Sleep after TBI: How the TBI Model Systems have advanced the field

Abstract

BACKGROUND:

Identification and management of comorbidities in TBI has become an increasing focus for optimizing TBI outcomes. Recent meta-analyses highlight sleep disturbance and sleep disorders following TBI (Mathias & Alvaro, 2012). Improving the recognition and treatment of sleep disorders in TBI should be a central focus of rehabilitation. The Traumatic Brain Injury Model System (TBIMS) has created an infrastructure allowing multi-center investigations into sleep dysfunction in those who have had a moderate to severe TBI and received inpatient rehabilitation.

OBJECTIVE:

This paper will describe the 1) infrastructure used to advance sleep dysfunction/disorders research following TBI, 2) preliminary findings from these studies, and 3) repository of data which can be accessed for secondary analyses by investigators outside of the TBIMS infrastructure.

METHODS:

Two internal mechanisms allow investigators at TBIMS sites to collaborate on projects of shared interest: Research Modules and Special Interest Groups (SIG).

RESULTS:

To date, five studies have resulted from the TBIMS collaborative process focusing on insomnia, circadian disruption, and sleep apnea.

CONCLUSIONS:

Keywords

Sleep disorders sleep dysfunction traumatic brain injury traumatic brain injury model systems

1 Background

In 2011, the Institute of Medicine (IOM), at the request of the Department of Veterans Affairs (VA), published a review of the literature to critically evaluate the neurologic, cognitive, and psychological/behavioral effects of traumatic brain injury (TBI). The IOM report indicated that TBI was associated with many chronic health conditions and disabilities (Rutherford & Corrigan, 2009). Identification and management of comorbidities in TBI has become an increasing focus for optimizing TBI outcomes. Recent meta-analyses highlight sleep disturbance and sleep disorders following TBI (Mathias & Alvaro, 2012); however, sleep dysfunction is also associated with accidents leading to TBI. Nonetheless, animal studies highlight the critical role that restorative sleep plays in facilitating mechanisms of neural repair post-neurologic injury and early neurodegeneration (Dash et al., 2009; Longordo et al., 2009; McDermott et al., 2003; Walker, 2009; Xie et al., 2013; Zunzunegui et al., 2011). Therefore, improving the recognition and treatment of sleep disorders in TBI should be a central focus of rehabilitation. The Traumatic Brain Injury Model System (TBIMS) has created an infrastructure allowing multi-center investigations into sleep dysfunction in those who have had a moderate to severe TBI and received inpatient rehabilitation.

2 Introduction to the TBIMS

2.1 History

The TBIMS was formulated in 1987 by the National Institute for Handicapped Research (NIHR; now the National Institute on Disability, Independent Living, and Rehabilitation Research, NIDILRR) to 1) demonstrate the components of health system management of TBI from hyperacute care to rehabilitation and integration of the injured person into the community, 2) establish research programs on the treatment of TBI, 3) demonstrate models of care for rehabilitation, and 4) to contribute to a longitudinal database to examine outcomes after rehabilitation treatment. Centers have comprehensive, integrated interdisciplinary clinical care that is closely associated with research and outcome measurement. Originally founded with 5 centers, the contributors to this database, have grown to involve 16 full centers and 3 follow up centers. Criteria for inclusion include age >16 years; medical documentation of post-traumatic amnesia >24 hours, CT findings of injury, or a loss of consciousness >30 minutes; admission to the model system site within 72 hours of injury; and informed consent (Dijkers et al., 2010).

Data is obtained from medical records and self-report at baseline (before discharge from acute rehabilitation), and at one, two, five years post-injury and every five years thereafter. The longitudinal database, the first of its kind in the United States, was developed with a comprehensive data dictionary; the data and dictionary are available to outside investigators as well. Over time, a number of cross-walk variable matches with other databases have been published and additional items derived from the Common Data Elements set developed by NIH have been incorporated. The database has permitted validation of a number of measures for the population with TBI as well as the development of new, population-specific measures.

2.2 Current structure – NIDILRR and Department of Veterans Affairs Polytrauma Rehabilitation Centers (VA PRCs)

During the decade following 2001 and the conflicts in Iraq and Afghanistan, which resulted in the identification of TBI as the signature injury for combat participants, discussions ensued regarding evaluating outcomes for the military and Veteran population. A major change in the structure of the TBI Model Systems occurred in 2008 when an interagency agreement was formulated between VA and NIDRR (then the National Institute for Disability and Rehabilitation Research) to develop a parallel longitudinal database for Veterans hospitalized with TBI. The five VA PRCs have become active partners in this research center with opportunities for comparison research and identification of specific prognostic indicators for the Veteran and civilian populations.

To date, over 16,000 subjects have been entered to date in the civilian TBIMS with outcomes available over 30 years. The VA PRCs have enrolled over 1100 subjects with outcomes available over the first 5-years. More than 850 peer-reviewed publications have derived from the longitudinal database and the single- and multi-site research projects that have been funded from this grant program; please see the TBIMS National Data and Statistical Center website (TBIMS NDSC)(NDSC, 2018b)). Additionally, an emphasis has been placed on the dissemination of information to both the scientific and medical communities and to consumers through the Model Systems Knowledge Translation Center (MSKTC) (MSKTC, 2018). The existence of the longitudinal database infrastructure has permitted the funding of a number of other large multicenter research trials funded by other agencies, such as the Patient-Centered Outcomes Research Institute (PCORI).

