theoret H. Persistent motor system abnormalities in formerly concussed athletes. J Athl Train. 2011;46(3):234–40.

    PubMed  PubMed Central  Google Scholar 

  30. Sosnoff JJ, Broglio SP, Shin S, Ferrara MS. Previous mild traumatic brain injury and postural-control dynamics. J Athl Train. 2011;46(1):85.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Fino PC, Nussbaum MA, Brolinson PG. Decreased high-frequency center-of-pressure complexity in recently concussed asymptomatic athletes. Gait Posture. 2016;50:69–74.

    Article  PubMed  Google Scholar 

  32. Catena RD, van Donkelaar P, Chou L-S. Cognitive task effects on gait stability following concussion. Exp Brain Res. 2007;176(1):23–31.

    Article  PubMed  Google Scholar 

  33. Howell DR, Osternig LR, Chou L-S. Dual-Task Effect on Gait Balance Control in Adolescents With Concussion. Arch Phys Med Rehabil. 2013;94(8):1513–20.

    Article  PubMed  Google Scholar 

  34. Parker TM, Osternig LR, Van Donkelaar P, Chou L. Gait stability following concussion. Med Sci Sports Exerc. 2006;38(6):1032.

    Article  PubMed  Google Scholar 

  35. Fino PC. A preliminary study of longitudinal differences in local dynamic stability between recently concussed and healthy athletes during single and dual-task gait. J Biomech. 2016;49(9):1983–8.

    Article  PubMed  Google Scholar 

  36. Alberts JL, Hirsch JR, Koop MM, Schindler DD, Kana DE, Linder SM, Campbell S, Thota AK. Using accelerometer and gyroscopic measures to quantify postural stability. J Athl Train. 2015;50(6):578–88.

    Article  PubMed  PubMed Central  Google Scholar 

  37. Alberts JL, Thota A, Hirsch J, Ozinga S, Dey T, Schindler DD, Koop MM, Burke D, Linder SM. Quantification of the Balance Error Scoring System with Mobile Technology. Med Sci Sports Exerc. 2015;47(10):2233–40.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Howell D, Osternig L, Chou L-S. Monitoring recovery of gait balance control following concussion using an accelerometer. J Biomech. 2015;48(12):3364–8.

    Article  PubMed  Google Scholar 

  39. King LA, Horak FB, Mancini M, Pierce D, Priest KC, Chesnutt J, Sullivan P, Chapman JC. Instrumenting the Balance Error Scoring System for Use With Patients Reporting Persistent Balance Problems After Mild Traumatic Brain Injury. Arch Phys Med Rehabil. 2014;95(2):353–9.

  40. Horak F, King L, Mancini M. Role of body-worn movement monitor technology for balance and gait rehabilitation. Phys Ther. 2015;95(3):461–70.

    Article  PubMed  Google Scholar 

  41. Mancini M, Horak FB. The relevance of clinical balance assessment tools to differentiate balance deficits. Eur J Phys Rehabil Med. 2010;46(2):239.

    CAS  PubMed  PubMed Central  Google Scholar 

  42. Dozza M, Chiari L, Horak FB. Audio-biofeedback improves balance in patients with bilateral vestibular loss. Arch Phys Med Rehabil. 2005;86(7):1401–3.

    Article  PubMed  Google Scholar 

  43. Dozza M, Horak FB, Chiari L. Auditory biofeedback substitutes for loss of sensory information in maintaining stance. Exp Brain Res. 2007;178(1):37–48.

    Article  PubMed  Google Scholar 

  44. Kansagara D, Maya Elin ON, Carlson K, Storzbach D, Brenner L, Freeman M, Quiñones A, Motu’apuaka M: Complications of Mild Traumatic Brain Injury in Veterans and Military Personnel: A Systematic Review. 2013.

  45. Katzman R, Brown T, Fuld P, Peck A, Schechter R, Schimmel H. Validation of a short Orientation-Memory-Concentration Test of cognitive impairment. Am J Psychiatry. 1983;140(6):734–9.

    Article  CAS  PubMed  Google Scholar 

  46. Pintelon R, Schoukens J. Measurement of frequency response functions using periodic excitations, corrupted by correlated input/output errors. IEEE Trans Instrum Meas. 2001;50(6):1753–60.

    Article  Google Scholar 

  47. Peterka RJ. Simplifying the complexities of maintaining balance. IEEE Eng Med Biol Mag. 2003;22(2):63–8.

    Article  PubMed  Google Scholar 

  48. Mancini M, King L, Salarian A, Holmstrom L, McNames J, Horak FB. Mobility lab to assess balance and gait with synchronized body-worn sensors. J Bioeng Biomed Sci. 2011;Suppl 1:007.

  49. Baron RM, Kenny DA. The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51(6):1173.

    Article  CAS  PubMed  Google Scholar 

  50. Herdman SJ. Role of vestibular adaptation in vestibular rehabilitation. Otolaryngol Head Neck Surg. 1998;119(1):49–54.

    Article  CAS  PubMed  Google Scholar 

  51. Shepard NT, Telian SA. Programmatic vestibular rehabilitation. Otolaryngol Head Neck Surg. 1995;112(1):173–82.

    Article  CAS  PubMed  Google Scholar 

  52. Weightman MM, Bolgla R, McCulloch KL, Peterson MD. Physical therapy recommendations for service members with mild traumatic brain injury. J Head Trauma Rehabil. 2010;25(3):206–18.

