Journal of NeurotraumaVol. 36, No. 13 AbstractsFree AccessAbstracts from The 37th Annual National Neurotrauma Symposium June 29–July 3, 2019 Pittsburgh, PennsylvaniaPublished Online:27 Jun 2019https://doi.org/10.1089/neu.2019.29100.abstractsAboutSectionsPDF/EPUB Permissions & CitationsPermissionsDownload CitationsTrack CitationsAdd to favorites Back To Publication ShareShare onFacebookTwitterLinked InRedditEmail Selected Oral PresentationsDBA Data Blitz A: Mini PresentationsDBA‐01 SIGNIFICANT EPIDEMIOLOGICAL CHANGES IN CERVICAL SPINAL CORD INJURIES IN THE STATE OF MARYLAND OVER AN 18‐YEAR PERIODBizhan Aarabi, Jennifer Albrecht, Kathirkamanathan Shanmuganathan, Benjamin Howie, Pallavi AtluriUniversity of Maryland School of Medicine, Neurosurgery, Baltimore, USACross‐sectional analysis of longitudinal data from Spinal Cord Injury Model Systems during the past 40 years analyzing over 30,000 traumatic spinal cord injuries (SCIs) have indicated that patients with SCI are getting older and the proportion of falls and incomplete SCI is increasing. The demographics, injury mechanism, AIS grade, ASIA motor score, intramedullary lesion length (IMLL), and hematomyelia on the admission MRI of 1251 patients who were admitted to the Shock Trauma Center with cervical SCI were analyzed for time trend change over an 18‐year period. During this period the incidence of cervical SCI in the state of Maryland did not change significantly. In this study, 78.7 percent of patients were male. Injury mechanism in 47.3 percent was due to falls, 33.6 percent due to motor vehicle collisions (MVCs), and 11.6 percent sport injuries. Miscellaneous causes were noted in 7.5 percent of patients. Close to 30 percent of injuries were complete (AIS grade A), and mean ASIA motor score was 45.3. Mean IMLL was 36.7 mm, and 35.8 percent of patients had evidence of hematomyelia on the admission MRI. From 2001 to 2018, the proportion of cervical SCIs caused by MVCs and the proportion of complete injuries decreased notably while age and motor score increased. These observed changes were not due to a change in referral patterns to University of Maryland, which remained steady during the study period (p for trend = 0.8). Mean IMLL and proportion of injuries with hematomyelia did not change significantly during this period. To determine whether the observed changes were independent of one another, we ran a series of regression models. For each factor that significantly increased or decreased during the study period, we tested the time trend, adjusting for all other factors. Time trends for age, MVC, and complete injury remained significant and in the same direction following adjustment for other factors. The time trend for motor score was no longer significant following adjustment.Keywords: Spinal Cord Injury, Epidemiology, Time Trend, Cross Sectional Study, Longitudinal Study, Intramedullary Lesion LengthDBA‐02 LONG TERM PROGNOSIS OF SEVERE TRAUMATIC BRAIN INJURED PATIENTS USING ACUTE PHASE NEUROPHYSIOLOGY TESTINGJoshua Bauer, Ava Puccio, Benjamin Zusman, David Panczykowski, Jeffrey Balzer, David OkonkwoUniversity of Pittsburgh School of Medicine, Neurosurgery, Pittsburgh, USAIntroduction: Current prognostic measures for severe traumatic brain injury (TBI) are initial Glasgow Coma Scale (GCS) score, pupillary response, and computerized tomography results; however, their predictive value is limited by confounding variables contributing to outcome and poor inter‐rater reliability. Use of somatosensory evoked potentials (SEPs) testing in the acute phase may provide a reliable objective mechanism for predicting patient prognosis.Methods: Under an IRB‐approved protocol, prospective clinical and outcome data were collected from all severe TBI patients (Glasgow Coma Score ≤8) admitted to UPMC between 2006‐2011. SEP recording was performed on post‐trauma day 5. N9, N13, P14, N20, and P30 latencies are measured; central conduction time (CCT) is defined as the delay between the P14 peak recorded at the C‐2 electrode and the N20 peak recorded at the somatosensory cortex. Peak‐to‐peak amplitudes measured are of N20/P30 using Fz reference. Criterion for pathologic SEP event was defined as either the reduction of the N20 amplitude below 50% of baseline or the increase