3 Multi-center study opportunities through TBIMS

One of the goals of the TBIMS is to utilize its nation-wide infrastructure to build capacity for multicenter studies. Two internal mechanisms allow investigators at TBIMS sites to collaborate on projects of shared interest: Research Modules, and Special Interest Groups (SIG).

Research Modules are time-limited studies involving 2 or more TBIMS Centers that begin and end during a single 5-year funding cycle. The Research Module process launched in 2007 when applicants were required to propose a module study as part of their TBIMS grant proposal. In the 2012–2017 grant cycle, funded centers proposed module projects at the start of the award period, and in the 2017–2022 grant cycle, the Research Module proposal process returned to being part of the TBIMS application. At the start of each cycle, Research Module proposals are shared with all funded centers, and a voting process is used to decide which modules will be implemented in that cycle. Each TBIMS Center is required to participate in at least one module study, and the center that proposed each chosen module will lead that project. Modules can also be proposed and implemented during the grant cycle if at least 2 centers have available resources and interest in participating. Modules are typically observational in nature, but can also include intervention studies.

SIGs are working groups of investigators across TBIMS centers who want to pursue projects, manuscripts, informational products and other collaborations in an area of shared interest. SIGs have also been used to develop new grant proposals. SIGs can be formed at any time, and have allotted meeting slots at bi-annual Project Directors Meetings as well as regularly scheduled conference calls to ensure continual progress outside of the Project Directors Meetings. The Sleep-Wake-Fatigue SIG was initially developed to facilitate collaborative projects. Members of the SIG have been involved in papers, joint presentations at conferences to facilitate interest and research, as well as sharing methodologies and other research challenges with advancing the science in this arena with TBI survivors.

Given that multiple centers have had an interest in sleep and fatigue after TBI, the module process has been used for two of the three studies in this area that have been conducted in the TBIMS. This strong interest also led to the development of a SIG on sleep, wake and fatigue after TBI. Together, the Sleep and Fatigue modules and the Sleep-Wake-Fatigue SIG have supported the development of several external multi-center grant applications. For example, a sub-group of the Sleep-Wake-Fatigue SIG developed a proposal to the PCORI, which led to the development of the third TBIMS study on sleep apnea.

4 Sleep disorder studies generated from the collaborative process

4.1 Insomnia

4.1.1 Description

This five-center observational study involved two cohorts of individuals with TBI with the goal of obtaining data at two follow-up time points. One cohort was followed at one and two years post-TBI and the other was followed at years two and five. Participants were contacted during their regular follow-up interview and asked if they would like to participate in an additional survey study regarding insomnia and fatigue after TBI.

4.1.2 Aims

The aims of this study were to provide information to support the development of new treatments for individuals with post-TBI insomnia and fatigue, including information about:

The relationship between fatigue and insomnia in individuals with TBI

Factors associated with time of onset of insomnia and onset of fatigue

The relationship between duration of insomnia and post-TBI fatigue (PTBIF)

The role of factors that may be predisposing (e.g., female gender, milder injury), precipitating (e.g., pain, psychiatric distress), perpetuating (e.g., poor sleep hygiene, substance abuse) and mitigating (e.g., sleep medication) for insomnia in individuals with TBI

The adequacy of existing treatments for post-TBI insomnia and fatigue

Outcomes (e.g., vocational status) associated with post-TBI insomnia and fatigue

4.1.3 Data elements

Willing participants completed additional questions shown in the Sleep/Wake category in Table 1. Other measures listed were used in data analysis projects stemming from this module but were gathered as part of the follow-up interviews that are a standard part of the TBI Model Systems data collection process.

Table 1
Insomnia Study Variables

Content Areas Measures

Sleep/Wake Pittsburgh Sleep Quality Index (PSQI)

Insomnia Severity Index (ISI)

Insomnia Interview Schedule (IIS)

Sleep Hygiene Index (SHI)

Epworth Sleepiness Scale (ESS)

Multidimensional Assessment of Fatigue (MAF)

Mood and Psychosocial Patient Health Questionnaire (PHQ-9) *

Generalized Anxiety Disorder Questionnaire (GAD-7) *

Satisfaction with Life Scale (SWLS) *

Participation Assessment with Recombined Tools - Objective (PART-O) *

Functional Status FIM™ Instrument *

Disability Rating Scale *

Content Areas	Measures
Sleep/Wake	Pittsburgh Sleep Quality Index (PSQI)
	Insomnia Severity Index (ISI)
	Insomnia Interview Schedule (IIS)
	Sleep Hygiene Index (SHI)
	Epworth Sleepiness Scale (ESS)
	Multidimensional Assessment of Fatigue (MAF)
Mood and Psychosocial	Patient Health Questionnaire (PHQ-9) *
	Generalized Anxiety Disorder Questionnaire (GAD-7) *
	Satisfaction with Life Scale (SWLS) *
	Participation Assessment with Recombined Tools - Objective (PART-O) *
Functional Status	FIM™ Instrument *
	Disability Rating Scale *

^*Included in the TBIMS follow-up interview.

