key: cord-0007033-v2m5l5wz authors: nan title: Neurocritical Care Society 14(th) Annual Meeting date: 2016-08-11 journal: Neurocrit Care DOI: 10.1007/s12028-016-0301-7 sha: 76fc614d39eeced179b78ad6d4926e4550de8c4e doc_id: 7033 cord_uid: v2m5l5wz nan In this exploratory analysis, CSF levels in the progesterone treated group were variable and not as supratheraputic compared to serum levels. This study highlights an additional factor that needs to be considered in the designing of clinical trials in TBI. Not only does the heterogeneity of the injury and subsequent outcome measures need to be refined, but the biomarker of pK levels also needs to be analyzed in CSF as well as serum to determine if the treatment is reaching the target organ, the brain. Arctic ground squirrels (AGS) are extreme hibernators capable of withstanding months of freezing temperatures by suppressing metabolic rate. Hibernation is characterized by hypoxia and low cerebral blood flow and interrupted by bouts of arousal in which perfusion is quickly restored. Curiously, AGS do not experience reperfusion injury which is hypothesized to be reflected in altered transcriptional signatures in an in vitro model of reperfusion injury. To investigate the influence of ischemia/reperfusion on AGS neuronal stem and neural progenitor cells (NSC/NPCs), we exposed AGS and murine NSC/NPCs to control conditions, hypoxia, oxygen and glucose deprivation or glucose deprivation alone or following return to normal conditions to model reperfusion. Cell viability and cell cycle state were assessed by automated cytometry; metabolic phenotype by in vitro oxygen consumption and extracellular acidification rate. To determine novel genes involved in AGS resilience to reperfusion injury, a cDNA library was constructed in a mammalian expression vector and introduced into murine NSCs that were then assayed for viability after ischemia/reperfusion. AGS NSC/NPCs demonstrated marked resistance to ischemia/reperfusion injury compared to murine NSC/NPCs. This survival phenotype is associated with suppressed mitochondrial oxidation and altered cell cycle regulation. AGS genes regulating mitochondrial function strongly modulated murine NSC/NPC viability following ischemia/reperfusion injury. A dynamic ability to suppress mitochondrial oxidation may underlie resilience to reperfusion injury in AGS by promoting a quiescent cell cycle phenotype. Development of therapeutic agents suppressing mitochondrial oxidation may induce a protective phenotype and promote survival following reperfusion injury. High Throughput Imaging of Motor System Connectivity in the Mouse Brain. Stroke results in profound alterations to architecture in the brain, particularly the corticospinal tract (CST). Some plasticity may contribute to functional recovery, while other changes may be maladaptive. Studies of CST connectivity have been limited by standard imaging methods which do not allow visualization and analysis of global axonal connectivity in the brain. We employed two novel imaging methods to visualize axonal projections to forelimb musculature. First, a pseudorabies viral (PRV) vector carrying green fluorescent protein (GFP) was injected into the left forelimb flexor in naïve 10-12 week-old C57 mice. PRV was transported retrogradely and transynaptically, labeling neurons in the motor cortex and other regions of interest. Whole slide imaging was performed using an automated slide scanner (Nanozoomer, Hamamatsu Photonics K.K., Hamamatsu City, Japan) producing images of serial coronal sections, allowing visualization of multiple levels of the brain in a single slide image. For serial two-photon tomography (STPT), UTSW Whole Brain Microscopy Facility used a TissueCyte 1000 imaging system (Tissue Vision, Somerville, MA) which uniquely performs automated sectioning and fluorescent imaging of the brain to produce 3-dimensional images with micronlevel resolution. This allows for unprecedented visualization of axonal connectivity in the whole brain. We have imaged motor systems in the brain of uninjured mice using two different methods, each with distinct benefits. Whole slide imaging allows for quantification of regions of interest in the brain on a single slide, while STPT produces a highly detailed image that improves our understanding motor systems in 3-D space. Future directions will investigate changes in connectivity following stroke injury and during recovery, allowing a greater understanding of the complexity of plasticity and how it contributes to beneficial and pathological circuit remodeling after injury. Andexanet alfa (AnXa) is a modified recombinant factor Xa (FXa) derivative that sequesters direct FXa inhibitors and reverses their anticoagulation effects. Non-specific prothrombin complex concentrates (PCCs) have been proposed as potential reversal strategies. The objective of these studies was to compare 3-factor (Bebulin) and 4-factor (Kcentra®) PCCs, approved for reversal of warfarin, with AnXa in reversing anticoagulation effects of rivaroxaban in a rabbit model of bleeding. NZW rabbits were treated with rivaroxaban (IV, 1 mg/kg) and 30 min later, either PCCs (3-factor PCC: 60 or 120 mg/kg; 4-factor PCC: 25, 50, or 100 IU/kg) or AnXa (75 or 125 mg/rabbit) was administered IV. Liver injury was then induced with 10 1-cm incisions following laparotomy, and blood loss was measured for 15 min. Plasma concentrations of unbound (active) and total rivaroxaban, as well as pharmacodynamic (PD) parameters (anti-FXa activity, PT, and aPTT) were determined. AnXa reduced blood loss in rivaroxaban-anticoagulated rabbits to levels seen in non-anticoagulated rabbits. In rabbits treated with AnXa, anti-FXa activity and unbound rivaroxaban were reduced dosedependently by >95% and >90%, respectively, within 5 minutes, and both parameters correlated with reduction in blood loss. In contrast, 3-factor or 4-factor PCCs had no significant effect on any of these markers in rivaroxaban-anticoagulated rabbits. AnXa effectively reversed the anticoagulation activity of rivaroxaban in a rabbit model of bleeding. In contrast, PCCs showed no reversal activity as assessed by blood loss or PD markers. These results suggest that use of non-specific PCCs as reversal agents for direct FXa inhibitors are not likely to be as effective as specific reversal agents that target FXa inhibitors. Investigation of AnXa vs. PCC to reduce hematoma expansion in models of intracranial hemorrhage is warranted. Outcomes after resuscitation from cardiac arrest (CA) remain poor. Preventable secondary injury from ongoing brain tissue hypoxia (BTH) may worsen injury burden. Unfortunately, markers to allow individualized, real-time care optimization are lacking. We performed a randomized crossover trial in a swine model of opioid-induced CA to 1) determine the prevalence of BTH with standard care (StdC), and 2) test whether neuromonitor-guided goal-directed care (NGDC) can prevent BTH. Female swine (25-30kg) were anesthetized with propofol and fentanyl. We placed femoral arterial and venous sheaths, a continuous cardiac output pulmonary artery catheter (Edwards LifeScience) and a right frontal intracranial access bolt (Hemedex) with probes for brain tissue oxygen (PbtO2), pressure (Raumedic), microdialysis (mDialysis70), cerebral blood flow (CBF) (Hemedex), and an 8-contact electroencephalographic depth electrode (AdTech). We induced apnea with 30mcg/kg fentanyl, extubated the animal and began ACLS 9min after apnea. After 1h stabilization, animals with return of spontaneous circulation (ROSC) were randomized to three alternating 6h care blocks: StdC (mean arterial pressure>65mmHg, oxygen saturation 94-98%, cardiac output>75% baseline) or NGDC (PbtO2>20mmHg, CBF>20mL/100g/min). Animals were euthanized at 18h post-ROSC. Our primary outcome was the effect of care block on PbtO2, which we analyzed at 1min resolution using generalized estimating equations with robust standard errors. Overall, 8 of 17 animals achieved ROSC after 13±3min. PbtO2 was higher during NGDC than StdC (P<0.001) and did not differ during NGDC from pre-arrest. PbtO2 was <20mmHg more during StdC than NGDC (40% of minutes vs 1%, P<0.001). CBF was lower during NGDC than StdC (P<0.001), and lower in both arms than pre-arrest (both P<0.001). Brain tissue hypoxia was common in this cardiac arrest model and prevented by neuromonitor-guided goal-directed care. Lower CBF and higher PbtO2 during goal-directed care implies preserved hypoxic cerebral vasodilation and diffusion-limited oxygen delivery. Future work will incorporate electroencephalographic and metabolic injury markers. SAGE-547 is a proprietary formulation of the endogenous neurosteroid allopregnanolone, being studied as a potential adjunctive therapy for the treatment of super-refractory status epilepticus (SRSE) . SRSE refers to a condition of persistent seizures that have failed treatment with first-, second-and third-line treatments. In preclinical models, prolonged seizures reduce the surface expression of synaptic GABAA receptors, exacerbating neuronal excitability and limiting target sites for GABAergic treatments (eg. benzodiazepines). Here we present preclinical data describing the pharmacological properties of SAGE-547 that support its further development as a potential treatment for SRSE. GABA-evoked currents were examined in vitro with whole cell patch clamp recordings in cells expressing -pilocarpine model of refractory status epilepticus (Pouliot 2013) was used to examine in vivo anticonvulsant activity. SAGE-547 or pentobarbital was administered intravenously 60 minutes after the onset of pilocarpine-induced seizures, a time point when benzodiazepines are ineffective in animal models (Pouliot 2013). SAGE-547 potentiated both synaptic-vitro, with EC50s of 60 nM and 80 nM, respectively. The concentration-GABAA receptors by SAGE-547 was 16,000-fold more potent than that observed with pentobarbital alone. When SAGEs were also observed in the rat model of RSE. When sub-active doses of SAGE-547 and pentobarbital were combined, electrographic seizure activity was significantly reduced. In vitro, SAGE-547 potently modulated both synaptic-type and extrasynaptic-type GABAA receptors, and the maximal potentiation at these receptors was further augmented by the co-application of pentobarbital. This enhanced in vitro potency and maximal effect at GABAA receptors provides further support for the development of SAGE-547 as a potential treatment for SRSE. Vasospasm has long been considered the primary mechanism underlying delayed cerebral ischemia (DCI) in subarachnoid hemorrhage (SAH), but increasing evidence shows that other processes such as cortical spreading depressions and inflammation. We propose that abnormal neural activity in the form of epileptiform abnormalities, we term ictal-interictal continuum abnormalities (IICAs), may contribute to DCI. These abnormalities may increase metabolic demands in injured brain tissue, thereby contributing to metabolic crisis and secondary neuronal injury. Here, we investigate whether the presence of IICAs predict DCI development. We analyzed EEG reports from 124 ICU patients with moderate-severe non-traumatic SAH. Continuous EEG data was recorded with daily review to identify electrographic seizures and interictal patterns. We tallied daily seizures, sporadic epileptiform discharges, lateralized or generalized periodic discharges (LPDs and GPDs), and lateralized or generalized rhythmic delta activity (LRDA and GRDA). Delayed cerebral ischemic events were also marked. Cumulative distribution curves and IICA-to-DCI time plots were calculated. IICAs are more prevalent in patients who develop DCI, especially when they begin several days after the onset of SAH. All IICA types except generalized rhythmic delta activity occur more commonly in patients who develop DCI. In particular, IICAs that begin later in hospitalization correlate with increased risk of DCI (LRDA day 9, ED day 6, LPDs day 12, GPDs day 5) Most IICAs also precede the onset of DCI. We next trend features of discharges to identify those most closely associated with DCI and will present our preliminary findings. IICAs represent a new marker for identifying early patients at increased risk for DCI. Moreover, IICAs might contribute mechanistically to DCI and therefore represent a new potential target for intervention to prevent secondary cerebral injury following SAH. Up to 80% of patients resuscitated from cardiac arrest remain in a coma, and the ability to predict longterm neurologic recovery in these patients is limited. Quantitative analysis of electroencephalography (qEEG) is objective and may facilitate outcome prediction. Consecutive patients with hypoxic-ischemic coma were enrolled. Continuous EEG was obtained on all patients. EEG was post-processed and analyzed by Fourier transform. Spectral analysis was conducted on artifact-free contiguous 5-minute EEG epochs from each hour. Whole band (1--4 Hz) --suppression ratio were computed as quantitative metrics of EEG for the entire EEG recording, and then statistically compared during the last 6 hours of EEG. Sedation, level of arousal, and body temperature were also analyzed. Good outcome (good neurologic outcome, GNO) was defined as consciousness recovery at any point in the acute hospitalization. Ten subjects were included in the study, with cEEG durations ranging from 12 -72 hours of recording. The mean age was 49.6 years (18-65). There were significant differences in alpha power (1.39 (0.66 -1.79) vs 0.27 (0.17 -0.48), median (IQR), p<0.05, GNO vs poor neurologic outcome [PNO] ), delta power (2.78 (2.21 -3.01) vs 0.55 (0.38 -0.83), median (IQR), p=0.01, GNO