of latency between P14 and N20 by 10%. SEPs were graded as bilaterally absent, unilateral absence, bilateral delay >23ms, unilateral delay with normal, and bilateral normal latency. Functional outcome was prospectively collected at 12‐month follow‐up, and included neurologic status and mortality as assessed by Glasgow Outcome Scale (GOS).Results: 145 patients were available for analysis. Median GCS was 6 (IQR 2), while median radiographic injury was a Marshall score 3 (IQR 2). Mean age was 38 ± 16. Univariable ordinal logistic regression analysis of SEP demonstrated significant ordinal prediction of outcome according to GOS; favorable outcome correlated with higher SEP grading [OR = 1.66, 95%CI (1.29‐2.13), p = 0.001]. A dichotomized model grouping higher/lower SEP grading significantly predicted favorable/unfavorable outcomes, respectively [OR 4.69, 95%CI (2.36‐9.32), p < 0.001]. These findings remained significant regardless of age, admission GCS, or radiographic injury severity (p > 0.59).Conclusion: Graded SEP testing in acute sTBI is a significant predictor of outcome at one year.Keywords: Neurophysiology, Outcomes, Severe, PrognosisDBA‐03 GUT MICROBIOTA DYSBIOSIS EXACERBATES HIPPOCAMPAL NEURONAL LOSS AND MALADAPTIVE FEAR MEMORY RESPONSE FOLLOWING TBIMarta Celorrio‐Navarro, Trey Rhodes, Victoria Goodwin, Sangeetha Vadivelu, Sophia Xiao, Brian Jwa, Ashley Steed, Stuart FriessWashington University in St. Louis School of Medicine, Pediatrics, St Louis, USAThe influence of the gut microbiota on traumatic brain injury (TBI), tissue repair, and subsequent neurological outcome is presently unknown. This knowledge gap is of paramount clinical significance as TBI patients are highly susceptible to alterations in the gut microbiota due to frequent antibiotic administration, prolonged hospitalization, and autonomic dysfunction. To induce microbial dysbiosis, mice were randomized to drinking water with broad spectrum antibiotics (vancomycin, neomycin, ampicillin, and metronidazole: VNAM) dissolved in Kool‐aid or Kool‐aid alone as control. Two weeks later, mice were randomized to controlled cortical impact or sham surgery with or without continuation of antibiotics post‐injury. Contextual fear conditioning and testing were performed on post‐injury days 5‐7 with sacrifice 90 minutes after testing on post‐injury day 7. As expected, we found microbial dysbiosis as determined by 16S sequencing and loss of gut epithelial integrity in animals treated with antibiotics. Importantly, we found increased neuronal cell loss in the ipsilateral (IL) CA3 region of the hippocampus in antibiotic treated mice 7 days after injury. We further characterized this cell loss and, specifically, found fewer parvalbumin GABAergic inhibitory interneurons in the IL hippocampi of the injured mice exposed to VNAM. Additionally, we observed increased c‐Fos staining in the IL ventral hippocampi of injured, VNAM treated mice. Furthermore, we found a maladaptive fear memory response with increased contextual freezing response in uninjured, VNAM treated mice compared to the sham controls. Importantly, we also observed an increased contextual freezing response in the injured, VNAM treated mice compared to both the injured/uninjured non‐antibiotic treated mice. Given our hypothesis that microbiota dysbiosis leads to poor outcomes after TBI via microbial‐host immune interactions, further studies are investigating the peripheral and/or local immunological environment and subsequent effect on tissue repair after TBI.Keywords: gut microbiota, dysbiosis, fear memory, traumatic brain injury, inhibitory neuronsDBA‐04 SALIVARY NON‐CODING RNAS: THE NEXT GENERATION OF BIOMARKERS IN CONCUSSED RUGBY FOOTBALL UNION PLAYERSValentina Di Pietro1,2,3, Patrick O'Halloran1, Callum N Watson1, Kamal M Yakoub2, Conor Bentley2, David J Davies1,2, David Menon4, Keith Stokes5,6, Matthew J Cross5,7, Simon PT Kemp6, Antonio Belli1,2,31University of Birmigham, Birmingham, UK2Queen Elizabeth Hospitals, Birmingham, UK3MiRNA diagnostics ltd, Birmingham, UK4University of Cambridge, Cambridge, UK5University of Bath, Bath, UK6RFU, London, UK7Premiership Rugby Ltd, London, UKThe Study of Concussion in Rugby Union through MicroRNAs (SCRUM) is a prospective observational study of