117 / 198
the corpus callosum serves as physical restraint for increase in hematoma
size. We recommend that conservative and clinical management of
ncSDH, be similar to tSAH. The more common cSDH should be dis-
tinctly categorized from the ncSDH.
Keywords: Acute traumatic subdural hematoma, Non-Convexity
Subdural Hematoma, Tentorial and Hemispheric Subdural Hematoma,
Convexity Subdural hematoma
C3-03
EARLY NUTRITIONAL THERAPY AND MINIMAL BLOOD
GLUCOSE VARIABILITY IMPROVE OUTCOMES AFTER
TBI
Stephanie Wolahan
, David McArthur, Paul Vespa, Neil Martin,
Thomas Glenn
UCLA Brain Injury Research Center, UCLA Neurosurgery, Los An-
geles, USA
Following severe TBI, dysfunctional cerebral metabolism is associated
with metabolic crisis, the coincidence of high cerebral LPR and low
glucose, and with poor recovery. Meanwhile, glycemic control in the
acute injury period to avoid severe hyperglycemia improves outcomes.
The objective of this study was to investigate the impact of nutritive
support provided in the neuroICU between post-injury days (PID) 0 and
10 on clinical outcomes, acute glycemic control, and metabolic crisis.
Nutritive support (enteral and parenteral routes), point-of-care (POC)
glucose, and cerebral microdialysis glucose and LPR were extracted
from the UCLA BIRC electronic database and patient charts for twenty-
six severely injured consented adults. The variability of clinical data was
quantified by the standard deviation of POC glucose readings (recorded
approximately every 4 hours), the rate of carbohydrate (CHO) caloric
delivery, and insulin provided (bolus and IV) between PID 0-10. Clinical
outcomes were Glasgow Outcome Scale extended (GOSe) scores ob-
tained 6 months post-injury. There was a significant correlation between
GOSe and kilocalories delivered PID 0 (0.432 [0.054 0.702], p
=
0.028)
but not between GOSe and average daily kilocalories PID 0 through 10 (
-
0.183 [
-
0.532 0.220], p
=
0.371). Additionally, there was a significant
negative correlation between GOSe and glucose variability (
-
0.414 [
-
0.691
-
0.032], p
=
0.035). Linear modeling significantly predicted GOSe
by POC glucose variability (p
=
0.011), by CHO caloric delivery rate
variability (p
=
0.013), and by total kilocalories received in the first 24
hours following injury (p
=
0.002). Although the incidence of metabolic
crisis was not a significant predictor of GOSe in this cohort, blood-brain
glucose correlations PID 0-10 were negatively correlated to incidence of
metabolic crisis (
-
0.406 [
-
0.690
-
0.013], p
=
0.044). Early nutritional
therapy, blood glucose variability and, to a lesser extent, CHO caloric
delivery rate variability were predictive of 6 month outcomes. While
delivery rate variability was not a significant indicator, the impact of
nutritional support, including makeup, timing, and rate of delivery, on
glycemic control and variability warrants further research.
Keywords: Glycemia, Microdialysis, 6-month outcome, Neurocri-
tical Care, Nutrition
C3-04
METABOLOMICS OF HUMAN TRAUMATIC BRAIN INJURY
Daniel Hirt
1
,
Stephanie Wolahan
1
, Alyson Thien
1
, Joshua Dusick
1
,
Daniel Braas
2
, Neil Martin
1
, Thomas Glenn
1
1
UCLA Brain Injury Research Center, UCLA Neurosurgery, Los
Angeles, USA
2
UCLA Crump Institute of Molecular Imaging, Molecular and Med-
ical Pharmacology, Los Angeles, USA
Understanding metabolomics is paramount in the individualized
treatment and prognosis of severe traumatic brain injury (TBI).
Analysis of the metabolome has gained significant momentum in this
endeavor in numerous disease processes, but has not yet received the
attention it deserves in TBI research. As a result, we conducted a
preliminary observational and feasibility study aimed at identifying
differences in the metabolite composition of TBI patients and normal
individuals. Eight subjects, 6 TBI patients (mean age: 31, mean GCS:
3.8) and 2 normal individuals (mean age: 22) were consented and
enrolled. Blood samples were collected from all 8 subjects and ana-
lyzed using LCMS. We identified a total of 103 metabolites. A
principle components analysis revealed a significant separation of the
metabolite contributions of TBI versus normal patients in PC2
(p
<
0.0001). Further investigation revealed that elevations of metab-
olites of glycolysis and the TCA cycle, such as glycine, lactate, and a-
ketogluterate, constituted the unique metabolome of a brain-injured
patient. In contrast, citrate, alanine, and phosphoglycerate of these
same pathways were elevated in normal individuals. Other differences
that were identified between these two groups, included changes in
numerous amino acids, neurotransmitters, and metabolites involved in
DNA repair. Investigations of the metabolome are of utmost impor-
tance in the understanding of post-traumatic metabolism, identifica-
tion of novel biomarkers, and ultimately improved clinical
management of these critically ill patients. While this study is limited
by the small number of subjects, it nonetheless demonstrates the
feasibility of this methodology and gives a small glimpse into the
possibility that this type analysis can offer in directing individualized
patient care.
Keywords: Metabolomics, Principal components, Individualized
medicine, Energy metabolism
C3-05
USE OF ASPIRIN AND P2Y12 RESPONSE ASSAYS IN DE-
TECTING REVERSAL OF PLATELET DYSFUNCTION
CAUSED BY ANTIPLATELET AGENTS IN TBI
Phillip Choi
, Phillip Parry, Joshua Bauer, Benjamin Zusman, David
Panczykowski, Ava Puccio, David Okonkwo
University of Pittsburgh School of Medicine, Neurological Surgery,
Pittsburgh, USA
Many adult patients suffering traumatic brain injury have pre-
existing comorbidities requiring antiplatelet therapy. These agents
potentially worsen outcome, which has resulted in a treatment ap-
proach of platelet transfusion in patients with evidence of intracra-
nial bleeding and history of antiplatelet therapy. We seek to
understand the utility of the aspirin and P2Y12 response (ARU and
PRU, respectively) assays as tools for detecting platelet dysfunction
and subsequent reversal following transfusion of platelets in trau-
matic brain injury patients. Between 2010 and 2015, we conducted a
prospective comparative cohort study of patients presenting with
a positive head CT and a reported history of antiplatelet ther-
apy. Platelet dysfunction was assessed with the VerifyNow assay
(Accumetrics, San Diego, CA) upon admission and 6 hours post-
transfusion, with a primary endpoint of disinhibition after platelet
transfusion. 107 patients were available for analysis [mean age
75.5
–
12.2 years, 65% male, median GCS 15 (IQR 13–15) and
Marshall score 2 (IQR 2–4)], with 62% taking aspirin, 8% taking
clopidogrel, and 30% taking both agents. According to the ARU and
A-81