vs PNO), burst suppression ratio (0.66 (0.02-2.42) vs 73.4 (48.0 -97.5), median (IQR), p=0.01, GNO vs PNO), and multiple measures of variability between GNO and PNO patients. Quantitative spectral analysis of continuous EEG may be predictive of consciousness recovery in patients with hypoxic-ischemic coma. Higher alpha power, lower burst suppression ratio, and higher variability were all correlated with good outcome. Because the media plays an important role in educating the public and impacting public perception on medical topics, we sought to evaluate whether mainstream media provides education or misinformation to the public about brain death through review of articles on two recent highly publicized brain death cases: 1) the Jahi McMath case, in which a teenage girl was declared brain dead and her family refused to allow organ support to be discontinued; and 2) the Marlise Muñoz case, in which a pregnant woman was declared brain dead and the hospital refused to terminate organ support until they were ordered to do so by a judge. media websites of 2014 using the search terms, "Jahi McMath" and "Marlise Muñoz." Each article was evaluated to determine whether it contained 1) teaching points, or 2) misinformation, defined as misleading, incomplete, or incorrect information. We reviewed 208 unique articles. The subject was referred to as being "alive" or on "life support" in 72% (149) of the articles, 97% (144) of which also described the subject as being brain dead. A definition of brain death was provided in 4% (9) of the articles. Only 7% (14) of the articles noted that organ support should be discontinued after brain death declaration unless a family has agreed to organ donation. Reference was made to wellarticles and 47% (16) of these implied both patients were in the same clinical state. Mainstream media provides poor education to the public on brain death. Because public understanding of brain death impacts organ and tissue donation, it is important for physicians, organ procurement organizations, and transplant coordinators to improve public education on this topic. Neurocrit Care (2016) 25:S1-S310 Post-operative hemorrhage: a possible predictor of delirium in brain tumor patients Post-operative delirium after brain tumor resection is frequent, difficult to manage, and may increase chart review of patients admitted to the Neurologic Intensive Care Unit (NICU) after brain tumor resection. We also evaluated the effect of agitated delirium on length of stay. Medical records of NICU admissions form 2013-2015 were reviewed to identify cases and controls. Cases were defined as patients with no pre-existing neuropsychiatric history who experienced significant agitated delirium post-operatively, defined by requirement for treatment with neuroleptics (quetiapine or dexmedetomidine) <48 hours after surgical resection. We compared these patients to a control group comprised of randomly selected patients admitted after brain tumor resection who did not experience agitated delirium. In a multifactors: sex, age, tumor location, pathology, postpost-operative hemorrhage, use of steroids and prophylactic anti-epileptics, particularly levetiracetam. There were 48 cases and 49 controls. Multivariate analysis revealed male sex (O.R 3.47; 95% CI 1.17-10.32; p=0.025) and a post-operative course complicated by hemorrhage within the resection cavity (O.R 10.13; 95% CI 3.27-31.64; p<0.001) as significant predictors of agitated delirium. The ICU length of stay was significantly longer in those with agitated delirium (6.3±5.1 days vs. 1.8±1.4 days; p<0.001). Neurointensivists caring for post-operative patients with brain tumor may consider resection site ctor for developing agitated delirium. Future studies may investigate -operative bleeding and delirium and the long term outcome of these patients. Cardiac arrest is a leading cause of death and disability, and predicting outcome in these patients is a challenge. Optic nerve sheath diameter (ONSD) on brain CT correlates closely with intracranial pressure. Therefore in this study we studied correlation between ONSD measured on the initial brain CT in patients after cardiac arrest and outcome. This is a retrospective study of patients with cardiac arrest admitted to the medical intensive care unit at our hospital between 2009 and 2014. Pati arrest were included. Demographics, neurological status on arrival and day 14 and outcomes were collected. ONSD on brain CT was measured bilaterally 3mm behind the optic nerve head and averaged for each patient. A total of 71 patients were included. Mean age was 66±13.6 years, 54 patients (76%) were male. 46 patients (65%) suffered out-of-hospital cardiac arrest. Mean Glasgow Coma Scale (GCS) on admission was 4.35±1.96 (range 3-9; median 3). Return of spontaneous circulation (ROSC) time was 15.63±9.37 intervention. 70 patients (98.59%) underwent therapeutic hypothermia. 17 patients (23.9%) had seizures. Average modified (19.7%) had a good outcome (mRS 0-3). Average duration from ROSC to CT was 1.3±2.04 days. Mean ONSD in patients with GCS 3-8 at day 14 was 5.76±0.69mm, while in those with GCS 9-15 at day 14, ONSD was 5.81±0.72mm (p=0.78). Mean ONSD in patients with mRS 0-3 at discharge was 5.56±0.62mm, while in those with mRS 4-6 was 5.88±0.71 (p=0.28). There does not appear to be a significant correlation between the ONSD o cardiac arrest and outcome at day 14. Traumatic brain injury (TBI) is a major public health problem. While the association between subarachnoid hemorrhage and systolic dysfunction (SD) has been established, the effect of TBI on the incidence of early SD in previously healthy patients following moderate-severe TBI, and 2) To describe the longitudinal change in We conducted a prospective cohort study among mild and moderate-severe TBI patients admitted to a level 1 trauma center with these inclusion criteria: 1) Age < 65 years, 2) No severe non-TBI injuries, 3) No prior cardiac disease, and 4) Minimal comorbidities. Transthoracic echocardiograms were performed at <24 hours, 3-4 days, and 7-9 days following TBI. Systolic function was assessed using fractional shortening (FS), and SD was defined as FS<25%. Descriptive statistics were used to compare the mild and moderate-severe TBI groups. Multivariable linear regression was used to compare fractional shortening between groups. 64 patients were studied (32 mild TBI and 32 moderate-severe TBI). Both groups were young (36.2 years mild TBI and 36.5 years moderate-severe TBI) and mostly male (69% mild TBI and 84% moderatesevere TBI). Early SD was present in 7 (22%) moderate-severe TBI patients and 0 (0%) mild TBI patients (p<0.01). On multivariable regression, moderate-severe TBI was associated with an absolute 4.2% reduced FS compared to mild TBI (95% CI 0.8% -7.6%, p=0.02). All patients with early SD recovered to normal systolic function by 7-9 days injury ( Figure 1 ). SD is common early after moderatehospitalization. TBI severity is independently associated with worse systolic function. Early echocardiography is a safe, applicable, and feasible procedure following TBI and may aid clinicians with hemodynamic management post TBI. Sophie Samuel. Department of Pharmacy. Memorial Hermann Medical Center, Houston, TX, USA. Paroxysmal sympathetic hyperactivity (PSH) is a neurological condition that occurs most frequently after traumatic brain injury (TBI). Sympathetic hyperactivity can manifest as increased heart rate, blood pressure, respiratory rate, temperature, sweating and posturing activity. In 2014, a consensus statement and probability of diagnosis. The objective of this study is to report the incidence, frequency and severity of PSH symptoms in the acute setting after TBI using the new diagnostic criteria. This was a retrospective study from July 2014-August 2015. Included were all adult patients admitted to a tertiary care intensive care unit with a primary diagnosis of TBI and a length of stay longer than 14 days. symptoms were recorded. The most frequent symptom was tachycardia (53%), followed by tachypnea (47%), hypertension (26%), 2 symptoms occurring simultaneously and 68% had symptom -6) vs 8(3-12); p=0.02] and at day 14; [10(7-14) vs 14(14-15); p<0.01.] Medications often used to control symptoms included, anti-pyretics, opioids, beta---38) vs 17(14-26); p=0.03], but no difference in mortality. Symptoms of sympathetic hyperactivity were seen commonly after TBI in the acute setting. Using a admitted with a hospital length of stay greater than 14 days. Hypertonic saline (HTS) appears to be more effective than mannitol in reducing raised intracranial pressure (ICP) after severe traumatic brain injury (TBI). In this study we investigated which agent had superior combined effects on ICP and cerebral perfusion pressure (CPP). The Brain Trauma Foundation TBI-Patients who received only HTS were identified and matched with patients who received mannitol only (1:1 and 1:2 match). The two groups were matched for age, pupillary reactivity, Glasgow Coma Scale (GCS), CT abnormality, craniotomy and occurrence of hypotension on day 1. Univariate analysis was performed to compare combined average and duration of ICP>25mmHg (ICPhigh) and CPP<60mmHg A total of 25 patients with severe TBI, who received only HTS were identified and matched with 25 (1:1) and 48 (1:2) patients who received mannitol only. In the 1:2 group 1 HTS patient was excluded, as there was no corresponding match in mannitol group. The mean age, GCS, incidence of abnormal pupils, hypotension, abnormal CT, craniotomy and day of ICP insertion were similar in the groups. There was no difference in number of days of ICP monitoring (p=0.09, 0.46; 1:1, 1:2 groups). Osmolar doses were comparable; all patients in HTS group received 3% HTS except one who received 23.4%. In 1:1 match, number of days with CPPlow (2.0±1.7 vs. 3.6±2.8, p=0.03) was significantly lower in the HTS group. In p=0.01) were significantly lower in the HTS group. These results were reproduced in the 1:2 analyses. HTS is superior in its combined effect on ICP and CPP after severe TBI when compared to mannitol. Statins constitute a class of medications commonly used in the treatment of elevated cholesterol. However, in experimental studies statins also have other non-cholesterol mediated mechanisms of action, which may have neuroprotective effects. The purpose of this study was to determine if administration of atorvastatin for 7 days after injury would improve neurological recovery in patients with mild traumatic brain injury (mTBI). The hypothesis was that atorvastatin administration would reduce post-concussion symptoms and also -injury would be safe. 140 patients with mTBI were placebo for 7 days starting within 24 hours of injury. Assessments of post-concussion syndrome, postthe Rivermead Post-Concussion Symptoms Questionnaire at 3 months was the primary outcome. Enrollment in the trial was stopped early because of difficulty in recruiting sufficient numbers of subjects. 52 patients with mTBI were enrolled; 28 patients received atorvastatin and 24 received placebo. The mean Rivermead score was 7.6 for the atorvastatin group compared to 9.2 for the placebo group at 3 months post-injury [F(1,4) = .1855, p=.6687)]. The change in the Rivermead score between baseline and 3 months was also analyzed. The mean change in score was a decrease of 8.1 for the atorvastatin group and 10.7 for the placebo group [F(1,47)=.4132, p=.5235]. No serious adverse events occurred, and there was no significant difference in the incidence of adverse events in the two treatment groups. Atorvastatin administration for 7 days post-injury was safe, but there were no significant differences in neurological recovery after mTBI with atorvastatin. Association between comorbidities, nutritional status, and anticlotting drugs and neurological outcomes in geriatric patients with traumatic brain injury An essential part of the management of traumatic brain injury (TBI) is the mitigation of secondary insults to the brain such as sustained increases in intracranial pressure (ICP). It would be beneficial to be able to predict increased ICP so as to facilitate safe transport of patients. Given the role of neuro-inflammation in increases in ICP. after admission. Receiver Operating Characteristic (ROC) curves were used to compare the predictive elevation of ICP above 20 or 30 mmHg for 5 min or more in the following 6 hours. 