players participating in the two highest tiers of senior professional male domestic rugby competition in England, the Premiership and Championship. The study was embedded within the ongoing Rugby Football Union Injury Surveillance Programme for the 2017‐2018 season. As defined in the Head Injury Assessment protocol, whenever a player suffered a head impact event with either clear/immediate signs of concussion OR with the potential for concussion, they were asked to provide saliva samples at the following timepoints: during the game (HIA1), post‐match (HIA2) and at 36‐48 hours following the game (HIA3). This created 2 categories of player suffering a head impact event; 1. Those with clear/immediate signs of concussion or diagnosed with concussion after multi‐timepoint clinical assessment and 2. Those in whom concussion was excluded after multi‐timepoint clinical assessment. In addition, uninjured control players and players removed from the game with other musculoskeletal trauma were asked to provide samples at the HIA2 and HIA3 timepoints in order to exclude contamination from non‐neurological injuries and orthopaedic‐trauma.An initial discovery‐set of 60 samples was evaluated through next generation sequencing. A panel of 94 small non‐coding RNAs (sncRNAs) survived to an FDR <0.05 or p‐value <0.05, comparing concussions and controls. The selected panel, was used for a validation study on 598 samples, by qPCR. When comparing the groups, 21 sncRNAs were found to be differentially expressed with p‐value <0.05, four of which survive the Benjamini‐Hochberg correction at a significance level of 0.05. These biomarkers can provide a novel, non‐invasive way to support the diagnosis of concussion.Keywords: microRNA, salivary biomarkers, Rugby Football UnionDBA‐05 DISRUPTIONS IN HIPPOCAMPAL CIRCUITRY INVOLVED IN LEARNING AND MEMORY FOLLOWING TBIKimberly Gagnon1, Carlo Cottone1, Christopher Adam1,2, John Wolf1,21University of Pennsylvania, Department of Neurosurgery, Philadelphia, USA2Corporal Michael J. Crescenz VA Medical Center, Philadelphia, USAHippocampal‐dependent memory loss is a common and persistent sequela following TBI, but the mechanisms underlying memory impairment are poorly understood. Previous studies using the fluid percussion injury (FPI) model have reported a loss of power in frequency‐bands necessary for the organization of neuronal ensembles that underlie spatial memory tasks. Learning and memory impairment following TBI may be due in part to communication disruptions within the hippocampus via altered neuronal oscillations and disorganized ensemble activity. To examine these potential mechanisms, rats were chronically implanted in the hippocampus with 64 channel silicon probes following injury, which allowed for recordings of single units and simultaneous laminar field potentials. Wireless recordings were performed while the animal was exploring a familiar and a novel environment to examine the effects of injury on the formation/organization of neuronal ensembles during memory encoding and retrieval. Preliminary results show that place cells are formed in rodents post‐injury, but these neurons show a reduction in spatial specificity. Importantly, CA1 firing rates remain unchanged compared to sham, suggesting disorganized timing. An entrainment analysis of single unit timing to oscillations involved in the organization of neuronal ensembles such as theta (4‐12 Hz) and high gamma (60‐110 Hz) showed a shift in phase preference in injured animals. This entrainment is coordinated by input from the entorhinal cortex and CA3 region. A laminar analysis showed a loss in the phase‐amplitude coupling of these oscillations at all levels of the dendritic arbor. Interestingly, the differences in entrainment to the theta oscillation specifically during encoding of memory is lost in the injured animal. Radial arm maze testing showed that rats were able to remember a previously learned pattern of rewards post‐injury, but were unable to encode a new pattern. These reductions in oscillatory power, as well as the disordered timing of CA1 unit firing, may disrupt hippocampal encoding leading to learning and memory deficits post injury.Keywords: hippocampus, diffuse axonal injury, spatial memory, ElectrophysiologyDBA‐06 