4.1.4 Progress

While data collection has been completed for this study, the analyses are ongoing with several dissemination products to date. One of the most comprehensive publications generated from this data examined 334 individuals with TBI who completed 1-year (n = 213) or 2-year (n = 121) follow-up interviews between 2008 and 2012 (Cantor et al., 2012). Insomnia occurred in 11% to 24% of the sample and a large majority of these individuals with insomnia 1 or 2 years after injury (78–93%) did not report having insomnia prior to their TBI. Insomnia was very rarely present without fatigue (<4%) whereas PTBIF without insomnia occurred in over 20% of individuals in the 1 or 2 years samples. Insomnia and fatigue were both related to a number of factors including sleep disturbance, sleep hygiene, life satisfaction, anxiety, and depression.

The rich nature of the data from these longitudinal cohorts allows for additional investigations extending the core aims of the original study to answer questions about the psychometric properties of the measures used within a sample of individuals with TBI, something that was previously lacking in the literature. One such study examined the Multidimensional Assessment of Fatigue, a measure that was developed within the context of rheumatoid arthritis (Lequerica et al., 2012). A Rasch analysis in the TBI sample revealed disordered thresholds for the 10-point rating scale. Collapsing the ratings down to a 4-point scale improved the properties of the measure. This study also found a poor fit for an item reflecting the degree to which fatigue interfered with the ability to walk. It was suggested that this might be due to the origin of the measure within a rheumatoid arthritis sample where the experience of fatigue is more likely to interfere with walking as opposed to the nature of fatigue described by individuals with TBI as being more of a mental exhaustion.

Another study emerging from this dataset examined the factor structure of the Pittsburgh Sleep Quality Index (PSQI) (Lequerica et al., 2014). Although this widely used measure of sleep quality had been used in TBI studies in the past, the factor structure of the measure among individuals with TBI was previously unknown. While this study found support for a 3-factor structure previously reported in a non-TBI sample, a two-factor structure was also found to have an acceptable fit. This 2-factor model separated quantitative, time-related items from qualitative items measuring subjective impressions of sleep and daytime performance. This factor structure may have greater clinical utility in the assessment of sleep disturbances after TBI such as insomnia or circadian rhythm disorders. A more recent publication showed that the persistence of PTBIF 1–2 years post-injury was associated with greater disability, sleep disturbance, and depression while persistence beyond 2 years was associated with reduced community participation (Lequerica et al., 2017). Ongoing analyses are now looking into the factors associated with the remission of post-TBI insomnia.

4.2 Sleep efficiency

4.2.1 Description

In the inpatient rehabilitation setting, altered sleep cycles can disrupt therapy interventions, delay recovery, and affect other symptoms such as mood, memory and attention. Studies attempting to treat sleep disturbance and wakefulness after TBI have focused on pharmacologic treatment, with limited benefit. Additionally, the sedative-hypnotic medication used to treat sleep disorders in the general population are often not optimal after TBI given concerns for cognitive and behavioral side effects. Treatment in other populations with bright white light therapy (BWLT) in the morning have resulted in improved sleep. This study sought to examine the feasibility and initial efficacy of BWLT vs. red light therapy (RLT) comparison in the treatment of sleep disruption in TBI survivors in the acute inpatient rehabilitation setting.

Three sites participated in this randomized, single-blind, comparison module study with goal recruitment of 128 subjects. Participants on the inpatient rehabilitation unit received 30 minutes of BWLT or RLT within 2 hours of waking (0730-0930) for up to 10 days using a Litebook placed 12–24 inches from the subject’s face. The primary outcome was sleep efficiency as measured by wrist actigraph. Participants were between the ages of 18–70 years old, within 3 months of a moderate to severe TBI, and rated on clinical evaluation to have sleep disturbance (e.g. use of sleep medications or sleep disturbance noted by nursing). Patients with contraindications to BWLT (e.g. history of bipolar disorder or light sensitivity), blindness, or those at high risk for obstructive sleep apnea (based on standard screening measures) were excluded.

4.2.2 Aims

The specific aims were 1) in persons with TBI, prospectively compare overnight sleep efficiency in a cohort exposed to morning BWLT with a comparison group exposed to a neutral stimulus, RLT, in an acute inpatient rehabilitation setting and 2) explore the relationship between BWL exposure during inpatient rehabilitation after TBI with mood, therapy participation, attention, rate of functional recovery, and length of stay.