250 serum samples from 24 patients were matched to subsequent 6 hour periods of monitoring.. NI-VS the predictive capacity of a combined model of NI-VS and IL4 level over NI-VS alone in predicting ICP elevation to >20 mmHg (0.8 vs 0.77, p30 mmHg (0.91 vs 0.79 p <0.005). levels when combined with physiological data. Even without invasive monitoring, predictions about measurements. Head Injury Neuroworsening (NW) after traumatic brain injury (TBI) is a major cause of added morbidity, however, there is no reliable way to predict NW. We hypothesized that autonomic nervous system dysfunction (ANS) measured by analysis of plethysmograph variability (PPGV) in the first hour after presentation may predict NW in the initial 48 hours after TBI. and head Abbreviated Injury Score(AIS)>1. Patients with systemic trauma were excluded. NW was defined as any of the following occurring in the first 48 hours: new asymmetric pupillary dilatation (>2mm), 2 point GCS decline, interval worsening of CT scan as assessed by the Marshall score, or requirement for neurosurgical intervention. The beat-to-beat variation of the PPG, and PPG morphologic features were calculated to quantify the ANS impact on the physiological status. Multivariate stepwise logistic regression was used to develop predictive models of NW. There were 205 patients (mean age 43 years old, GCS 13, ISS 12, 27% women) who met criteria between December 2011 and May 2013. NW occurred in 24(12%) patients. PPGV analysis at 15 (PPG15) and 60(PPG60) minutes post-admission demonstrated predictive capability for NW(P<0.01). PPGV15 was able to better discriminate NW as compared to a baseline model of age, sex, initial VS (ROC 0.91 v. 0.78, P=0.01). PPGV60 better discriminated future NW as compared to the model of age, sex, admission VS and GCS (ROC 0.98 v 0.86,P=0.02), and marginally better than a model combining admission VS, GCS, and Marshall score on CT(ROC 0.98 v 0.91,P=0.05). ANS dysfunction assessed by continuous PPG waveform analysis in the first hour represents a nonclinical factors to more accurately predict NW, potentially leading to automated algorithms for earlier therapeutic interventions. Tanzania Severe traumatic brain injury (TBI) is the number one cause of death and disability among young adults worldwide. Formulation and subsequent adherence to the Brain Trauma Foundation (BTF) Guidelines has been associated with reduced mortality after severe TBI. In this study we studied epidemiology and treatment of severe TBI at a tertiary referral hospital in Tanzania in reference to the BTF Guidelines. Patients with TBI hospitalized at Bugando Medical Centre, a Tanzanian tertiary referral hospital were recorded in a prospective registry. Demographics, cause of trauma, clinical characteristics, hospital care, and mortality were recorded for 14 days and on the day of discharge. Between September 2013 and October 2015, of 371 patient 47.8%) were hospit pressure (ICP) monitoring were not performed for any patient. Thirty-eigh - Pre-hospital and routine ICU care, CT imaging, blood pressure and ICP monitoring are underutilized or unavailable in the management of severe TBI in the tertiary referral hospital setting. TBI associated mortality is significantly higher than that in high-income countries. Improving outcomes after severe TBI will require concerted investment in pre-hospital care as well as improvement in availability of neuroimaging, ICU resources and expertise in multidisciplinary care. Establishment of comprehensive traum Elevated intracranial pressure (ICP) is thought to mediate secondary brain injury by decreasing cerebral perfusion pressure (CPP) and reducing cerebral blood flow. Clinical trials targeting ICP thresholds have not demonstrated benefit. We hypothesized that stratifying elevations in ICP based on their effect on CPP would be feasible with the use of continuous, time-resolute neuromonitoring data. We studied a convenience sample of five patients undergoing intracranial neuromonitoring after severe traumatic brain injury per our institutional protocol. Patients had a parenchymal ICP monitor time-synced with continuous arterial blood pressure. Waveform data were recorded into Moberg CNS monitors, and second-bywere identified and their duration was measured along with CPP. Elevations were stratified into those that exhibited a reciprocal decrease in CPP (ICP[neg] ) and those with stable or elevated CPP (ICP [preserved] ). The mean patient age was 33; three were male. A total of 290 individual elevations in intracranial pressure were observed over a monitoring duration of 31,573 minutes. We found ICP elevations that reciprocally decreased CPP (ICPneg) were identifiable using bedside neuromonitoring devices, and that these elevations resulted in a CPP drop of nearly 17 mmHg, despite a similar maximum ICP despite a shorter duration. The majority of these ICP elevations occurred in patients with poor outcome, and may represent a target for aggressive ICP lowering therapy. Mild traumatic brain injury (mTBI) is defined as an initial Glasgow Coma Scale (GCS) 13-15. Current recommendations include a follow-up computed tomography (CT) scan of the head prior to discharge. Often, imaging and neurological exam remains stable, questioning the role of routine repeat imaging. A retrospective chart review was completed on TBI patients evaluated at a level 1 trauma center between August 2009 and December 2015. Inclusion criteria included: initial GCS 13-15, blunt head injury, and available repeat imaging. Exclusion criteria included GCS <13, penetrating trauma, those that required immediate surgery, or those without repeat imaging. A total of 198 patients were included in the analysis. Statistics were done with Mann-U Whitney or Chi-Square testing. Age was 49.3±21 years. There were 123 males and 75 females. The most common mechanism was falls (43.9%), followed by motor-vehicle collision (22.7%), motor-cycle accident (12.6%), assault (9.1%), pedestrian--related (5.6%). Polytrauma occurred in 42.6%, better, 59.1% were stable, 24.7% were worse. Only 19 patients (19.6%) exhibited neuro-exam changes, where 16 patients received repeat imaging which ultimately demonstrated stable findings. Ultimately, only one patient required a neurosurgical procedure, an external ventricular drain, due to significant decline. Age, gender, mechan neuro-exam changes. On the other hand, neuro-exam changes significantly correlated with CT changes (p = 0.01). Repeat imaging tend to show worse findings when associated with neurological changes. On the other hand, results for repeat imaging were variable without neurological changes and generally did not alter repeat imaging is only warranted for neurological changes that may necessitate a neurosurgical procedure. create a meaningful tool, we explored family members' and physicians' perspectives on prognosis communication during goals-of-care discussions for ciTBI patients employing mixed-methods. We conducted semi-structured interviews with 16 ciTBI surrogate decision--1 trauma centers, and 20 attending physicians representing geographic (Northeast,Mid-Atlantic,South,West,Midwest] and subspecialty diversity (neurocritical care,neurosurgery,trauma,palliative care). Two independent reviewers analyzed transcribed interviews using deductive and inductive approaches (NVIVO-software). The sample size was determined by theme saturation. prognosticated outcomes expressed as percentages, and 100% preferred prognosis in a "more direct" manner. Surrogates favored percentages because they were "more clear, more concise, and less confusing". In contrast, 75% of physicians stated that they do not use precise percentages when discussing prognosis in ciTBI due to distrust in the predictive accuracy of existing data: "Better have damn good data to do that with, and most often, we do not." Physicians also voiced concern over families' judgment…they become simplified and [these numbers are] used against you later." The dissimilar preferences for the use of percentages and numbers during prognostication represent an important difference between surrogates and physicians. These findings have a direct impact on the design of a goals-of-care SDM tool for ciTBI. A future goals-of-care decision aid will require iterative Decompressive craniectomy (DC) is not a new procedure, however, it has gained momentum in recent years, in the management of refractory intracranial hypertension. However, the timing of DC has not been -based guidelines for the optimal timing of DC has resulted in a wide variability in practice patterns. In most instances, DC has been performed based on neurosurgical evaluation of the patient with or without intracranial pressure (ICP) monitoring and evidence of increased shown variable outcomes in emergency (within 4 hours of injury) neurosurgical procedure in Traumatic Brain Injury (TBI). The purpose of this study was to evaluate the impact of emergency DC on in-hospital mortality following blunt TBI. craniectomies performed on patients with a blunt mechanism of head injury within twenty-four hours of admission, were included in the study. In-hospital mortality was the main outcome of interest. 942 patients qualified for the study. 669 patients (71%) underwent a craniectomy within 4 hours (emergency group)and 273 patients (29%) had craniectomy performed between >4hours hours to 24 hours following hospital arrival (late group). Propensity matched analysis identified 268 pairs of patients in both groups. The mean standardized differences were less than 10% after matching. There were no significant differences in mortality [odd ratio 1.018, CI (0.689 [0.004, 95% CI (-0.078, 0.085), P=1.00] and length of stay between the groups [hazard ratio, 0.770, 95% CI (0.56, 1.059], P= 0.108]. No difference was seen on in-hospital mortality between patients operated within 4 hours versus patients operated between 4 and 24 hours of admission. Spreading depolarizations (SDs) are pathological waves of neuronal depolarization that occur in 56% of patients with traumatic brain injury (TBI) who require surgical treatment of focal lesions. The incidence of SDs in non-surgical TBI i (dEEG) placed at bedside via burrhole. We hypothesized that the incidence of SDs recorded using dEEG in non-surgical patients would be similar to that documented in surgical TBI. -penetrating TBI who did not require urgent neurosurgery on admission. All patients underwent bedside burrhole placement of intracranial pressure, tissue oxygen, regional blood flow and dEEG monitors via a single quad-lumen bolt per institutional protocol. Data were recorded on Moberg CNS with DC-coupled amplifiers. Over a 12--19; 80% male; GCS range 3-7) underwent monitoring for a mea -43.5 hours, beginning 11.0 hours (median, quartiles: 8.1-(80%) monitoring devices were placed in nonpatients had focal pathology in the monitored lobe, w died, including those with SDs. There were no significant hematomas or infections related to invasive neuromonitoring. The incidence of SDs detected with dEEG placed in non-dominant frontal lobe was lower than previously reported with injury-targeted placement of subdural strips in surgical TBI patients. This may be due to targeting to injured periinnate incidence of SD in patients with non-surgical or more diffuse injuries. Targeted placement of subdural electrodes through burr holes may be warranted in non-surgical TBI patients. University of Utah General Surgery, Salt Lake City, UT, USA. Traumatic brain injury (TBI) is a prevalent condition that is responsible for a significant amount of disability and healthcare expenditures. Clinicians can and do use the IMPACT prognostic calculations to inform o examine self-reported individual and institutional use of the IMPACT prognostic calculations in an effort to identify trends and effects on treatment. We conducted an international and multidisciplinary survey examining self-reported awareness and use of the IMPACT prognostic calculator. Factors associated with awareness and use of the calculator including provider specialty, years in practice, personal and institutional volume of TBI patients treated, and institutional trauma level were als voluntary and anonymous survey in an email. Study data was collected and managed using REDCap. 76 respondents (49.7%) were aware of the calculator, only 5.3% (4) said that they often, and 48.7% (37) said they sometimes used it. Volume of TBI cases and specialty both were positively associated with awareness and use of the calculator. 42 providers often or sometimes used calculator, 88.1% (37) stated that it had some influence on their care for the patients. 88.1% used the information to better of care and 19% used it to provide more aggressive care. Of those aware, still only slightly more than half (54%) used it. The use of the IMPACT calculator was mainly to better communicate with patient family, but a portion, 26.2% and 19% of providers, said it influenced their care in other ways. These results provide direction to increase awareness and use of the IMPACT prognostic calculations. The Brain Trauma Foundation guidelines suggest individualizing cerebral perfusion pressure (CPP) goals based on tissue oxygenation (PbtO2), pressure reactivity (index; PRx), and metabolism (lactate-pyruvate ratio; LPR). Our objectives were to investigate practices pertaining to bedside hemodynamic and neuromonitoring in TBI patients, and to analyze differences among "neurointensivists" (NIs; defined as clinical Electronic survey of 22 items including a TBI case-scenario; endorsed by SCCM (9,000 recipients) and ESICM (on-line newsletter) in 2013. Chi-square test was used to compare proportions of responses between NIs and OIs with a significance p< 0.05. There were 655 responders (66% completion rate); 422(65%) were classified as OIs and 226(35%) as NIs. Use of neuromonitoring-derived variables to optimize CPP in patients with severe TBI, for the entire cohort: PbtO2 (18%), transcranial Doppler(TCD) (12%), jugular venous bulb (11%), CT Perfusion(CTP) (4%), PRx (4%), and LPR (3%). NIs use more PbtO2 (28% vs. 10%, p=0.0001) and CTP (7.6% vs. 2.5%, p=0.012). More NIs have a hemodynamic protocol (44.5% vs. 33.3%, p=0.007) for TBI, use more arterial waveform analysis (45% vs. 35%, p=0.019), and bedside ultrasound (37% vs. 27.7%, p=0.023), while more OIs monitor mixed venous oxygen saturation (54.1% vs. 45%, p=0.045). In the case scenario of raised ICP, low PbtO2, and preserved pressure autoregula (vasopressor use 34.5% NIs vs. 23% OIs, p=0.014). "Neurointensivists" employ more hemodynamic and neuromonitoring to patients with TBI. Intracranial pressure and CPP remain cornerstones of management, however the use of other physiologic variables -specific PP goals. The predominant experience of penetrating traumatic brain injury (pTBI) derives from battlefield settings, but the civilian experience in Western settings in patients treated after 2005 is limited to only small and