MICROGLIA ELIMINATION RECOVERED PERIPHERAL INFLAMMATION‐INDUCED SLEEP BUT PROLONGED TBI‐INDUCED SLEEP IN MICEKatherine Giordano1,2, Tabitha Green1,2, JB Ortiz1,2, Maha Saber1,2, Yerina Hur1,2, Helena Morrison3, Jonathan Lifshitz1,2,4, Rachel Rowe1,2,41University of Arizona COM‐Phoenix, Phoenix, USA2Barrow Neurological Institute at Phoenix Children's Hospital, Phoenix, USA3University of Arizona College of Nursing, Tucson, USA4Phoenix VA Healthcare System, Phoenix, USASleep is regulated by physiological processes such that inflammation and cytokine production may disrupt the signaling required to maintain a healthy sleep profile. Traumatic brain injury (TBI) evokes inflammation and microglia activation, with subsequent sleep disturbances that can exacerbate inflammation and neurological symptoms after TBI. As such, a dynamic feedback loop connects sleep, inflammation, and TBI. We hypothesized that microglia elimination with a CSF‐1R inhibitor (PLX5622) would attenuate cytokine levels and differentially regulate sleep after peripheral‐induced and TBI‐induced inflammation. Mice were administered PLX5622 or control diet (21d) and baseline sleep was measured. PLX5622 eliminated microglia (< 0.5% remained) without significant differences in physiological sleep or peripheral cytokine levels (IL‐6, TNF‐α, IL‐1β) compared to control diet. Mice on PLX5622 and control diet received lipopolysaccharide (LPS;1.2mg/kg i.p) or midline fluid percussion injury (mFPI) and sleep was recorded for 3 days. LPS increased sleep (cumulative minutes/day) regardless of diet (F(1,8) = 11.28, p < 0.01). In LPS mice on control diet, sleep remained increased. In mice on PLX5622, sleep returned to baseline by day 3 post‐injection, with elevated peripheral IL‐6 at day 1 post‐injection (F(2,10) = 25.99, p < 0.0001). TBI increased sleep regardless of diet (F(1,4) = 17.91, p < 0.01). In mFPI mice on control diet, sleep returned to baseline by day 2 post‐injury. In mFPI mice on PLX5622, sleep remained higher than baseline over 3 days post‐injury. Regardless of diet, peripheral cytokine levels were not elevated (1 day post‐injury). Thus, microglia elimination recovered sleep disturbances after peripheral‐induced inflammation, but prolonged sleep disturbances after TBI, which establishes a role for microglia in the feedback loop among sleep, inflammation, and TBI. Microglia represent a plausible therapeutic target to mitigate inflammation, using sleep outcomes as a pharmacodynamic measure in treating TBI and other neuroinflammatory diseases.Funding: PCH‐Mission‐SupportKeywords: Sleep, Microglia, Inflammation, Midline fluid percussion injury, MouseDBA‐07 GLIAL FIBRILLARY ACIDIC PROTEIN LEVELS RELATE TO CHANGES IN NEUROIMAGING FINDINGS AFTER MILD TRAUMATIC BRAIN INJURYVivian De Alvarenga Guedes, Jordan Peyer, Cassandra Pattinson, Candace Moore, Katie Edwards, Christine Turtzo, Christina DeVoto, Chen Lai, Xuemin Zhang, Lawrence Latour, Jessica GillNational Institutes of Health, Bethesda, USAMild traumatic brain injury (mTBI) is associated with variable clinical outcomes and might result in lasting symptoms. Subtle signs of injury that are not seen on CT scans can be detected via MRI, including diffuse axonal injury, which is associated with higher risks for neurologic symptoms. However, MRI is often not available or financially viable in clinical settings. Thus, identifying biomarkers that relate to MRI findings would aid clinicians in injury assessment, mitigating the costs of MRI and allowing for the development of precise treatment models. We aimed to determine whether the blood‐based biomarkers glial fibrillary acidic protein (GFAP), tau, and neurofilament light chain (NFL) can discriminate between mTBI patients who do and do not recover from MRI‐identified injuries. Study participants (n = 157) presented to the emergency department with history of head injury and suspected mTBI. Blood collection and MRI imaging were performed within 48 hours of injury and up to 7 days after the mTBI. Participants were classified into two groups according to MRI findings: MRI‐persistent (MRI‐positive at both timepoints, n = 127) and MRI‐recovered (MRI‐positive at baseline and MRI‐negative at follow‐up, n = 30). Plasma samples were analyzed using an ultrasentive