    Article  PubMed  Google Scholar 

  53. Mayagoitia RE, Nene AV, Veltink PH. Accelerometer and rate gyroscope measurement of kinematics: an inexpensive alternative to optical motion analysis systems. J Biomech. 2002;35(4):537–42.

    Article  PubMed  Google Scholar 

  54. Jacobson GP, Newman CW. The development of the dizziness handicap inventory. Arch Otolaryngol Head Neck Surg. 1990;116(4):424–7.

    Article  CAS  PubMed  Google Scholar 

  55. Weathers FW, Litz BT, Herman DS, Huska JA, Keane TM: The PTSD Checklist (PCL): Reliability, validity, and diagnostic utility. In: annual convention of the international society for traumatic stress studies: 1993: International Society for Traumatic Stress Studies San Antonio; 1993.

  56. Bennett RM, Friend R, Marcus D, Bernstein C, Han BK, Yachoui R, Deodhar A, Kaell A, Bonafede P, Chino A. Criteria for the diagnosis of fibromyalgia: validation of the modified 2010 preliminary American College of Rheumatology criteria and the development of alternative criteria. Arthritis Res. 2014;66(9):1364–73.

    Article  Google Scholar 

  57. Friend R, Bennett RM. Distinguishing fibromyalgia from rheumatoid arthritis and systemic lupus in clinical questionnaires: an analysis of the revised Fibromyalgia Impact Questionnaire (FIQR) and its variant, the Symptom Impact Questionnaire (SIQR), along with pain locations. Arthritis Res Ther. 2011;13(2):1.

    Article  Google Scholar 

  58. McCrory P, Meeuwisse WH, Aubry M, Cantu B, Dvořák J, Echemendia RJ, Engebretsen L, Johnston K, Kutcher JS, Raftery M. Consensus statement on concussion in sport: the 4th International Conference on Concussion in Sport held in Zurich, November 2012. Br J Sports Med. 2013;47(5):250–8.

    Article  PubMed  Google Scholar 

  59. Beck AT, Steer RA, Brown GK. Beck depression inventory-II. San Antonio: The Psychological Corporation. 1996:78204–2498.

  60. McHorney CA, Ware JE, Lu J, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). Med Care. 1994;32(1):40–66.

    Article  CAS  PubMed  Google Scholar 

  61. 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.

    Article  Google Scholar 

  62. Reeves DL, Winter KP, Bleiberg J, Kane RL. ANAM® Genogram: Historical perspectives, description, and current endeavors. Arch Clin Neuropsychol. 2007;22:15–37.

    Article  Google Scholar 

  63. Eckner JT, Kutcher JS, Richardson JK. Pilot evaluation of a novel clinical test of reaction time in National Collegiate Athletic Association Division I football players. J Athl Train. 2010;45(4):327–32.

    Article  PubMed  PubMed Central  Google Scholar 

  64. Shumway-Cook A, Horak FB. Assessing the influence of sensory interaction on balance. Phys Ther. 1986;66(10):1548–50.

    CAS  PubMed  Google Scholar 

  65. Nashner LM, Peters JF. Dynamic posturography in the diagnosis and management of dizziness and balance disorders. Neurol Clin. 1990;8(2):331–49.

    CAS  PubMed  Google Scholar 

Download references

Acknowledgements

The authors would like to thank audiologists Sean Kampel and Marco Jurado, and research assistants Heather Belding, Merissa Walls, Emily Sippel, and Clayton Swanson for assisting in set up and data collection.

Funding

This work was supported by the Assistant Secretary of Defense for Health Affairs under Award No. W81XWH-15-1-0620. Opinions, interpretations, conclusions and recommendations are those of the author and are not necessarily endorsed by the Department of Defense.

Availability of data and materials

De-identified data generated from this study will be deposited into the Federal Interagency Traumatic Brain Injury Research (FITBIR) Informatics System.

Authors’ contributions

PF participated in design regarding gait and balance testing of study and drafting of the manuscript; RP participated in design of study, development of the novel CSMI test, and editing of the manuscript; TH participated in design regarding vestibular testing of study and editing of the manuscript, FH participated in design of the study and editing of the manuscript; CM was responsible for power analysis and will be the primary statistician responsible for analyzing data for this study, JC participated in design of study, editing of the manuscript; LK conceived of the study, participated in design of study, editing of the manuscript. All authors read and approved the manuscript.

Competing interests

Dr. Horak has a significant financial interest in, and is an employee of ADPM, a company that has a commercial interest in the results of this research and technology. This potential institutional and individual conflict has been reviewed and managed by OHSU and the VA Portland Health Care System. No other authors have a competing interest.

Consent for publication

Not applicable.

Ethics approval and consent to participate

Informed written consent will be obtained from all participants. An investigator will verbally explain the consent form, allow the person ample time to read through the consent form and then will acknowledge consent by signing the form. All protocols have been approved by a joint OHSU and VA Portland Health Care System Institutional Review Board (Study #00015010).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Laurie A. King.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Fino, P.C., Peterka, R.J., Hullar, T.E. et al. Assessment and rehabilitation of central sensory impairments for balance in mTBI using auditory biofeedback: a randomized clinical trial. BMC Neurol 17, 41 (2017). https://doi.org/10.1186/s12883-017-0812-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12883-017-0812-7

Keywords