single-center studies. As a result, outcome predictors of civilian pTBI in modern trauma and neurocritical care settings are poorly defined. The aim of this study was to identify predictors associated with survival in a contemporary, large, diverse two-center pTBI cohort, and to develop a parsimonious survival prediction score for civilian pTBI. Our cohort comprised 415 pTBI patients retrospectively identified from the local trauma registries at two U.S. level-1 trauma centers, of which one was predominantly urban and the other predominantly rural. Predictors of in-hospital and 6-month survival identified in univariate and multivariable logistic regression were used to develop the simple Surviving Penetrating Injury to the BraiN (SPIN) Score. at hospital discharge and 6-months post pTBI was 42.4%. Motor Glasgow Coma sub-score, pupillary reactivity, self-inflicted injury, transfer from other hospital, female sex, Injury Severity Score and INR were independently associated with survival (all p<0.001; area-under-the-curve 0.962). Important radiological factors associated with survival were also identified but their addition to the full multivariable would have resulted in model overfitting without much gain in the area-under-the-curve. We developed the SPIN Score, a logistic regressionafter pTBI. While external validation is warranted, this clinical survival prediction tool may provide important information to guide families and physicians during intervention-and goals-of-care decision- Real-time visualization of the cumulative pressure and time dose of intracranial pressure in individual traumatic brain injured patients. The 'dose' of intracranial hypertension, a summary measure of duration and intensity of elevated intracranial pressure (ICP) episodes, is associated with worse outcome in traumatic brain injury ( Retrospective analysis of minute-by-minute ICP monitoring data from a large multicenter database of TBI population-based color-coded plots by Güiza et al, where 'bad' ICP episodes are red, and 'good' ICP episodes are blue. The ICP insult currently experienced by the patient, together with his ICP episodes of the previous 6 hours, and the cumulative ICP burden since ICU admission, are shown. When playing these minute-by-minute snapshots consecutively, an animation is created showing the current and cumulative burden of ICP of the patient. We present the clinical course of 2 patients, 1 with good outcome (GOS 5), and 1 who died (GOS 1). Clinical trials aiming at aggressively treating ICP below a fixed threshold of 20 mmHg have given thresholds, could present a new way to define secondary injury by ICP, and a future target for therapy. The proposed method visualizes the current and cumulative time and pressure burden of ICP for individual patients, which could help a neuro-intensivist in identifying when a patient is currently in a state of potentially harmful elevated ICP, or when his outcome is at a turning point. Acute Blood pressure variation and mortality in severe Traumatic Brain injury We retrospectively evaluate 249 temperations (282hematomas) from April 2011 to March 2015 operated by one young neurosurgeon. During the operation, we made sure to put the burr hole the highest, not to drainage tube into the right position. We scaled the amount of hematoma by hand free ROI using head CT before operation and also the next day to evaluate the improvement rate. The recurrence rate was 8.03% (20 cases). The average amount of air contaminated was 12.3ml and the improvement rate was 56.1%, both of which didn't relate to the recurrence. No relevance among the recurrence rate, the amount of air contaminated and the hematoma improvement rate could be found. Though the amount of air contaminated during temperation doesn't relate to the recurrence rate, it was useful to scale the hematoma improvement rate and the amount of air in order to objectively assess the d recurrence became less, which suggest that we need to investigate the operation for chronic sundial hematoma further after we operate more. The aim of this prospective observational study was to evaluate if trans-cranial Doppler (TCD) ultrasonography can be used as an inference tool of cerebral hypoxic episodes in patients with moderate to severe traumatic brain injury. Recruited patients had serial TCD studies to assess blood flow velocity of the middle cerebral artery (MCA). Measurements were done on bilateral MCAs as soon as logistically possible after the insertion of PbtO2 monitoring, once a day for a total of 5 days, and during dynamic challenge tests when feasible. Multiple physiologic parameters were registered concomitantly with each TCD measurement, with a particular focus on determinants of PbtO2 and potential confounding factors. We studied 17 consecutive patients with a total of 85 TCD studies, of which 29 (34%) were performed 5-14.5h) after TBI. When considering all readings, we found no correlation between PbtO2 and MCA's readings > 24h. For value level, ICP, and CPP. to correlate with brain tissue hypoxia and could be use as a screening tool to help minimise timesensitive secondary injury during that period. Otherwise, Vmean is not correlated to PbtO2. A precise assessment of brain condition after severe traumatic brain injury (TBI) is crucial to reduce secondary injuries and sequelae. Multimodal neuromonitoring permits to assess multiple systemic and brain parameters, but these data are complex to interpret continuously, especially in the overwhelmed environment of intensive care unit (ICU). Computerized Decision Support Systems (CDSS) can assist the clinicians in optimizing care. This study aims to evaluate an algorithm for classifying the cerebral condition, as a first step in the development of a CDSS. The study was approved by the local Research Ethics Committee. Patients with severe TBI (Glasgow coma score < 8) with a monitoring of intracranial pressure and brain tissue oxygenation pressure were eligible. Data were extracted from the existing ICU electronical medical records (Semi Solutions Médicales). An incremental learning fuzzy minparameters online was implemented. The different cerebral status categories included: control condition, ntracranial hypertension. Previously validated and published datasets were used to train the system. The system was then tested with the patients' data and compared to a classification made by two clinical experts. Eight 2-hour recording periods from adults with severe TBI were analyzed. The pathophysiological status was appropriately classified by the CDSS in (median) 93 % (interquartile: 91-97%) of time. Every critical event was detected, but brief misclassifications were frequently observed during the transition periods. In this preliminary cohort of patients with severe TBI, the CDSS was able to adequately classify the brain condition in a large proportion of time, but some errors occurred during brief transitional periods. Further training of the CDSS with a larger dataset may improve the system accuracy, which should be tested in a larger patient population. Mild traumatic brain injury (TBI) is a commonly seen pathology at trauma centers. Neurosurgical consultation is a routine practice; however, the vast majority do not require surgical intervention or invasive monitoring during the entirety of their hospital stay. In certain trauma centers, neurologycentered neurocritical care solely evaluate and manage mild TBI. We provide a retrospective analysis of this practice at our level 1 trauma center. A retrospective chart review was completed on TBI patients evaluated at a level 1 trauma center between September 2009 and December 2015. Inclusion criteria included; initial GCS 13-15, blunt head injury, available repeat imaging, and management by neurology-centered neuro-intensivists. Exclusion criteria included GCS <13, penetrating trauma, those that needed immediate surgery, those with neurosurgical consultation, and those without available imaging. A total of 87 patients were included in the final analysis. Age was 48.8 ± 21 years. There were 50 males and 37 females. The most common mechanism was falls (51.7%), followed by motor-vehicle collision (16.1%), motor--related (8.0%), assault (6.9%), and pedestrian-60.9% were stable, 24.1% were worse. Only 10 patients (11.5%) exhibited neuro-exam changes, where 7 patients received repeat imaging which ultimately demonstrated stable findings. No patients required a neurosurgical procedure. Average hospital stay was 2.8 ± 3.7 days. Neurology-centered neuro-intensivists can manage mild TBI appropriately without official neurosurgical consultation. This practice can streamline TBI management and potentially reduce hospital costs. Bulic, Natasha N. Renda, May M. Kim-Tenser, Gene G. Sung, Benjamin B. Emanuel. USC, Los Angeles, CA, USA. Measurements of optic nerve sheath diameter (ONSD) using bedside ultrasound (US) have been shown to correlate with clinical and radiologic signs and symptoms of increased intracranial pressure (ICP). Eleven patients (7 males, 4 females) with traumatic brain injury (TBI) and GCS<9 were evaluated. All patients had extraventricular drainage (EVD) monitors, 9 right and 2 left. Three patients had right decompressive hemicraniectomy, 4 had left hemicraniectomy, 2 had suboccipital craniectomy and 2 did not have decompression. A total of 29 examinations were obtained with invasive ICP measurements, pulsatility indices (P balance were recorded. Twenty-nine ocular USs were performed on 11 individual patients. In 14 ONS assessments, bilateral ONSD was 5.7 mm, while ICP was 13mmHg, however, later developed ICP 37mmHg within 24 hours. Another patient had bilateral ONS 5.8mm, while ICP was 16mmHg, however, later developed ICP of >30mmHg within 24 hours. Two patients had bilateral ONS measurement> 3.8mm with corresponding ICP >20mmHg All patients with ONSD20mmHg had ONS >5.7mm. Although, patients with ONS >5.7mm and ICP 20mmHg within the next 24h. There was no correlation between PI on TCD, thus TCD was not useful in this dataset. There was no correlation between increased temperature or elevated blood pressure with ICP> 20mmHg. This small sample size suggests that ONSD may predict future ICP elevations, however, a larger sample size is needed to confirm these results. The precise threshold differentiating normal and elevated intracranial pressure (ICP) is variable among individuals. In the context of several pathophysiologic conditions, elevated ICP leads to abnormalities in global cerebral functioning and impacts the function of cranial nerves (CNs), either or both which may contribute to ocular dysmotility. The purpose of this study is to assess the impact of elevated ICP on eye watching a 220-second continuously playing video moving around the perimeter of a viewing monitor. Pupil position was recorded at 500 Hz and metrics associated with each eye individually and both eyes together were calculated. Linear regression with generalized estimating equations was performed to test performed at ICP levels ranging from -3 to 30 mm Hg in twenty-three patients (twelve female, eleven male, mean age 46.8 years) on fiftycorrelating with cranial nerve function linearly decreased with increasing ICP (p-value 12 mm Hg was 0.798. intracranial pressure. Increasingly elevated ICP was associated with increasingly abnormal eye tr physiologic impact of elevated intracranial pressure. This represents a new non-invasive automatable means for assessing the physiologic impact of elevated ICP. Use of shared decision--of-care decisions in critically-ill traumatic brain injury (ciTBI) offers the hope to decrease variation and bias in goals-of-care discussions. SDM guidelines demand the inclusion of an evidenceand acceptance of the "International-Mission-for-Prognosis-and-Analysis-of-Clinical-Trials-in-TBI"(IMPACT)-model by physicians holding such discussions with ciTBI families. We conducted a mixed-methods study with semi-structured interviews in 20 attending physicians representing geographic (Northeast,Mid-Atlantic,South,West,Midwest) and subspecialty diversity (neurocritical care, neurosurgery, trauma, palliative care). We explored methods of prognosis derivation and communication, ciTBI outcome model use, and, specifically, awareness and perceived utility of the IMPACT-model. We analyzed transcripts in NVIVO-software with the investigator-triangulated-inductive--approach. Theme saturation determined the final sample size. Overall,42% of physicians use the IMPACT-model, 25% were not aware of it, and 50% don't rely on any TBI outcome models. Positive IMPACT-model views included: "helpful in getting an idea where your confidence should be"; "to ground physicians a little bit"; "reduces the variability of prognosis that a large some participants "do not thin only"; "those calculators are about populations; they're not about individuals, use of those calculators for this purpose is a perversion of the original construct". We identified substantial physician variability in the awareness of, use, and attitude toward the IMPACTmodel, which crucially informs the development and successful implementation of future goals-of-care SDM tools in ciTBI. The Analyses of Acute Subdural Hematoma Using Acute subdural hematoma (ASDH) is associated with cerebral contusion and laceration of bridging veins following a head injury, however a few cases of ASDH without head injury had been reported. The purpose of this study was to detect the difference of traumatic ASDH and non-traumatic one. 204 cases of ASDH hospitalized at our institute from March 2014 to March 2016 were retrospectively reviewed. Traumatic groups were 189 patients (92.6%), and non-traumatic groups were 