assay. Concentrations of plasma GFAP were significantly higher in the MRI‐persistent group compared to the MRI‐recovered group at baseline (p = .007) and at follow‐up (p = .001). Levels of tau and NFL did not differ significantly between groups. GFAP was a poor method for distinguishing groups at baseline (AUC 0.66; 95% CI 0.54‐0.79), but satisfactory at follow‐up (AUC 0.77; 95% CI 0.69‐0.86). Our findings suggest the potential of blood GFAP as a marker of brain abnormalities in patients with acute mTBI. This biomarker strategy could prove valuable in the stratification of patients in clinical trials and to guide clinical decisions. Funding: NINR, NINDS, CNRM.Keywords: axonal injury, MRI, blood, postconcussive symptomsDBA‐08 MTOR PATHWAY ACTIVATION IMPLICATED IN EPILEPTOGENESIS AFTER TBILauren Hanlon1, William Stewart2, Josephine Atkinson2, John Wolf1, Frances Jensen3, Delia Talos3, Victoria Johnson11University of Pennsylvania, Neurosurgery, Philadelphia, USA2Queen Elizabeth University Hospital, Neuropathology, Glasgow, Scotland3University of Pennsylvania, Neurology, Philadelphia, USAPost‐traumatic epilepsy (PTE) represents a debilitating complication of traumatic brain injury (TBI). The mammalian target of rapamycin (mTOR) pathway is increasingly implicated in various genetic and acquired epilepsies, with compelling data supporting a role in epileptogenesis. Notably, mTOR activation has also been reported following rodent models of TBI in association with mossy fiber sprouting and PTE, both of which were mitigated by mTOR inhibition via rapamycin. However, the role of mTOR in human post‐traumatic epileptogenesis is unknown. Here, we examine mTOR pathway activation using post‐mortem tissue from the Glasgow TBI Archive from individuals with a history of PTE (n = 31), severe TBI without epilepsy (n = 37; survival 2wks‐12yrs), epilepsy without TBI history (n = 26) and controls (n = 26). Immunohistochemistry specific for markers of mTOR activation (pS6‐ser235‐236 and pS6‐ser240‐244) was performed and findings semiquantitatively scored within hippocampus. Mossy fiber sprouting was examined via dynorphin staining. A subset of PTE cases demonstrated marked immunoreactivity for pS6‐ser235‐236 and pS6‐ser240‐244 in the dentate granule cells, compared with normal controls (p < 0.05), and was comparable to that in non‐trauma‐associated epilepsy. Importantly, a subset of TBI cases without epilepsy also displayed increased immunoreactivity for pS6‐ser240‐244 across the survival range (p < 0.05 vs controls). Yet, curiously, pS6‐ser235‐236 immunoreactivity was not increased in this group. While pS6‐ser240‐244 is mTOR pathway‐specific, pS6‐ser235‐236 is also driven by mTOR‐independent pathways and may account for differential findings. Mossy fiber sprouting was more frequent in all groups versus controls (p < 0.05) and was associated with mTOR activation. However, this was not absolute, with sprouting in the absence of mTOR activation observed in some cases. These data indicate mTOR activation may be an important feature in some cases of PTE. Given the recognition that the mTOR pathway contributes to epileptogenesis, the observed mTOR activation in TBI without epilepsy suggests this may also be of mechanistic significance in post‐traumatic epileptogenesis. Funding: CURE‐TAPTE GrantKeywords: TBI, Post Traumatic Epilepsy, Hippocampus, mTORDBA‐09 EXPLORING ADDITIONAL APPROACHES TO THERAPY RANKING IN OPERATION BRAIN TRAUMA THERAPYPatrick Kochanek1, W. Dalton Dietrich2, Deborah Shear3, Helen Bramlett2, Stefania Mondello4, Audrey Lafrenaye5, Kevin K.W. Wang6, Ronald Hayes7, Janice Gilsdorf3, John T. Povlishock5, Sanuel Poloyac1, Philip Empey1, C. Edward Dixon11Safar Center for Resuscitation Research, Univ. of Pittsburgh, Department of Critical Care Medicine, Pittsburgh, USA2University of Miami, School of Medicine, MIami, USA3Walter Reed Army Institute of Research, Silver Spring, USA4Messina University, Messina, Italy5Virginia Commonwealth University, Richmond, USA6University of Florida, Gainesville, USA7Banyan Biomarkers, Alachua, USAOperation brain trauma therapy (OBTT) is a multicenter pre‐clinical therapy and biomarker screening consortium. It screened 12 therapies (nicotinamide, erythropoietin, cyclosporine, simvastatin, levetiracetam, glibenclamide, Kollidon