4.2.3 Data elements

Measures of nursing staff evaluation of sleep (Delirium Rating Scale (Del-R-98, Item 1) (Trzepacz et al., 2001) and risk for obstructive sleep apnea Berlin Questionnaire (Netzer et al., 1999) were used to identify exclusion criteria (Makley et al., 2008). Participants wore an actigraph watch, and sleep data were recorded 24 hours before the first light exposure, and then continuously throughout the remainder of the study. Actigraph data was to be analyzed on a pre-set epoch from 10:00 PM through 6:00 AM. ACG-generated variables included total sleep time, sleep efficiency, and number of awakenings. Baseline and post-treatment assessments included demographic questionnaire, Symbol Digit Modalities Test for attention, Positive and Negative Affective Schedule for mood, and the Karolinska Sleepiness Scale. Rate of recovery way measured with FIM™ efficiency. Nursing staff were surveyed quarterly to assess if light exposure was a feasible addition to daily care on the rehabilitation unit, and to rank the perceived nursing care burden of light exposure. Therapy staff were asked to evaluate subject participation/cooperation with therapy on a 0–100 scale at baseline and at the end of light exposure.

4.2.4 Progress

Recruitment and data collection is complete. Of 131 initial participants from three sites, 108 subjects had valid actigraphy data of at least five days in length. Some participants had a shorter length of stay on inpatient rehabilitation that did not allow for the planned 10 days of light exposure. Multivariate analyses were utilized to analyze between group differences for the BWLT and RLT exposure conditions. Light exposure was well tolerated and rated by nursing staff to be a feasible addition to daily care with low burden. Some research participants were interested in utilizing light exposure at home (after discharge from inpatient rehabilitation) and were provided with general information about BWLT and sleep from what is known in the general population. Analysis is being completed and the publication of results is under preparation.

4.3 Sleep apnea

4.3.1 Description

Sleep apnea is thought to occur in one-third of those with TBI (Holcomb et al., 2016; Webster et al., 2001). This number may be an underestimate given that sleep apnea may be largely undiagnosed and untreated in both the general and TBI populations. Sleep apnea is a sleep-related breathing disorder that is characterized by repeated cessation (i.e., apnea) or near cessation (i.e., hypopnea) of ventilation during sleep. Subtypes of sleep apnea include obstructive, central or mixed type. Obstructive sleep apnea (OSA) is diagnosed when there is repetitive collapse or obstruction of the airway resulting in oxygen desaturation (hypoxemia) and/or disturbance of sleep (arousals [<15 seconds] or awakenings [>15 seconds]) resulting in sleep fragmentation and non-restorative sleep. Collectively, chronic nightly hypoxemia and frequent arousals and awakenings (disturbing the continuity of sleep) due to cessation of breathing in sleep apnea reducing total sleep time may serve as mechanisms to hinder neurologic repair in acute stages (Zunzunegui et al., 2011) and explain earlier cognitive decline at younger ages in chronic TBI relative to persons with non-TBI (Xie et al., 2013).

Table 2
Circadian Disruption Variables

Content Measures

Screening Berlin Questionnaire

Delirium Rating Scale, Item 1

Demographic information Age

Sex

Injury Severity

Sleep Wake Actigraphy

•Sleep Time (minutes)

•Sleep Efficiency

•Number of Awakenings

Karolinska Sleepiness Scale

Barrow Neurological Institute Fatigue Scale

Cognition/Attention Symbol Digit Modalities Test

Mood/Affect Positive and Negative

Affective Schedule

Function/Recovery Functional Independence

Measure

Participation in Therapy

Content	Measures
Screening	Berlin Questionnaire
	Delirium Rating Scale, Item 1
Demographic information	Age
	Sex
	Injury Severity
Sleep Wake	Actigraphy
	•Sleep Time (minutes)
	•Sleep Efficiency
	•Number of Awakenings
	Karolinska Sleepiness Scale
	Barrow Neurological Institute Fatigue Scale
Cognition/Attention	Symbol Digit Modalities Test
Mood/Affect	Positive and Negative
	Affective Schedule
Function/Recovery	Functional Independence
	Measure
	Participation in Therapy

Recognition of sleep apnea at the earliest stage possible after TBI was a recommendation from the Galveston Brain Injury Conference focused on optimizing long-term health. Further, comparison of screening algorithms and tools to diagnose sleep apnea is considered a high priority future research need by the Agency for Healthcare Research and Quality (Balk et al., 2012). As such, investigators within special interest groups (SIGs) of the TBI Model Systems received funding the Patient Centered Outcomes Research Institute (PCORI) to advance sleep apnea research utilizing the TBI Model System infrastructure.

Comparison of Sleep Apnea Screening and Diagnostic Tools to Improve Rehabilitation Outcome (C-SAS) is a comparative effectiveness study conducted during inpatient rehabilitation at six TBIMS sites for those with moderate to severe TBI (PCORI, 2016). The study is a registered clinical trial with detailed information about it online (PCORI, 2018). The overall goal of the study is to promote earlier detection of sleep apnea by comparing screening and diagnostic tools. The study sites include James A. Haley Veterans Hospital and Polytrauma Rehabilitation Center, Tampa, FL; University Washington, Seattle, WA; Craig Hospital, Denver CO; Ohio State University, Columbus, OH; Moss Rehabilitation Research Institute, Philadelphia, PA; and Baylor Scott and White Rehabilitation, Dallas, TX. Collaborating sites include Stanford University, North Florida South Georgia Veterans Hospital, and University of Texas Southwestern. A key feature of the PCORI funding mechanism is engagement of stakeholders to promote pragmatic trials and adoption of study findings to promote knowledge translation. Stakeholders in the project include national directors of TBI care (Department of Veterans Affairs, Defense and Veterans Brain Injury Center), rehabilitation clinicians, hospital administrators, patients, and families. Stakeholders participate in regular meetings and provide feedback on study methodology, findings, and dissemination.