15 patients (7.4%). The results were statistically analyzed by logistic regression to use the various factors: age, gender, Glasgow Coma Scale (GCS) Score, the presence of light reflection, Danti--coagulation agents, neurological outcome and so on. Neurological outcome was evaluated using Glasgow Outcome Scale, and it was classified into two groups: the good prognosis group (GR, MD) and the poor prognosis group (SD, VS, D). Traumatic groups were mean age 67.8 ± 19.7 years, and were comprised of 121 males and 68 females. Non-traumatic groups were mean age 74.5 ± 19.5 years, and were comprised of 6 male and 9 females. There were significant difference in patients` characteristics, the presence of light reflection, D-dimer and neurological outcome (p <0.05). Non-We will be described the detail of them and collect further cases in the future. Traumatic brain injuries (TBI) are of significant importance due to increased morbidity and mortality. We retrospectively analysed TBIs to assess clinical profile and factors predicting in-hospital mortality. Electronic database at a private, urban tertiary care centre was screened (2011 to 2016) to include all TBI -hospital mortality outcome were assessed. Mean age of population was 39.2 ±17.2 years with 81.8% being males. Most injuries (90%) resulted from road traffic accidents (RTAs). Bleeding from either ear, nose or throat (ENT) was most common presentation (29.2%) followed by vomiting (25.8%) and convulsions (7.5%). On presentation, severe brain d Subdural (36.7%) haematoma was most common followed by sub-arachnoid (25.8%), extradural (16.7%) and intracerebral (2.5%). Threatened airway was observed only in 7.5% cases. Bony trauma (25%) followed by face (14.2%) ,chest (3.3%), and spine injuries (0.8%) were associated injuries. In-hospital in-hospital mo 4.82, 50.88; p<0.0001), tachycardia (OR 3.0, 95% CI 1.21, 7.43; p=0.014) and with development of hyponatremia (OR 5.0, 95% CI 2.17, 11.53; p<0.0001) or fever (OR 3.18, 95% CI 1.29, 7.61; p=0.010) during hospitalization. Ventilator support was necessary in 29 (23%) cases out of which 16 (55%) died. Hospital stay (days) did not vary significantly in survivors and non-survivors (5.5±4.0 Vs 8.5±7.8, p=0.156). development of hyponatremia or fever and requirement of assisted ventilation were associated with -hospital mortality. Casey 1 We established a team that included attending physicians, fellows, advanced practice providers (APP), specific guidelines. We surveyed staff regarding team communication and discussion of QSIs during rounds. We designed a DGT that defined team member roles, structured communication patterns, and prompted standard discussion of QSIs. Following implementation, we evaluated team compliance with the DGT, as well as rates of catheter-associated urinary tract infections (CAUTI) and ventilatorassociated pneumonias (VAP). 11% respiratory therapists). For most QSIs, a minority of staff reported that these were always addressed during morning rounds (% staff indicating QSI "always" addressed: 14% pressure ulcers, 19% code status, 33% CAM ICU, 36% mobility goal, 46% central line catheter removal, 50% urinary catheter removal, 66% DVT prophylaxis plan, 69% ventilator weaning). Shared understanding of daily goals between nurses and physicians was reported by 58% of staff, with a significant difference between nurse 33%), p = 0.047. DGT audits spanning 212 patient days demonstrated median compliance greater than 80% for discussion of all QSIs. There was an 11% reduction in CAUTIs (95% CI 3%, 19%), p = 0.011, and a trend towards reduction in VAP that was not significant. Team discussion of QSIs can be enhanced by DGTs. By promoting adherence to evidence-based best practice, DGTs may reduce hospital-associated infections. Follow-up is ongoing to determine the impact of DGTs on clinical outcomes and team communication. Catheter-associated urinary tract infection (CAUTI) is the most common health-care associated infection accounting for > 560,000 nosocomial infections annually (Gould,2016) . According to the Center for Disease Control, CAUTIs are also a leading cause of secondary blood stream infection resulting in development related to cognitive, motor, and sensory deficits. NeuroICU's goal was to eliminate CAUTIs, defined as a rate of zero. With Initial efforts (2008 ( catheter days (Titsworth et al, 2012 . In February, 2016 NeuroICU launched a patient-centered quality improvement effort to further reduce CAUTIs. The NeuroICU Interdisciplinary Comprehensive Unit Safety Program (CUSP) performed an extensive literature review of evidence based best practices specific to urinary catheter management. A preexisting, staff nurse driven urinary catheter management protocol was revised to better fit the needs of the neurocritical care population, including assessment of and interventions for acute and chronic scan assessments to every 4 hrs, revision of urinary d-ofconducted daily rounds to evaluate the necessity and management of indwelling urinary catheters. After three months of implementation, compliance with use of the NeuroICU's Urinary Catheter Management Algorithm (UCMA) was greater than 95%, urinary catheter utilization was reduced from 65% to 38%, and CAUTI rates were reduced to zero. Implementing a neurocritical care patient-centered, interdisciplinary approach to urinary catheter management significantly impacted urinary catheter utilization, CAUTI rates, and unit culture. Dedicated neurointensivists have previously been shown to improve various outcome measurements in patient and family satisfaction. The purpose of this study is to evaluate the impact of newly appointed neurointensivists on quality outcome measures in a NSICU. This is an observational cohort study of adult patients (>18 years) in a 10-bed NSICU at an academic, tertiary care center evaluating quality outcome measures pre-and post-neurointensivists. Outcome measurements include catheter associated urinary tract infection (CAUTI), central line blood stream infection (CLABSI), ventilator associated pneumonia (VAP), patient acuity, mortality, and length of stay (LOS). Patient satisfaction questionnaires from discharged patients were compared to historical controls. tentiveness of doctors; 5) Recommend the hospital to others. Statistics include 1-sample binomial and N-1 Chi-squared (categorical) and t-test (continuous). For questionnaire data, 1considered significant. 2547 total patient days occurred pre-and 2560 days post-neurointensivist coverage. Patient acuity decreased 24.2% (p=0.043). CAUTI (66%, p=0.084), CLABSI (100%, p=0.32), central line days (16.5%, p<0.0001), ventilator days (18.7%, p=0.3), and VAP (200%, p=0.15) also decreased. These saved the hospital an estimated $65,000 based on Health Services Advisory Group Data. 89 questionnaires were returned. Patient satisf on physicians' attentiveness (p=0.03). Patients recommending the hospital to others increased 67% (p=0.03). Dedicated neurointensivists positively impact quality outcome metrics, particularly significantly improving patient satisfaction. Future studies should evaluate the direct impact of neurointensivists on Medicare reimbursement from improved patient satisfaction. Bertan Hallacoglu, Tanmayi T. Oruganti, Chandran C. Seshagiri. Research & Development, Boston, MA, USA. Cephalogics has developed a wearable Diffuse Optical Tomography (DOT) imaging device to help clinicians monitor perfusion and oxygenation from multiple brain regions on the bedside in disease states the system to changes in cerebral tissue oxygenation (SctO2) induced by hyperventilation in a pig and human subjects. DOT sensor was positioned on the pig's head along the sagittal line, and the second sensor was positioned on the hind leg muscle for monitoring systemic tissue oxygenation (SstO2). DOT measurements were performed continuously during baseline (8 mins, PaCO2 = 40mmHg), transient hypercapnia (4 mins, PaCO2 = 15mmHg), and recovery (25 min, PaCO2 = 40mmHg) periods controlled by the respiratory rate of the ventilator. DOT data were recorded to a laptop for off-line analysis. StO2 -SD) were computed for comparison of results across measurements. The approach was also investigated in three human volunteers, who were instructed to hyperventilate during DOT recordings. --6%), consistent with reports of reduced cerebral blood flow during hypocapnia. In contrast, SstO2 estimates -1%), indicating elevated systemic perfusion. Both parameters fully recovered to baseline values during the recovery period. SctO2 response to hyperventilation in human volunteers were consistent with the results in pig. Hypercapnia induced reduction in SctO2 was noninvasively imaged in human subjects and a pig despite the large scalp-cortex distance in pig. The results of this study demonstrate the sensitivity of the Cephalogics' DOT system to SctO2 values and its ability to separate SctO2 from systemic perfusion. Brittany Doyle, Michael M. Rogers, DaiWai D. Olson, Venkatesh V. Aiyagari. UT Southwestern, Depts. of Neurological Surgery, Neurology and Neurotherapeutic, Dallas, USA. Multidisciplinary rounds play a pivotal role in optimizing care in the Neurosciences Critical Care Unit (NCCU). Care providers were frustrated with inconsistency and the need for manual data entry to conduct multidisciplinary rounds. The purpose of this project was to develop an integrated computerized form that incorporated "smart" features within EPICTM. The NCCU Nursing Council collaborated with the physicians to design a standardized system-based approach to multidisciplinary rounds, and the elements that would be addressed within each system. Input was collated from nursing, neurology, neurosurgery, pharmacy, and critical care into a paper version which was beta tested before the EPIC tool was officially rolled out. The rounds template is now a perpetually editable note with fields that include drop-down menus, copy--populate with up-to-date data (e.g., ICP values, lab values). While there was a learning curve to the use and un -led rounds became much more efficient, comprehensive, and less frustrating for the entire healthcare team. Within four months, nursing management saw the benefit of such a standardized tool for clear communication and the tool also became the standard for nurse-to-nurse handover at change of shift. Efficient and orderly presentation of information during multidisciplinary rounds is greatly assisted by the use of a standardized electronic tool. Having all of the current, relevant data available in a single location has greatly improved the quality of nurse-led Neurocritical Care rounds. This tool can be replicated and customized to the needs and patient populations of other units and hospitals. The degree of burst suppression on continuous critical care EEG (CCEEG) monitoring is used to guide dosing of intravenous anesthetic drugs (IVADs) in the treatment of refractory elevated intracranial pressure (ICP) and refractory status epilepticus (SE). However, medication titration is performed only as frequently as CCEEG review (potentially as little as 2-3 times a day). Quantitative EEG (QEEG) may provide a continuous, objective assessment of the level of burst suppression that would allow for more precise and rapid titration of IVADs. compared software-generated QEEG burst suppression ratio (QEEG BSR) with three raw EEG variables as determined by two board-certified neurophysiologists (R1 and R2): bursts per minute (BPM), total burst duration per minute (BD) and average inter-burst interval (IBI) for 10 one minute segments per patient. A total of 90 EEG segments were analyzed. Agreement between readers was very high for the three raw EEG variables: BPM, BD and IBI (correlation coefficient 0.92, 0.96, and 0.95, respectively). The best correlation was observed between BD and QEEG BSR (-0.93 for both R1 and R2). The correlation between BPM and QEEG BSR (-0.80 R1 and -0.75 R2) and IBI and QEEG BSR (0.70 R1 and 0.65 R2) was not as strong. Left and right hemispheric QEEG BSR did not differ statistically from the generalized QEEG BSR (p=0.11 and 0.08, respectively) despite the presence of focal intracranial pathology. The depth of therapeutic burst suppression can be accurately assessed by generalized QEEG BSR. Although CCEEG bursts per minute is the most commonly used CCEEG metric of burst suppression, it shows a lesser correlation with QEEG BSR than total burst duration per minute. There is little operational data on optimal Neuro-ICU physician staffing. This prospective study evaluated the introduction of night-time in-hospital neurocritical care fellows (NCFs) in an urban academic hospital. The goal was to determine if the new staffing model enhances patient care and provider and patient satisfaction. IRB approval was obtained. The new staffing model was rolled out on 9-1-2015. 121 providers (NCFs; neurology residents and attendings; neurosurgery residents, physician assistants, advanced practice nurses and attendings; Neuro-ICU nurses) were surveyed. A pre-roll out survey of providers' perceptions of the existing model (night-time NCF coverage from home with on-call in-hospital neurology residents and neurosurgery physician assistants or residents, not dedicated to the Neuro-ICU) was administered prior to new model roll-out. Two follow-up surveys, a separate night-shift nurses survey, and patient (or surrogate) surveys were administered between 9-1-15 and 5-10-16. Surveys were electronic and responses were voluntary and anonymous except for the night-shift nurses survey which was paper, required, and anonymous. Response rates were 52% (pre-roll out), 28% (first follow-up), and 25% (second follow-up). 