VA‐64, AER‐271, amantadine, minocycline, E64d and P7C3‐A20) in >1500 rats across three TBI models (parasagittal fluid percussion [FPI], controlled cortical impact [CCI], and penetrating ballistic‐like brain injury [PBBI]). We assigned 22‐points per model across behavioral and histological outcomes, generating an overall score for each therapy. Levetiracetam and glibenclamide ranked the highest. We now take two additional therapy ranking approaches. First we examined therapeutic efficacy within models. Second, to maximize positive signals, we examined the impact of eliminating negative points. Several findings emerged. Using a minimum threshold of ≥+3‐points to reflect efficacy within models, no therapy produced a signal in PBBI. Levetiracetam, which produced the highest overall score, produced its strongest signal in FPI, however, no other therapy in FPI reached threshold. In contrast, levetiracetam, glibenclamide, nicotinamide, and Kollidon VA‐64 all met/exceeded threshold in CCI (in that order). Eliminating negative points did not impact scoring for levetiracetam; it generated no negative points across models. However, glibenclamide, nicotinamide, Kollidon VA64, minocycline, and simvastatin all reach threshold (in that order). Eliminating negative points, levetiracetam and glibenclamide remained the highest scoring therapies; both generated ≥2X the points of any other therapy. Glibenclamide showed maximal efficacy in CCI. We conclude, levitiracetam and glibenclamide merit additional pre‐clinical/clinical study, with model dependence for glibenclamide. Eliminating negative points minimally impacted our conclusions. Alternative therapeutic approaches should also be explored across OBTT, particularly in PBBI (i.e., targeting regeneration, cell‐based therapies, or combination therapy). Support: DAMD W81HWH‐14‐2‐0118; W81XWH‐10‐1‐0623Keywords: Consortium, Biomarker, Drug Screening, TranslationDBA‐10 MILD TBI RESULTS IN HIPPOCAMPAL AND CORTICAL OSCILLATORY CHANGES THAT RESOLVE OVER TIMEKatelynn Ondek, Aleksandr Pevzner, Amber Schedlbauer, Ali Izadi, Consuelo Jimenez‐Ornelas, Kiarash Shahlaie, Gene GurkoffUniversity of California‐Davis, Neurological Surgery, Davis, USADespite the high prevalence of mild to moderate traumatic brain injury (TBI), little is known about the factors that play a role in recovery and the potential development of chronic cognitive deficits. We hypothesized that TBI would alter low frequency theta oscillations, corresponding with spatial learning deficits. Following lateral sham injury (n = 13) or lateral fluid percussion (n = 37) in rats, we analyzed electroencephalography (EEG) in the local field potentials from depth electrodes implanted in the hippocampus (HIPP) and anterior cingulate (ACC) in the first two weeks of recovery. All animals received a craniotomy over the right hippocampus, through which TBI rats received an injury (average 2.12 ATM), after which linear arrays of recording electrodes were implanted in both HIPP and ACC. TBI animals had significantly increased righting times (p < 0.001) and significantly greater weight decrease (p < 0.0001) compared to sham. Injured animals did not exhibit elevated latency on the Barnes maze (PID8‐11), although they did exhibit a greater percentage of random searches (p < 0.05), suggesting a mild injury. EEG from both ACC and HIPP collected on post‐injury days (PID) 3 and 7 indicated oscillatory aberrations that resolve over ∼10 days. TBI animals had decreased theta power (p < 0.0001), percent time in theta (p < 0.001), and a change in theta‐delta ratio (p < 0.01) in the HIPP compared to sham on PID3 but not PID7. Analysis of the ACC revealed significantly increased theta power (p < 0.001) on PID 3 and 7 with a decrease in peak frequency (p < 0.01) and significantly decreased theta‐delta ratio (p < 0.05) on PID3. Theta coherence between the HIPP and ACC was decreased on PID3 but resolved over time. These changes correspond with the timeline of neurochemical changes following TBI, such as abnormal calcium accumulation and alterations in glucose metabolism. These data highlight the therognostic potential of EEG analysis to noninvasively diagnose or track recovery following even a more mild TBI.Keywords: