Traumatic brain injury (TBI), particularly moderate to severe TBI, is a major cause of physical and neuropsychiatric disability in Veterans, preventing community reintegration and return to employment. Early prognostication of outcome from TBI is a critical need which is currently lacking, and is necessary to optimally allocate scarce resources from existing VA treatment programs, and to better inform patients and their families about the prognosis. Current predictors of long-term functional outcome after TBI are generally based on demographic/socioeconomic and clinical markers, and have shown only moderate predictive ability; they have not been sufficiently precise to direct therapy decisions in individuals. To address the critical need for sensitive/specific functional outcome predictors after TBI, we propose to test novel predictors of functional outcome after TBI, which are based on quantitative analysis of electroencephalography (EEG) during sleep. The structure of the analysis is based on measures of cross- frequency couplings (CFC) between EEG frequency bands, reflecting coordination between neural circuits that generate the underlying rhythmic and oscillatory pattern of EEG. In our preliminary animal studies, we have identified several EEG neuromarkers that were highly sensitive to the TBI group. Next, we examined these same neuromarkers and their correlation with functional outcomes (Functional Independence Measure (FIM) and Disability Rating Scale (DRS)) in a small cohort (n=7) of Veterans with moderate-to-severe TBI who received neuro-rehabilitation. We found that one particular sleep-based EEG neuromarker, delta-gamma cross-frequency coupling, very strongly and significantly predicted functional improvement after rehabilitation (DRS regression model: R2=0.95, F=86, p < 0.0002; FIM regression model: R2 =0.91, F= 50, p < 0.001). Importantly, we found the same strong predictive capability at a follow-up time point one-year later, not only indicating robust internal consistency, but also highlighting the potential to identify a valuable, ultra-long term predictor f outcome, (FIM regression model: R2 =0.89, F= 40, p < 0.002) . The main objective of this SPiRE project is to comprehensively evaluate the ability of EEG CFC-based neuromarker to distinguish TBI from healthy control, and to predict Functional Outcomes of Veterans with moderate-severe TBI. This will be accomplished by computing the delta-gamma neuromarkers using an existing database of recorded sleep-EEG studies from 45 healthy individuals, and 80 Veterans with moderate- severe TBI, in conjunction with their measures of functional outcome (DRS and FIM) at baseline, at discharge from neuro-rehabilitation, and at one year post discharge. We will also explore the development of potentially improved neuromarkers by generalizing our analyses to include computation of EEG cross frequency couplings between an expanded pairs of frequency bands including theta, alpha, delta and gamma, and for 3 states of awake, non-REM, and REM sleep. Expected Outcomes: This study will evaluate novel objective neuromarkers that can predict functional recovery and response to treatment in moderate-severe TBI patients.

Public Health Relevance

Traumatic brain injury (TBI), particularly moderate to severe TBI, is a major cause of physical and neuropsychiatric disability in Veterans, preventing community reintegration and return to employment. Early predictors of functional recovery after TBI is a critical need as it can be utilized to direct appropriate rehabilitation treatment, optimize utilization of resources, and infrm patients and relatives about the prognosis. This study will evaluate a new method of predicting of long-term functional outcome after TBI based on analysis of brain function during awake and sleep periods.

Agency
National Institute of Health (NIH)
Institute
Veterans Affairs (VA)
Type
Veterans Administration (I21)
Project #
5I21RX001923-02
Application #
9315586
Study Section
Rehabilitation Research and Development SPiRE Program (RRDS)
Project Start
2016-08-01
Project End
2018-07-31
Budget Start
2017-08-01
Budget End
2018-07-31
Support Year
2
Fiscal Year
2018
Total Cost
Indirect Cost
Name
Veterans Health Administration
Department
Type
DUNS #
097378632
City
Gainesville
State
FL
Country
United States
Zip Code
32608