93% of providers indicated pre-roll out that night-time in-hospital NCFs would be beneficial; 94% (first follow-up) and 92% (second follow-up) indicated satisfaction with the new staffing model. Major reasons for satisfaction included: physician response to emergencies, physician -nurse communication, and patient outcome. The reason for dissatisfaction was decreased educational opportunity for residents. 94% of night-shift nurses reported preference for the new model. 62% of 72 patients (or surrogates) agreed to participate: 87% reported satisfaction with the care they received in the ICU; dissatisfaction was not More than 92% of providers were satisfied with a new staffing model featuring night-time in-hospital NCFs. Additional studies are necessary to determine optimal Neuro-ICU physician staffing with increasing patient and treatment complexity and decreasing resources. Therapeutic plasma exchange (TPE) is a first-line therapy for Guillain-Barre syndrome, myasthenia gravis, chronic inflammatory demyelinating polyneuropathy and polyneuropathy associated with paraproteinemias. TPE is also a second-line therapy for neuromyelitis optica, chronic focal encephalitis, and acute multiple sclerosis. The therapy is also used in autoimmune encephalopathies, refractory status care service has provided a neurointensivist run TPE program using membrane-based technology with the Gambro® Prismaflex System. A benefit of a neurointensivist managed membrane -based service is lower direct costs compared to centrifugal-based therapies. Since a consultation to another service (often available during business hours) is avoided, delays in therapy can be reduced (potentially further decreasing costs) and the therapy is also available for emerging indications such as refractory status epilepticus. This review is of the initial 15 patients treated with TPE in our Neurocritical Care Unit between April 2015 and May 2016. All patients were treated with 5% albumin with a targeted exchange of 1.2 to 1.5 plasma volumes. Overall dosing, timing, indications as well complication were reviewed. Indications for therapy included Guillain-Barre syndrome, Myasthenia Gravis, autoimmune encephalitis, refractory status epilepticus, autoimmune vasculitis, and parainfectious transverse myelitis. All patients received a minimum prescription of 1.2 plasma volumes and had the therapy started sooner than historical controls. No changes in nurse staffing were required for the therapies and we report no bleeding, infectious or access related complications. Mild coagulopathy was seen in most patients after previously not available. A neuro-intensivist lead TPE program is feasible and safe. The therapy can be delivered more timely and can be offered for a wider variety of indications. A cost analysis of the program is now underway. Specialty neurocritical care improves hospital length of stay and mortality in patients with critical neurologic illness. However, clinical practices are often informed by studies focused on mortality or gross functional outcome. Both staff and family members face uncertainty about neurological outcome, which performing detailed followup for all patients admitted to the University of Cincinnati Neurosciences Intensive Care Unit (NSICU). All patients admitted during a twotrained Outcome Scale-Extended (GOSE), cognitive outcome using the Teleph and quality-of-life using the Euro-QOL. We assessed the duration of each follow-up call in order to determine feasibility. n=50 patients were admitted to the NSICU; mean age was 59; 63% were male. Overall follow-up was obtained in 80%; 64% of survivors were contacted. Disposition from NSICU included: home (20%), acute rehab (27%), long-term care (24%), hospice (4%). In-hospital mortality was 16%. Median GOSE (IQR) was 4 (3-7) and median mRS (IQR) was 3 (1--30. An average of 13:28 (mm:ss) was required for each assessment; comprehensive assessments required 21:54. Overall, a total time of approximately 11 hours was required for 50 patients. Our pilot study demonstrated feasibility of following patients admitted to the NSICU. These findings have the potential to guide in-hospital care and out-of-hospital resources when used as a quality improvement metric, and to provide valuable information for retrospective research. Our Neurocritical Care Unit restructuring provided an opportunity to revise our staffing model. The prior model ratio of 1:2 resulted in at least one nurse experiencing a 1:3 ratio when high acuity patients -specific data to support a higher staffing. Prospective observational study of nursing time using the 28-item Therapeutic Intervention Scoring System (TISS-28) and manual timing of discrete nursing interventions including off-unit transport. Baseline data was obtained to include measures of nursing experience. Measures of central tendency and regression analysis was performed using SAS v9.4. The average time to complete a neurologic assessment was 4.17 minutes; the average time to chart a neurologic assessment was 4.05 minutes. For time spent off the unit the average time spent traveling to CT was <20 minutes, average time spent in MRI was 80 minutes twice a day, average time in IR was 82.5 minutes. We found that nursing experience was not associated with patient acuity, duration of assessment, nor time spent doc ents. Assessment time was associated with documentation time. Moreover, higher acuity was a predictor of both assessment time and documentation time. The data were used to support an increased acuity model with 3 nurses having planned 1:1 ratio. Neurologic critical are patients require more hands on nursing care and time spent in diagnostic testing than an average critical care patient. Neurocritical Care Units staffing ratios should be adjusted to provide time to ensure nursing care is complete. Melissa Panter, Sonja S. Stutzman, DaiWai D. Olson, Venkatesh V. Aiyagari. UTSouthwestern/Neurocritical Care, Dallas, USA. Venous access is a basic yet critical component of care. Determining which venous access devices (VAD) to use depends on the type, duration, and frequency of infusion. Prolonged continuous infusion of drugs or fluids, benefit from midline peripheral catheter (midline) or a peripherally inserted central catheters (PICC). Midlines are safe and effective but use is declining in favor of PICCs, which have similar insertion costs and added benefits, such as the delivery of toxic drugs harmful to peripheral veins. However, an infection associated with a PICC is classified as a central line associated blood stream infection which has important financial implications for the hospital. Therefore, it is important that the type of VAD be specifically tailored based on -physician collaborative study to design and test a decision support tool to assist clinicians in determining the appropriateness of the PICCs vs midlines for patients. This study has three phases. First, a retrospective chart review of VAD decisionexamine the current degree of appropriateness for each type of VAD for each specific patient scenario. Next, we will design a decision support tool to help decide which VAD should be used. The tool will be developed based on expert clinician, pharmacist review and a review of the literature. Finally, we will their patients. The first phase of the study is ongoing and we will present the results of the retrospective part of the study and the decision support tool at the annual Neurocritical Care Society meeting. A decision support tool to guide clinicians choosing between different types of VAD will help improve current clinical practice and patient outcomes. Clinical trials in neurocritical care require a predictable set of baseline, monitoring, treatment, and clinical outcomes data. However, interoperability standards restrict automated real-time streaming of this data, resulting in inefficiencies performing clinical trials, preventing real-time clinical trial oversight and constraining collaborative research. We investigated available data systems and developed a conceptual e. Examining current traumatic brain injury interventional trials, we considered technical interoperability -based clinical trial oversight and collaborative analytic research. We elaborated a vendor-neutral interoperability schema for data extraction, repositories, analysis, annotation, and visualization. The proposed conceptual solution is described. Nodes of data acquisition include: 1) continuously streaming devices including physiologic monitors and infusion pumps; 2) discrete data from highpenetrance EHR and laboratory platforms; and 3) biospecimen, radiology, and clinical outcomes repositories. An application program interface performs function calls to utilize individual episodes of data. A data management system queries and manages multiple patient records for batch processing of ondemand or prefor real-time or post-hoc assessment of raw and derived parameters (e.g., percent time in target range or on-protocol compliance). We enumerate the variety of current nodes requiring interoperability interfaces, and propose an open standard to promote a highly efficient platform for n -based clinical research, featuring automated case report form data extraction, a programmable interface for oversight and early warning detection, and a platform for annotation and crowdsourcing of novel algorithms. This conceptual architecture for a modular, vendor-neutral, data collection and management system for -based clinical trials in neurocritical care and offer new functionality for real-time oversight and collaborative analytics. Improving Cancer is a devastating illness; with a rise in brain and spine tumors specialized care is more important that ever. With advances in technology and treatment strategy, those too ill to previously receive care reased acuity has translated to the need for higher levels of hospital care. Recognizing the unique expertise required to care for this emerging population, the decision was made to merge neurology and oncology specialties within critical care. The purpose of this poster is to describe the admission criteria, patient population, utilized technology, staffing model, and patient outcomes of a newly created neuro-oncology critical care unit (NCCU). In January 2014, a 24 bed neurocritical care unit was opened in a midwestern academic medical center. Six of the beds housed within this 24 bed unit were designated and budgeted to the oncology medical center. Separate nursing staff and management teams were created to support the care within the unit while maintaining close collaboration with university hospital NCCU nursing teams. While nursing teams are distinct, nurse practitioners, neurointensivist, and other multidisciplinary team members are shared between both services. To evaluate the benefit of patient diagnosis, care needs, NCCU length of stay (LOS), number of ventilator acquired pneumonias (VAPs), urinary tract infections (UTIs), and central line acquired bloodstream infections (CLABSIs). At the time of this submission, specific results are still being tabulated. Merging access to academic research trials and oncologic and neurointensive specialists, this has created an environment that promotes care reflective of the most up to date evidence based practice. Through the utilization of creative staffing and focused onboarding, this unit has been able to treat this subspecialized population holistically utilizing a multidisciplinary approach to minimize hospital acquired complications and LOS. In 2014, the Neurocritical Care Society (NCS) published a consensus statement in support of multimodal neuromonitoring, emphasizing the essential need for systems to integrate data in meaningful ways to t barrier to the integration of data from multiple Group on Neurocritical Care Informatics was established in 2015; it includes experts from the clinical and research side of neurocritical care and representatives from medical device manufacturers. Through an open meeting and continued discussion, a recommendation document "Medical Device Connectivity" was produced. Its objective is to provide guidance to medical device manufacturers desiring to design a communication protocol that allows external systems to acquire data from their devices. Our research on this project uncovered a high percentage of errors in medical device protocols as well as undocumented characteristics and safety issues. The group developed recommendations for providing robust communications as well as address usability and safety concerns. The document outlines the content that should be transmitted which includes device identification, protocol version identification, patient identifier, events, alarm conditions, system status, data labels, and units. Finally, it must be thoroughly documented and validated by the manufacturer. Multimodal monitoring (which requires medical device connectivity) reportedly improves quality of care through reduction of errors and increased detection of adverse events. However, widely adopted connectivity are on the horizon. In the meantime, the document created in this project will provide guidance for manufacturers in their communications protocol development. In doing so, they will then further the NCS's recommendations on device connectivity for multimodal monitoring. Warfarin-related intracranial hemorrhage (ICH) is associated with increased mortality due to higher rates of hematoma expansion. Current guidelines recommend rapid anticoagulation reversal using intravenous vitamin K and prothrombin complex concentrate (PCC). Previous studies show high prevalence of incomplete anticoagulation reversal with 3-factor PCC. We therefore sought to assess the impact of PCC type in reversing warfarin in ICH patients. This was a retrospective study of 89 ICH patients (47 traumatic and 42 non-traumatic) with warfarinassociated coagulopathy who were admitted to a level II trauma center between January 2010 and September 2015 and received at least one dose of 3-factor or 4-factor PCC. Post-PCC INR of £1.2 was considered successful INR reversal. Multivariable model using logistic regression was performed to assess the impact of PCC type on successful INR reversal after adjusting for age, sex, BMI, and baseline INR. Overall, the prevalence of successful INR reversal was 72.2%. There was a higher proportion of patients with successful INR reversal in those who received 4-factor PCC than 3-factor PCC (53.8% vs. 18.4%, respectively, p=0.005). In the multivariable model, 4-factor PCC (OR 18.1; 95% CI: 3.1 to 106.5) and baseline INR (OR per unit of INR 0.4; 95% CI: 0.19-0.84) were independent predictors of successful INR reversal. The change in INR post-PCC was significantly greater in those who received 4-factor PCC than 3-factor PCC (3.60 ± 4.16 vs. 1.26 ± 1.11 respectively, p<0.0001). 