4.3.2 Aims

The first study aim compares the sensitivity and specificity of traditional self-report screening tools (i.e., Berlin, STOP-BANG) and objective actigraphy monitoring devices (2 medical grade actigraph devices) to the criterion standard for diagnosing sleep apnea. The criterion standard for sleep apnea includes polysomnography with EEG and sleep technologist monitoring the overnight study in a sleep laboratory (Level 1 polysomnography [PSG]). The second aim tests the non-inferiority of a portable polysomnography (Level 3) that has a smaller number of respiratory and oxygen indices and no EEG. For this study, all six sites hired sleep technologists and were equipped with identical Level 1 and 3 PSG technology. Participants wear both the Level 1 and Level 3 device for simultaneous overnight monitoring. Following the night of PSG, participating sites submit locally collected data to the TBI Model System and Statistical Center website including chart abstraction, medical record review, physical exam, and interview. All PSG and ACG data are submitted in a de-identified format to a FTP site for transfer to the centralized scoring center. A single centralized certified sleep technologist (RSPGT) scores all studies which are then interpreted by a board-certified sleep medicine physician. Polysomnography reports are generated and sent back to study sites within seven business days for clinical action if indicated.

4.3.3 Data elements

Demographic and injury characterization from the TBIMS data elements will be merged with the C-SAS generated data elements to characterize the sample. Information about the TBIMS variables are described on an online syllabus for detailed review. Similarly, data elements collected as part of this clinical trial can be found on an online syllabus (NDSC, 2018a; PCORI, 2018). A table describing the C-SAS study specific variables is summarized in Table 3. Public access to the database is summarized below.

Table 3
Sleep Apnea Variables

Pre-Injury Sleep Pre-Injury Sleep Apnea Diagnosis

History If so, Sleep Apnea Treatment History

Preinjury Insomnia History

If so, Insomnia Treatment History

History of Sleep Medication Usage

Medical Record Height

Review Weight

Amputation

PEG Tube Status on Day of PSG

Diet Status on Day of PSG

Medications

Manual Muscle Testing

Ashworth Spasticity Score

Physical Exam (Night of PSG) IV Status

Baseline O2 Saturation

Neck Circumference below thyroid

cartilage

Waist Circumference

Hip Circumference

Mallampati Classification

Agitated Behavior Scale Rating

Self-Report Sleep STOP-BANG

Apnea Screening BERLIN

Tools MAPI

Actigraphy Data Epoch Recording Interval (seconds)

During PSG Total Sleep Time (in minutes)

Sleep Latency (in minutes)

Wake After Sleep Onset

Sleep Efficiency

Number of Awakenings

Average Activity Per Minute

Average White Light (Illuminance

[Lux])

Time spent above 50 lux in minutes

Polysomnography Lights Off

(PSG) Lights On

PSG AHI Indices OSA Index (Obstructive AHI);

#/hour

REM Obstructive AHI

Non-REM Obstructive AHI

Supine Obstructive AHI

Non-Supine Obstructive AHI

Central Sleep Apnea Index (Centeral AHI);

#/hour

Cheyne Stokes Respiration

Overall AHI

PSG Oxygen Indices SpO2 NADIR

Desaturation Index 4%

Desaturation Index 3%

SpO2 Less than 90% Saturation

Total Sleep Time Less than 90%

(minutes and percentage)

Total Sleep Time Less than 85%

(minutes and percentage)

PSG Sleep Indices Time in Bed

Total Sleep Time (minutes)

Time Asleep Supine and Non-Supine

(minutes)

Sleep Efficiency

Wake After Sleep Onset (WASO)

WASO as Total % of Sleep

Clinical Sleep Latency

REM Latency

% Time in Stages of Sleep (N1, N2,

N3, REM)

Arousal Index

Diagnoses OSA, CSA, Mixed

PSG Scoring 4-Level Scoring Challenge Rating

Difficulty Laterality of Distortion of EEG for

Scoring

Presence of theta and high frequency

activity distorting EEG

ACG – Actigraphy, AHI-Apnea Hypopnea Index, CSA-Central Sleep Apnea, O2 – Oxygen, OSA-Obstructive Sleep Apnea, PSG-Polysomnography, Mixed – Mixed Sleep Apnea, WASO- Wake After Sleep Onset; #/hour – number per hour.

