66 / 162
Mild traumatic brain injury (mTBI) is a rising epidemic affecting
millions of people each year. Our understanding of mTBI is still in its
infancy and to gain a greater understanding relevant animal models
should replicate many of the features seen in human mTBI. These
include changes to diffusion tensor imaging (DTI) parameters, the
absence of anatomical lesions on conventional neuroimaging, and
changes to neurobehavioral outcomes. These changes are transient in
majority of the mTBI victims. The Maryland Model of TBI causes the
anterior-posterior plus sagittal rotational acceleration of the brain
frequently observed with motor vehicle and sports related TBI injuries
without skull fracture (Kilbourne et al., 2009). The goal of our study
was to characterize longitudinal pathophysiological changes following
a single mTBI using magnetic resonance imaging (MRI), behavioral
assays, and histology. On DTI we observed a significant difference in
fractional anisotropy (FA) and longitudinal and radial diffusivities in
the internal capsule 72 hours after injury compared to baseline mea-
sures (n
=
11). A significant difference in longitudinal diffusivity was
also observed in the genu of the corpus callosum also at 72 hours
compared to baseline measures. The exploratory activity computer-
ized activity box showed significant decrease in the ambulatory dis-
tance, average velocity, stereotypic counts, and vertical counts
compared to baseline measures at 72 hours. Histological examination
of the mTBI brain sections indicated a significant decrease in the
expression of myelin basic protein in the internal capsule. A signifi-
cant increase in the number of apoptotic cells was observed by cas-
pase-3 labeling in these brains as well as compromise of the blood
brain barrier by immunoglobulin-G detection. The changes with DTI
and neurobehavioral outcomes were only observed during the acute
phase of injury, similar to what was observed in human mTBI (Nar-
ayana et al. 2015). Thus this experimental TBI replicates the obser-
vations in human mTBI and can be used to investigate the long-term
effects of mTBI.
Keywords: fractional anisotropy, longitudinal and radial diffusiv-
ities, neurobehavioral
B6-22
ACUTE CHANGES IN FDG PET AFTER SINGLE AND RE-
PEAT MTBI IN RATS CORRELATE WITH CLINICALLY
RELEVANT SYMPTOMS OF CONCUSSION
Casandra Cartagena
1
, Scott Jones
2
, Andrea Mountney
1
, Deborah
Shear
1
, Stephanie Braverman
1
, Colin Wilson
2
, Shalini Jaiswal
2
, Frank
Tortella
1
, Reed Selwyn
2
1
Walter Reed Army Institute of Research, Brain trauma neuropro-
tection and neurorestoration, Silver Spring, USA
2
Uniformed Services University Health Sciences, Translational Ima-
ging Facility, Bethesda, USA
Mild traumatic brain injury (mTBI) can lead to immediate symp-
toms (headache, dizziness, confusion, loss of consciousness) but
lack gross pathology on computerized tomography (CT). Most
patients recover within 10 days but a minority have continued
deficits. Change in 18F-fluorodeoxyglucose (FDG) positron emis-
sion topography (PET) has been suggested as a potential prognostic
indicator of this minority. Here we evaluated FDG and CT 24 hr
post-injury in a rat concussion model using single and repeat
projectile impacts (sPCI, rPCI) to the right cortex. Unlike other
mTBI injury models this model lacks any gross pathology/injury.
Post-injury, animals were evaluated for righting reflex (RR; im-
mediately) and gait analysis (2 hr). Injured animals were compared
to their respective single (sSham) or repeat (rSham) sham controls
(equivalent anesthesia, no impact). All rats were negative for CT
findings including skull fracture. Following sPCI, RR time was
increased and gait analysis showed decreased swing speed com-
pared to sSham; FDG levels were significantly increased in right
olfactory bulb (OB). rPCI also showed increased RR time and
decreased limb swing speed compared to rSham; OB FDG levels
were increased. In addition, following rPCI FDG levels were in-
creased in right cortex and decreased in left hypothalamus. In-
creases in energy utilization may indicate cells in these regions
have initiated recovery processes rather than committing to cell
death. Increased FDG levels in OB correlated with increased RR
and decreased swing speed, suggesting this area may be an acute
marker of injury severity in the absence of gross lesions or positive
CT. Although not well studied, damage to the OB and smell
dysfunction have been reported following mTBI. Given the prox-
imity of the OB to the skull, this area may be particularly sensitive
to coup counter coup injury following concussion. Ongoing studies
will determine whether FDG-PET alterations continue, or have
prognostic value, for long-term deficits.
Keywords: Concussion, mTBI, PET, FDG
B6-23
SOFTWARE TOOL FOR LOADING TRAUMATIC BRAIN
INJURY NEUROIMAGE DATA INTO AN EXTERNAL RE-
POSITORY
Rich Hammett,
Justin Senseney
, Terry Oakes, Gerard Riedy
Walter Reed National Military Medical Center, National Intrepid
Center of Excellence, Bethesda, USA
Objective:
Our overall objective is to load neuroimage data into an
external data repository from a military population at the National
Intrepid Center of Excellence (NICoE). NICoE has image data from
a large number of chronic TBI subjects (N
=
800). The Federal In-
teragency Traumatic Brain Injury Repository (FITBIR) is a database
for sharing research data on TBI subjects and relating subjects
across studies while protecting their privacy. The objective of this
work is to create a software tool to facilitate transfer of data into the
FITBIR repository, in order to increase accuracy and reduce operator
time.
Methods:
We created a Python software tool, FITBIR Import of
Neuroimage Data (FIND), which takes a list of coded subject iden-
tification numbers, finds selected anonymized image data from our
archive/processing pipeline, subject image database and extracts in-
dexing metadata from these files and other coordinated databases.
FIND builds an upload package of the image files and associated
metadata for each subject and stores them in a FITBIR submission
package. This upload package can be pushed through the FITBIR
image submission preparation program which verifies our extracted
metadata with the metadata stored in each image file, and generates a
thumbnail image for FITBIR. The package is then validated and up-
loaded by the FITBIR submission tool.
Results:
FIND successfully imported 40 full MRI datasets for 6
subjects, and made this task much simpler than available methods. It
is being expanded to include PET data, to generate a thumbnail image
for each image type, and to extract other useful metadata from DI-
COM files to give more control than allowed by the FITBIR image
processing tool.
Conclusion:
We have created a tool to automate and simplify the
importation of large numbers of image files into FITBIR. We are
making it available to other users via FITBIR. We have made
FIND available to others through GitHub, a web-based software
repository.
Keywords: Software, Database, Sharing, FITBIR, military, blast
A-66