4-factor PCC more reliably reversed warfarin in ICH patients compared to 3-factor PCC. Cortical spreading depolarizations (CSDs) as highly active metabolic event commonly occur in patients with intracerebral hemorrhage (ICH) and may contribute to secondary brain injury. Fever is an independent predictor for unfavorable outcome after ICH and may trigger CSDs. Here, we investigated the dynamics of brain-temperature (Tbrain) relative to CSDs and core-temperature (Tcore). Twenty comatose patients with ICH and multimodal electrocorticograpy (EcoG) monitoring were prospectively enrolled. A subdural EcoG strip was placed adjacent to the evacuated ICH. A combined intracranial pressure (ICP) and Tbrain probe was inserted in the white matter ipsilateral to the ICH. Monitoring data were averaged to 5-minute-means for longitudinal analysis and to one-hour-means. -burden was defined as % of temperature >38.0° C per 24-hours. Data were analyzed using GEE-models and are presented as median and interquartile range (IQR). During 3097 hours (173 hours [81-CSDs occurred in clusters. Baseline Tcore and Tbrain were 37.3°C (36.9-37.8) and 37.4°C (36.7-37.9), respectively. Tbrain but not Tcore significantly increased 5 minutes preceding the CSDs by a median of 0.2°C (0.1--0.4]; p<0.001) but not Tcore (p=0.34) was higher during clusters compared to episodes of single CSDs. CSDs probability was highes OR=1.2 per %; Tcore: p<0.001; OR=1.1 per %) independent of MAP and ICP. CSDs were triggered during episodes of fever. Our data suggest an association between CSDs and cerebral heat production, especially during clusters. Integration of ECoG monitoring in trials investigating prophylactic normothermia after ICH may help to understand the potential beneficial effect of this intervention. Anticoagulation reversal is recommended for patients with intracerebral hemorrhage (ICH) on vitamin K antagonists. We propose a window for reversal of Vitamin K antagonists, in which ICH volume remains below the average growth in a control population on follow up imaging A retrospective review of Neuro ICU patients at Henry Ford from 2013-2014 was conducted on patients with the ICD9 code for diagnosis of ICH. Inclusion criteria: Brought to ED at onset of symptoms, either not on anticoagulation or were on warfarin with therapeutic INR, and had received PCC administration. Fifty eight total patients were identified. Patients were approximately matched for GCS on admission, ICH w -1cc difference in volume estimation. Forty-seven control patients were identified: Mean age 67 (44-92), GCS on admission 11 (3-15), SBP 170, ICH volume on admission 20.28cc, ICH volume on repeat scan 23.94cc, change in ICH volume 21.45%, and mean time between initial and stability scans was 1150 minutes. Eleven patients on VKAs were identified who underwent reversal with 4-factor PCC. Mean age 74 (58-88), GCS on admission 11 (3-15), SBP 149, ICH volume on admission 31.81cc, ICH volume on repeat scan 42.42cc, and time between scans was 630 minutes. Average INR on presentation was 4.71. All patients on warfarin Patients who had PCC administered before 195 minutes (7) had a mean change in ICH volume of 16.95%. Patients who had reversal completed after the 195 minutes (4) had a mean change in ICH volume of 60.28%. (P value= 0.0032). We propose a potential "recommended reversal time" of less than 200 minutes for Vitamin K antagonists in our institution. Limitations of study include small sample size. The Full Outline of UnResponsiveness (FOUR Score) is a validated scale that provides the essentials of a coma examination by incorporating motor response, eye opening and eye movements, brainstem reflexes, and respiratory pattern. We incorporated the FOUR Score into the existing ICH Score and Consecutive patients admitted to our institution from 2009-2012 with spontaneous ICH were reviewed. Using patient age, hemorrhage location, hemorrhage volume, evidence of intraventricular extension and GCS, the ICH Score was calculated. The FOUR Score was then incorporated into the ICH Score as a substitution for the GCS (ICH-FOUR Score). The ability of the two scores to predict mortality at 1 month was then compared. In total, 274 patients met inclusion criteria. The median age was 73 years (IQR 60-82) and 138 (50.4%) were male. Overall mortality at one month was 28.8% (n=79). The area under the ROC curve was 0.6 (95% CI 1.11-2.08) for the ICH Score, and 0.6 (95% CI 1.02-1.86) for the ICH-FOUR Score. For ICH Scores of 2, 3, 4 and 5, one-month mortality was 18.8%, 29.0%, 44.7% and 66.7%. In the ICH-FOUR Score model, mortality was 20.6%, 27.8%, 39.6% and 75.0% for scores of 2, 3, 4 and 5, respectively. The ICH Score and the ICH-FOUR Score predict 1-month mortality with comparable accuracy. As the FOUR Score provides additional clinical information regarding patient status, it may be a reasonable substitute for the GCS into the ICH Score. Depressive symptoms in patients with intracerebral hemorrhage (ICH) are common and are associated with worse outcomes. It is not well described how often depressive symptoms are appropriately ascertained and treated in a multicenter cohort, and whether this is a potential target for improving outcomes. We retrieved diagnostic codes from four university health systems across Chicago (multicenter cohort). Separately, we prospectively screened for depressive symptoms (NIH Patient Reported Outcomes Measurement Information System, PROMIS, T Score 60), in patients at one prospectively assessed cohort center at one, three and twelve months after ICH onset. We compared detection rates of depressive symptoms between the two samples. Diagnostic codes for depressive symptoms up to three months after ICH onset were recorded in 132 of 3422 (3.8%) of the multicenter cohort versus 26 of 116 (22.4%) in the prospectively screened cohort (OR 7.20, 95% CI 4.5 -11.5, p < 0.0001). Results were similar considering depressive symptoms up to 12 months after ICH, 140 of 3422 patients in the multicenter (4.0%) versus 36 of 116 prospective patients (31.0%; OR 10.55, 95% CI 6.87 -16.2, p < 0.0001). In the multicenter cohort less than 10% of patients months of ICH onset. The prevalence of depressive symptoms in survivors of ICH is more common than would be suggested treated and SSRIs may be an under-utilized therapeutic option. Vitamin K antagonist (VKA)are associated with higher mortality than primary ICH. Prompt reversal of international normalized ratio (INR) with prothrombin complex concentrate (PCC) may promote hemostasis and decrease hematoma impact of an electronic order set designed to standardize and facilitate more timely reversal of coagulopathy in VKA-associated ICH. We identified all adults that received PCC for VKA-associated ICH from June 2012 to March 2015 at UCSF Medical Center, which included a period before and after an electronic order set became available in 2014. We abstracted baseline demographics and clinical data from electronic medical records. The primary outcome was time from radiographic identification of ICH to administration of PCC. Secondary outcomes included PCC dosing accuracy based on actual weight and baseline INR as well as time from PCC order to follow-up INR. We identified 31 patients that received PCC for VKA-associated ICH, including 17 patients before and 14 patients after the order set became available. Baseline demographics and clinical features were similar. Order set use was associated with a significant decrease in the time from identification of ICH on imaging to the administration of PCC (median 45 vs. 83 minutes; p=0.02), more accurate doses delivered (92.9% vs 29.4%; p<0.01), and a shorter time from the PCC order to follow-up INR (median 85 vs 164 minutes, -hospital mortality were similar in the two groups. An electronic order set for administering PCC for VKA-associated ICH was associated with significantly faster time to PCC administration and increased accuracy in dose administered. Andrew Naidech, Alan A. Long, Kathryn K. Muldoon, Rajbeer R. Sangha. Northwestern Medicine, Chicago, IL, USA. crucial to identify patients. Both active contrast extravasation ("spot sign") and lower platelet activity have been associated with hematoma growth. We tested the hypothesis that patients with a spot sign had lower platelet activity. We prospectively identified patients with acute ICH, measured platelet activity on admission, routinely obtained CT angiography and graded the presence or absence of a spot sign. We limited the analysis to patients who underwent CT angiography within 12 hours of ICH symptom onset. Platelet activity was measured with the VerifyNow-ASA (Accumetrics, CA). Non-normally distributed data were compared with the Mann-Whitney test, and binary variables with chi-squared or logistic regression. -14.2 years, 51% were women and 73% had a history of hypertension. Sixteen (9.6%) had a positive spot sign. A spot sign was associated with lower platelet activity (479 [433 -532] vs. 563 [456 -637] ARU, P=0.03, where <=550 ARU indicates an aspirin effect). Of 16 patients with a spot sign, 13 had platelet activity <=550 ARU. Platelet activity <=550 ARU was associated with increased odds of a spot sign (OR 5.7, 95% CI 1.55 -20.7, P=0.009). The presence of a spot sign on CT angiography was associated with lower platelet activity, suggesting a hematoma growth. Introduction: Clot dissolution is a biochemical process catalyzed by enzymatic proteins, requiring a specific temperature range for optimal function. CLEAR III, a randomized, double-blinded, placebocontrolled trial tested whether extraventricular drainage (EVD) plus intraventricular alteplase improved outcome by removing IVH compared to EVD plus saline. Methods: Retrospective assessment of prospectively collected temperature data q4H over first 7 days post randomization. Blinded assessment o functional outcome (mRS4-6). Results: Median (interquartile range) daily temperature was 37.5 (37.4, 37.6)°C and did not differ between -6 (vs. 0at 30 days [37.42 (37.11, 37.72) vs. 37.20 (36.98, 37 .43), p37.5°C was significantly associated with faster IVH clot lysis rate (Spearmans rho 0.12; p=0.006), but despite significantly higher temperatures, patients with mRS4-6 (vs. mRS0-3) had significantly lower average percentage IVH removal at day 30 (51.23±45.77%vs. 63.06±28.72%, p<0.001) and 180 (48.82±51.47% vs. 58.34±31.02%, p=0.008). Temperature functional outcome of 1.91 (95%CI:1.45-2.52) and 1.54 (95%CI:1.26-1.90) at 30 and 180 days respectively, and was an independent predictor of poor functional outcome at day 30, but not 180 after adjustment for early infection and other severity predictors. associated with faster clot lysis rate, but also with poor functional outcome. Negative associations with Of 162 ICH patients, 26 patients (16%) had WAICH and received Kcentra. The INR ranges were: 1.4-1.9 in 5 patients (19%), 2.0-3.9 in 17 (65%), 4.0-6.0 in 1 (4%) and >6.0 in 3 patients (12%). Within 30 minutes of -minute repeat INR ranges were 1.4-1.5 and 1.0-1.1, respectively. FFP was administered to 1 (20%), 3 (18%), 0 (0%) and 1 (33%) patient in each group for a persistently elevated INR. At 6h post-PCC, INR reversal occurred in 100%, 69% (rest had INR 1.4), 100%, and 67%, respectively. At 24h and one had DIC following PCC administration. In this small case series of WAICH patients, INR correction with Kcentra was adequate except for those with INR 1.4- Prevalence of stroke following craniotomy or craniectomy for spontaneously intracranial hemorrhage Intensive care resources, particularly, neuroscience intensive care resources are limited and costly. In most institutions in the country, all intracerebral hemorrhage (ICH) patients are admitted to the Neuroscience intensive care unit. We sought to identify what criteria will allow us to determine which primary intracerebral hemorrhage patients will not need admission to an intensive care unit (ICU). We studied retrospectively 423 patients with primary ICH from January 2013 to the end of Dec 2015. We reviewed multiple admitting characteristics: demographics, hematoma volume, location of hemorrhage, any brain compression, blood pressure, respiratory status, INR, Glasgow Coma Score (GCS). The reasons for neuroscience intensive care unit admission requirements are the need for mechanical ventilation, hydrocephalus, increased intracranial pressure, Low GCS score, hematoma expansion, or the unit, and who did not require any ICU care intervention. This group had the following characteristics: Supratentorial ICH, ICH volume 12. We called this criteria the "Non-admission Criteria for Patients with Primary Intracerebral Hemorrhage". 