4.3.4 Progress

The initial cohort of study participants were exclusively TBIMS participants. However, delays in obtaining TBIMS consent resulted in short-time intervals to obtain consent for C-SAS during the remaining acute inpatient length of stay. Approximately 12 months into the study, the requirement for TBIMS inclusion was eliminated. Comparable TBIMS data elements will be collected on non-TBIMS participants. Data collection is scheduled to end in November of 2018.

5 TBIMS capacity building possibilities

5.1 Banking of data for internal and external investigator access

Datasets described in this publication are archived with the TBI Model System National Data and Statistical Center. Standard Operating Procedures (SOP) have been developed for both TBIMS internal investigators and non-TBIMS external investigators to use the data once study aims have been accomplished. The Insomnia Module study is currently archived and available for unrestricted analyses. Once publications addressing primary study aims for the remaining studies have been generated, they will be available for request and further analysis (NDSC, 2018b).

5.2 Ongoing collaborations

The TBIMS infrastructure has successfully yielded collaborations receiving additional funding from the Department of Defense, Centers for Disease Control, National Institute of Health, and recently two large Patient Centered Outcomes Research Institute Awards with one including a focus on sleep after TBI. The 30-year history of the TBIMS has resulted in over 850 peer-reviewed publications generated from TBIMS-related data and are logged in an online registry (MSKTC, 2018). At the time of this writing, twenty-three were focused on sleep and TBI. Collaboration with the NIDILRR-funded Model Systems Knowledge Translation Center has helped to yield sleep-related educational products including info-comics, slide-shows, and fact-sheets targeting patients and families. Systematic reviews and special issues focusing on sleep and TBI in scientific journals led by TBIMS investigators targeting researchers. Finally, a recent special issue in a clinician magazine for brain injury professionals addressed the significance of addressing sleep after TBI (Nakase-Richardson, 2018). Articles addressing sleep apnea (Nakase-Richardson & Schwartz, 2018), insomnia (Dahdah et al., 2018), circadian rhythm disorders (Zeitzer et al., 2018), sleep hygiene on rehabilitation units (Monden et al., 2018), and healthy sleep initiatives (Nallu, 2018) were written by TBIMS investigators and subject matter experts in the sleep medicine and sleep research fields. The TBIMS collaborative infrastructure is uniquely positioned to facilitate TBI and sleep research and disseminate findings to key stakeholders.

6 Future directions in advancing sleep research and TBI

Future directions for the SIG include continued development of available knowledge and understanding of the multidimensional factors that contribute to TBI-related sleep disturbance, optimal assessment tools, effectiveness of available treatments, and treatment compliance in this population. We aim to accomplish this through the evaluation of the rich data that has been collected through TBIMS to date and by developing and obtaining funding for new research guided by our current work and planned analyses of existing data. We anticipate that this will contribute to the empirical knowledge needed for our group to improve the precision and efficacy of existing diagnostic measures and treatments and develop new measures and treatments for individuals affected by TBI-related sleep disturbance. This will also contribute to future dissemination of information and products that will bring useful and accessible tools to consumers, clinicians, researchers, and other stakeholders.

Conflict of interest

The Polytrauma Rehabilitation Center Traumatic Brain Injury (TBI) Model System collaboration is an Interagency Collaboration between the Department of Veterans Affairs and the Department of Health and Human Services (National Institute on Disability, Independent Living, and Rehabilitation Research). This research is sponsored by VHA Central Office VA TBI Model System Program of Research, Subcontract from General Dynamics Health Solutions (W91YTZ-13-C-0015, PI: Nakase-Richardson) from the Defense and Veterans Brain Injury Center within the Defense Health Agency, contract with the Patient Centered Outcomes Research Institute (PCORI Project CER-1511-33005, PI: Nakase-Richardson), U.S. Department of Veterans Affairs Health Services Research and Development COIN grant (1 I50 HX001233-01; CINDRR).

The views, opinions, statements, and/or findings contained in this article are those of the authors and should not be construed as an official Department of Defense, Department of Veterans Affairs, National Institute on Disability, Independent Living, and Rehabilitation Research, Patient-Centered Outcomes Research Institute (PCORI), not its Board of Governors or Methodology Committee position or any other federal agency, policy, or decision unless so designated by other official documentation.

Footnotes

Acknowledgments

The participating agencies’ institutional review boards approved the studies described and informed consent was obtained after the details of the study were thoroughly explained to participants.

References

Balk, E. M., Chung, M., Moorthy, D., Chan, J. A., Patel, K., Concannon, T. W., … Chang, L. K. W. (2012). In Future Research Needs for Diagnosis of Obstructive Sleep Apnea: Identification of Future Research Needs From Comparative Effectiveness Review No. 32. In Rockville (MD).

Cantor, J. B., Bushnik, T., Cicerone, K., Dijkers, M. P., Gordon, W., Hammond, F. M., & Spielman, L. A. (2012). Insomnia, fatigue, and sleepiness in the first 2 years after traumatic brain injury: An NIDRR TBI model system module study. Journal of Head Trauma Rehabililtation, 27(6), E1–14.