Translational, Stereo/Intracranial EEG, Learning and Memory, Lateral fluid percussionDBA‐11 MACROPHAGE INFLAMMATORY PROTEIN‐1ALPHA (CCL3) AND ITS RECEPTORS CONTRIBUTE TO SECONDARY DAMAGE FOLLOWING SPINAL CORD INJURYNicolas Pelisch, Antje KronerMedical College of Wisconsin, Neurosurgery, Milwaukee, USASecondary damage after spinal cord injury (SCI) occurs due to a sequence of events after the primary injury, contributing to increased lesion size and poor locomotor recovery. Thus, understanding secondary damage is critical to minimize tissue damage and improve neurological outcome. We have identified the expression of CCL3, a member of the CC chemokine family, and its receptors at different timepoints after SCI. We used adult female 6‐8 week old and were subjected to a moderate T11 contusion, and the expression levels of CCL3 and its receptors CCR1, CCR4 and CCR5 were characterized at the lesion site. CCL3 was upregulated after injury, with a peak at 6 hours and stayed upregulated for up to 28 days. Similarly, CCR1 and CCR5 expression was also increased at day 3 and 7‐post injury, respectively. CCR4 in contrast did not show any significant change (p < 0.05, n = 4‐5). Next, we compared locomotor recovery in CCL3 knockout and wild type mice. The Basso Mouse Scale locomotor score (BMS) showed that CCL3 knockout mice initially recovered better compared to wild type group, but at later time points, the scores started to decline, while the wild type group showed a slight improvement (p < 0.05, n = 9‐10). Then, we tried to identify the influence of CCL3 under inflammatory conditions in vitro using bone marrow derived macrophages (BMDM). When stimulated with LPS (1μg/ml), CCL3 knockout BMDMs showed significantly lower expression levels of M1 related factors TNF and IL12b compared to wild type. In addition, CCL3 deficient BMDMs had a higher baseline expression of the M2 related marker CD163 (p < 0.05, n = 4‐5), indicating a shift towards reduced pro‐inflammatory phenotype in the absence of CCL3. Collectively, we show that the proinflammatory chemokine CCL3 and its receptors CCR1 and CCR5 might be critically involved in the pathophysiology of inflammation and contribute to secondary damage following SCI, thereby providing a new potential target for SCI therapy.Acknowledgement: Supported by Wings for Life 5088‐01.Keywords: Microglia, Neuroprotection, Behavioral function, Gene expressionDBA‐12 CEREBROVASCULAR REACTIVITY ASSESSED WITH FNIRS AND MRI AS A BIOMARKER OF TRAUMATIC CEREBROVASCULAR INJURYMichael Sangobowale, Justin Morrison, Erika Silverman, Hannah Zamore, Ramon Diaz‐ArrastiaUniversity of Pennsylvania, Neurology, Philadelphia, USAChronic traumatic encephalopathy (CTE) is a neurodegenerative disease found most often in professional contact sport athletes (e.g. football players), military veterans, and others with a history of repetitive head impacts. Presently, CTE can only be diagnosed after death through neuropathological examination. The primary effects of repetitive mild traumatic brain injury (rmTBI) involve not only mechanical neuronal damage but also traumatic cerebral vascular injury (TCVI). TCVI is common after a single traumatic brain injury (TBI) and an attractive target for therapeutic intervention. We hypothesize that cerebro‐vascular reactivity (CVR) will be decreased in patients with TBI. Furthermore, we predict that 5‐week administration of a phosphodiesterase 5 (PDE5) inhibitor (sildenafil) will augment CVR in patients with a history of TBI. Hemodynamic changes were assessed using fNIRS, a noninvasive and portable neuroimaging modality that detects changes in blood oxygenation related to brain function. We assessed CVR by measuring the changes in oxygenated hemoglobin (ΔHbO) and deoxygenated hemoglobin (ΔHbR) concentration produced by mild hypercapnia (5% CO2) with complicated mild TBI (GCS 13‐15 with neuroimaging abnormality) in the acute (within 72hrs after injury, n = 17) and subacute (14 days after injury, n = 15) stages in addition to 11 age‐ matched healthy controls (HC), who were studied once. The change in CVR one hour after the administration of single dose of sildenafil citrate (60 mg orally) was also assessed. Mean (± SD) CVR was comparable in TBI patients and HC at 72 hours (HC:0