Dahdah, M. N., Bell, K., Lequerica, A. H., & Merfeld, A. (2018). Effectiveness of cognitive behavioral therapy for treating insomnia in healthy individuals and those with brain injury. Brain Injury Professional, 14(4), 12–14. https://issuu.com/braininjuryprofessional/docs/bip february 2018?e=1121786/58553316. Accessed online May 8, 2018

Dash, M. B., Douglas, C. L., Vyazovskiy, V. V., Cirelli, C., & Tononi, G. (2009). Long-term homeostasis of extracellular glutamate in the rat cerebral cortex across sleep and waking states. Journal of Neuroscience, 29(3), 620–629.

Dijkers, M. P., Harrison-Felix, C., & Marwitz, J. H. (2010). The traumatic brain injury model systems: History and contributions to clinical service and research. Journal of Head Trauma Rehabilitation, 25(2), 81–91.

Holcomb, E. M., Schwartz, D. J., McCarthy, M., Thomas, B., Barnett, S. D., & Nakase-Richardson, R. (2016). Incidence, characterization, and predictors of sleep apnea in consecutive brain injury rehabilitation admissions. Journal of Head Trauma Rehabilitation, 31(2), 82–100.

Lequerica, A., Bushnik, T., Wright, J., Kolakowsky-Hayner, S. A., Hammond, F. M., Dijkers, M. P., & Cantor, J. (2012). Psychometric properties of the multidimensional assessment of fatigue scale in traumatic brain injury: An NIDRR traumatic brain injury model systems study. Journal of Head Trauma Rehabilitation, 27(6), E28–35.

Lequerica, A., Chiaravalloti, N., Cantor, J., Dijkers, M., Wright, J., Kolakowsky-Hayner, S. A., & Bell, K. (2014). The factor structure of the Pittsburgh Sleep Quality Index in persons with traumatic brain injury. A NIDRR TBI model systems module study. NeuroRehabilitation, 35(3), 485–492.

Lequerica, A. H., Botticello, A. L., Lengenfelder, J., Chiaravalloti, N., Bushnik, T., Dijkers, M. P., & Rosenthal, J. (2017). Factors associated with remission of post-traumatic brain injury fatigue in the years following traumatic brain injury (TBI): A TBI model systems module study. Neuropsychological Rehabilitation, 27(7), 1019–1030.

10.

Longordo, F., Kopp, C., & Luthi, A. (2009). Consequences of sleep deprivation on neurotransmitter receptor expression and function. European Journal of Neuroscience, 29(9), 1810–1819.

11.

Makley, M. J., English, J. B., Drubach, D. A., Kreuz, A. J., Celnik, P. A., & Tarwater, P. M. (2008). Prevalence of sleep disturbance in closed head injury patients in a rehabilitation unit. Neurorehabilitation and Neural Repair, 22(4), 341–347.

12.

Mathias, J. L., & Alvaro, P. K. (2012). Prevalence of sleep disturbances, disorders, and problems following traumatic brain injury: A meta-analysis. Sleep Medicine, 13(7), 898–905.

13.

McDermott, C. M., LaHoste, G. J., Chen, C., Musto, A., Bazan, N. G., & Magee, J. C. (2003). Sleep deprivation causes behavioral, synaptic, and membrane excitability alterations in hippocampal neurons. Journal of Neuroscience, 23(29), 9687–9695.

14.

Monden, K. R., Gerber, D., Newman, J., Phillipus, A., Biggs, J., Schneider, H., & Makley, M. (2018). Sleep hygiene: A novel, nonpharmacological approach to treating sleep-wake cycle disturbance after moderate to severe brain injury on an inpatient rehabilitation unit. Brain Injury Professional, 14, 20–21. https://issuu.com/braininjuryprofessional/docs/bip_february_2018?e=1121786/58553316. Accessed online May 8, 2018.

15.

MSKTC. (2018). Model System Knowledge Translation Center. Retrieved from www.msktc.org/publications?sys=T. Accessed online May 9, 2018.

16.

Nakase-Richardson, R. (2018). Improving the recognition and treatment of sleep disorders in neurorehabilitation. Brain Injury Professional, 14, 7.

17.

Nakase-Richardson, R., & Schwartz, D. J. (2018). Sleep apnea and traumatic brain injury. Brain Injury Professional, 14, 8–10. https://issuu.com/braininjuryprofessional/docs/bip_february_2018?e=1121786/58553316. Accessed online May 8, 2018.

18.

Nallu, S. (2018). Recommended sleep duration for adults and children: An American Academy of Sleep Medicine and Sleep Research Society Consensus Statement. Brain Injury Professional, 14, 22. https://issuu.com/braininjuryprofessional/docs/bip_february_2018?e=1121786/58553316. Accessed online May 8, 2018.

19.

Netzer, N. C., Stoohs, R. A., Netzer, C. M., Clark, K., & Strohl, K. P. (1999). Using the Berlin Questionnaire to identify patients at risk for the sleep apnea syndrome. Annals of Internal Medicine, 131(7), 485–491.

20.

PCORI. (2016). Comparison of sleep apnea assessment strategies to maximize TBI rehabilitation participation and outcome. PCORI R-1511-33005. Retrieved from http://www.pcori.org/research-results/2016/comparison-sleep-apneaassessment-strategies-maximize-tbi-rehabilitation. Accessed online May 9, 2018.

21.

PCORI. (2018). Comparison of sleep apnea assessment strategies to maximize TBI rehabilitation participation and outcome clinical trial. Retrieved from https://clinicaltrials.gov/ct2/show/NCT03033901. Accessed online May 9, 2018.

22.

Rutherford, G. W., & Corrigan, J. D. (2009). Long-term consequences of traumatic brain injury. Journal of Head Trauma Rehabilitation, 24(6), 421–423.

23.

TBIMS NDSC. (2018a). Comparison of Sleep Apnea Assessment Strategies to Maximize TBI Rehabilitation Participation and Outcome Study Website. Retrieved from apnea.tbindsc.org. Accessed online May 9, 2018.

24.

TBIMS NDSC. (2018b). TBI Model Systems National Data and Statistical Center. Retrieved from https://www.tbindsc.org/. Accessed online May 9, 2018.

25.

Trzepacz, P. T., Mittal, D., Torres, R., Kanary, K., Norton, J., & Jimerson, N. (2001). Validation of the Delirium Rating Scale-revised-98: Comparison with the delirium rating scale and the cognitive test for delirium. Journal of Neuropsychiatry and Clinical Neuroscience, 13(2), 229–242.

26.

Walker, M. P. (2009). The role of sleep in cognition and emotion. Annals of N Y Academy of Sciences, 1156, 168–197.

27.

Webster, J. B., Bell, K. R., Hussey, J. D., Natale, T. K., & Lakshminarayan, S. (2001). Sleep apnea in adults with traumatic brain injury: A preliminary investigation. Archives of Physical Medicine and Rehabilitation, 82(3), 316–321.

28.

Xie, L., Kang, H., Xu, Q., Chen, M. J., Liao, Y., Thiyagarajan, M., & Nedergaard, M. (2013). Sleep drives metabolite clearance from the adult brain. Science, 342(6156), 373–377.

29.

Zeitzer, J. M., Watson, E. M., & Holcomb, E. (2018). Circadian Rhythms: Disruptions and Treatments in Acute Traumatic Brain Injury. Brain Injury Professional, 14, 16–19. https://issuu.com/braininjuryprofessional/docs/bip_february_2018?e=1121786/58553316. Accessed online May 8, 2018.

30.

Zunzunegui, C., Gao, B., Cam, E., Hodor, A., & Bassetti, C. L. (2011). Sleep disturbance impairs stroke recovery in the rat. Sleep, 34(9), 1261–1269.

Pre-Injury Sleep	Pre-Injury Sleep Apnea Diagnosis
History	If so, Sleep Apnea Treatment History
	Preinjury Insomnia History
	If so, Insomnia Treatment History
	History of Sleep Medication Usage
Medical Record	Height
Review	Weight
	Amputation
	PEG Tube Status on Day of PSG
	Diet Status on Day of PSG
	Medications
	Manual Muscle Testing
	Ashworth Spasticity Score
Physical Exam (Night of PSG)	IV Status
	Baseline O2 Saturation
	Neck Circumference below thyroid
	cartilage
	Waist Circumference
	Hip Circumference
	Mallampati Classification
	Agitated Behavior Scale Rating
Self-Report Sleep	STOP-BANG
Apnea Screening	BERLIN
Tools	MAPI
Actigraphy Data	Epoch Recording Interval (seconds)
During PSG	Total Sleep Time (in minutes)
	Sleep Latency (in minutes)
	Wake After Sleep Onset
	Sleep Efficiency
	Number of Awakenings
	Average Activity Per Minute
	Average White Light (Illuminance
	[Lux])
	Time spent above 50 lux in minutes
Polysomnography	Lights Off
(PSG)	Lights On
PSG AHI Indices	OSA Index (Obstructive AHI);
	#/hour
	REM Obstructive AHI
	Non-REM Obstructive AHI
	Supine Obstructive AHI
	Non-Supine Obstructive AHI
	Central Sleep Apnea Index (Centeral AHI);
	#/hour
	Cheyne Stokes Respiration
	Overall AHI
PSG Oxygen Indices	SpO2 NADIR
	Desaturation Index 4%
	Desaturation Index 3%
	SpO2 Less than 90% Saturation
	Total Sleep Time Less than 90%
	(minutes and percentage)
	Total Sleep Time Less than 85%
	(minutes and percentage)
PSG Sleep Indices	Time in Bed
	Total Sleep Time (minutes)
	Time Asleep Supine and Non-Supine
	(minutes)
	Sleep Efficiency
	Wake After Sleep Onset (WASO)
	WASO as Total % of Sleep
	Clinical Sleep Latency
	REM Latency
	% Time in Stages of Sleep (N1, N2,
	N3, REM)
	Arousal Index
Diagnoses	OSA, CSA, Mixed
PSG Scoring	4-Level Scoring Challenge Rating
Difficulty	Laterality of Distortion of EEG for
	Scoring
	Presence of theta and high frequency
	activity distorting EEG