423 patients were identified as primary ICH. 69 patients (16.3%) fulfilled the non-admission criteria to the neuro-intensive care unit. Of 69 patients 27 patients (39.1%) discharged home, 31 patients to acute rehab facility (44.9%), 10 patients discharged to SNF (14.5%), 1 patient died (1.5%); elderly patient with existing DNR comfort care orders. None of the 69 patients had to be readmitted to neuroscience intensive care unit, and none required neurosurgical procedure. We propose that ICH patients that fulfill the non-admission criteria do not have to be admitted to an ICU and can safely be monitored in a step down unit This represented about 16% of ICH patients at our tertiary academic medical center. Future prospective studies are required to validate the criteria. Joseph R. Blunck, Justin J. Shewmaker. Saint Lukes Hospital / Pharmacy, Kansas City, MO, USA. Current guidelines recommend the use of 4-factor prothrombin complex concentrate (PCC) rather than fresh frozen plasma for reversal of warfarin-related intracranial hemorrhage (WRICH). There is no consensus regarding an optimal PCC dosing strategy, but limited data suggest that PCC dosing should be based on weight and international normalized ratio (INR). In November of 2013, our health system implemented a WRICH reversal protocol with a 4-factor PCC dosing nomogram that utilized fixed-dose options of 2000, above or below 4. The purpose of this study is to evaluate the effectiveness and safety of this simplified 4-factor PCC dosing protocol. Patients given 4-factor PCC for WRICH reversal in our health system were retrospectively identified with billing codes. Chart review was completed to evaluate the primary endpoint of achieving a post treatment evidence of a venous thromboembolic event, time to goal -hospital mortality. We identified 82 WRICH patients from November 2013 through April 2016 that were reversed with our simplified 4-factor PCC dosing protocol. Seventy-seven (93.9%) patients achieved a post reversal goal -six (68.3%) patients received a PCC dose equivalent to rounded PI dosing and the mean (SD) dose difference between groups was 2321 (365) units vs. 2256 (528) units, protocol vs. PI dosing, respectively. One patient (1.2%) had a thromboembolic event that occurred 20 days post reversal. A simplified 4-factor PCC dosing protocol is a safe and effective strategy for WRICH reversal. -hospital, 1 year and 10 year mortality, and has been influences functional outcomes. We sought to evaluate the association between CCI and outcomes in patients with primary intracerebral hemorrhage (ICH). Patients admitted to our center with primary-ICH from 2008-2013 were included. Demographic and clinical data were collected. The primary outcome measures were the proportion of patients with discharge mRS (dmRS) of 4-6, death and poor discharge disposition (any disposition other than home or inpatient-rehabilitation). Crude and adjusted logistic regression were used to evaluate the association between CCI and outcomes. 383 patients were identified. There were 37 (9.7%) patients with a CCI of 0 or 1, 242 (63.2%) patients with a CCI of 2-5, and 104 (27.2%) with a CCI of 6 or greater. While the continuous CCI was not significantly associated with a dmRS of 4-6 (OR 1.08, 95% CI 0.99-1.19, p=0.09), it was associated with disposition. The odds of poor disposition increased 18% with each increase in CCI (OR 1.18, 95% CI 1.08-1.28, p=0.0003). The odds of death increases 12% with each point increase in CCI (OR 1.12, 95% CI 1.02-1.23, p=0.021). After adjusting for baseline ICH score, CCI remains significantly associated with poor disposition (OR 1.13, 95% CI 1.00-1.27, p=0.042), however the association between CCI and death was not statistically significant (OR 1.05, 95% CI 0.91-1.21, p=0.52). In contrast to previous studies, CCI was not associated with poor short-term functional outcome or inhospital mortality in ICH patients after adjusting for ICH score. However, it was significantly associated with poor discharge disposition. This suggests that cumulative comorbidities only predict disposition in ICH, because the ICH score strongly impacts poor functional outcome and in-hospital mortality. Accurate assessment of renal function remains a unique challenge in patients with intracerebral hemorrhage (ICH). Mathematical estimates of creatinine clearance (CrCl) routinely used are often inaccurate in this setting. Subsets of critically ill patients have been shown to exhibit a hyperdynamic response leading to an enhanced renal clearance. No studies exist evaluating the directly measured creatinine clearance of patients with ICH. This was a single-center prospective observational study of adult patients with ICH admitted to the NSICU between January 2015 and July 2015. Eight-hour urinary creatinine clearances were performed daily to directly measure CrCl until the patient no longer had a foley catheter or the patient left the NSICU. Urinary -Gault equation. Statistical significance was defined as p-value < 0.05. Thirty patients with ICH were enrolled in the study. The study sample was 60% male with a mean age of 70±13.7 years. The median admission ICH score was 3 (IQR 2 -4) with a mean ICH volume of 64±64.1 mL. The median admission GCS was 7.5 (IQR 5 -13) and median admission SOFA score was 4.5 (IQR 2 -Additionally, the mean urinary CrCl was significantly higher than the estimated CrCl each individual study Patients with ICH consistently experienced urinary CrCl greater than estimated CrCl predicted based on -Gault equation. As renally eliminated medications are routinely dosed based on mathematical estimates of renal function, further study is needed to optimize medication regimens in this patient population to prevent underexposure. The Cognitive reserve hypothesis suggests that variations in patient pre-morbid status such as education, occupation and brain morphology influence outcome. This has been extensively validated in patients with Alzheimer's disease. An important component of Cognitive reserve is brain morphology, which can be quantified with measures such as whole brain-or gray matter volume. This study examines use of novel measures of brain morphology to measure cognitive reserve in patients with SDH. 56 patients with CT-documented SDH were screened. We identified patients who had a clinically indicated post-morbid MRI, Telephone Interview for Cognitive Status (TICS) and Barthel Index (BI) at discharge and follow-up. MRI was used to measure several volumes such as whole brain-, gray matter-, left vs right caudate-, hippocampal and intraventricular CSF volume using a Freesurfer pipeline. Primary outcome measures were TICS at 3 and BI at 12 months Results: -up. Regional -213) (SD), ----9 at discharge and 3 -8.3 at follow-up (3--6.4) at 12 mo. We found a positive correlation between intraventricular CSF volume and BI at 12mo (r = 0.77, p=0.04). There was also a significant correlation between left caudate volume and BI at 12months. Brain morphology did not correlate with TICS outcome at 3 months. Quantitative imaging can be used to predict functional outcomes in patients with intracranial hemorrhages. As we continue enrollment we hope to generate meaningful no and brain tumors. Christian Hernandez, Vivien V. Lee, Bichun B. Ouyang, Torrey T. Birch. Rush University Medical Center/Department of Neurological Sciences, Chicago, IL, USA. It remains unclear which patients with intracerebral hemorrhage (ICH) benefit from surgical hematoma evacuation, and the patient factors associated with the decision to pursue surgical intervention are largely early surgical intervention for ICH and to investigate if an age bias exists at our institution. One hundred and twenty consecutive patients who were admitted to the Neurosciences Intensive Care Unit with primary ICH between April 2010 and January 2012 were retrospectively reviewed. Multivariate regression analysis was used to analyze if select patient factors were associated with the decision to pursue early surgical intervention. In total 120 patients were analyzed; 60 patients (50%) were female (mean age 65.6 ± 13.3), 104 patients (87%) had supratentorial ICH, and 17 patients (14%) underwent hematoma evacuation. In univariate analysis age, race, and gender were not significantly a intervention. The only factors associated with hematoma evacuation were GCS score and ICH volume. After controlling for significant variables, multivariate analysis showed that the only factor associated with surgical intervention was ICH volume (OR 24.74, p=0.0002) . Surgical intervention did not affect discharge disposition (p=0.31), but was associated with a longer length of stay (14 vs 6 days, p=0.001). In this analysis, ICH volume was the only predictor of hematoma evacuation in patients with primary ICH. Age and sex did not influence patient selection for surgical intervention. Surgical treatment did not affect patient disposition at discharge, but was associated with longer hospitalizations. Further investigation is needed to determine which patients with ICH benefit most from early hematoma evacuation. studied. We investigate factors associated with admission systolic blood pressure (SBP), including hemorrhagic transformation (HT) and discharge outcome. This is a retrospective study of consecutive AIS patients presenting from April 2014 to March 2015. Demographic and clinical data were collected. Admission SBP was divided into three tiers: 165. The primary outcome measure was in-hospital mortality. Seven Hundred Seventy Six patients were included (meanwere 287 patients with SBP>165 (40.0%), 251 with SBP 140-165 (32.3%), and 238 with SBP165 were ry of hypertension (87.1%; p165. Premorbid use of antihypertensives did not differ among the three groups. There were no differences in proportion of HT (15.7% vs. 18.7% vs.12.6%; p=0.1462). Although patients treated with IV-tPA were evenly distributed among tiers (14.4% vs. 13.6% vs. 15.7%), more patients with SBP165 had mRS 3-6 (p=0.0382). Compared to SBP165 was associated with lower odds of in-hospital mortality (OR 0.536, 95%CI=0.295-0.975, p=0.041). This remained significant after adjusting for age and NIHSS (OR 0.431, 95%CI 0.193-0.962, p=0.0399). Normal presenting SBP (<140) in patients with AIS was associated with worse discharge functionaloutcome and higher in-hospital mortality. These differences may be related to other associated medical conditions such as pre-existing heart failure. Further research is needed to define the ideal range to maintain SBP after AIS. Between April 2009 and February 2011, 237 patients with non-traumatic, non-aneurysmal and nonmalignant parenchymal cerebral ICB were identified. The associations of NIHSS at presentation (NIHSSp), NIHSS at 48 hours (NIHSSd2), size of ICB, comorbidities, and infection with LOS were investigated retrospectively. The mean age for the 237 patients was 68.7±13.8 years and 52% were male while most patients were white (68%). The most prevalent comorbidity was hypertension (83% NIHSSd2 is a useful measure of LOS and should be collected for patients with ICB. NIHSSd2 at upper teens present most challenges to discharge and should be the aggressive focus for discharge planners. Incidence of infection and ventilator need is high in this population, adding to the challenges. Although oral Factor Xa inhibitors (FXaI), used for the prevention and treatment of venous racranial hemorrhage (ICH) than warfarin in clinical studies, intracranial bleeding is still associated with high morbidity and mortality. Moreover, there are no specific guidelines for managing these bleeds other than empirical institution-based hemorrhage protocols. There is a need to understand the real-world management, outcomes, and resource utilization of FXaI-associated major bleeding in order to potentially improve morbidity and mortality in these patients. Five US medical centers participated in a retrospective study of patients admitted to the hospital with lifethreatening bleeding on or after January 2014 while on apixaban, rivaroxaban, or low molecular weight heparin (LMWH). Baseline characteristics, treatment patterns, outcomes, and resource utilization were assessed. ICH patients were reviewed and are reported specifically. This interim report includes 33 major bleed patients, including 8 ICH. The majority of ICH were hypertensive [7(88%)], 3(38%) were on concomitant anti-platelet with FXaI, 2(25%) were diabetic and 2 e ICH was spontaneous (57%) and trauma (43%). In the management of ICH bleeding, 50% of patients received clotting factors (e.g. PCC), whereas 25% received interventions (e.g. radiological embolization). Within 30 days of discharge, 5 (63%) died, 4 of which occurred during the admission. For the 3 patients who did not die, length of hospitalization was 3.5(3.1-47.0) days, with only 1 discharged patient restarting on an anticoagulant. This study provides a real-world picture of patients receiving FXaIs or LMWH, experiencing lifethreatening intracranial bleeds. Despite efforts to restore hemostasis, mortality remains high and substantial healthcare resources are expended. This highlights the need to develop specific strategies for managemen Prophylactic seizure medications are not recommended by guidelines for patients with intracerebral hemorrhage, yet are prescribed to nearly 40% of them. There are few data on disparities in their use. We tested the hypothesis that there are differences in the administration of seizure medication, specifically We electronically retrieved information from patients who were diagnosed with intracerebral hemorrhage from HealthLNK, a multi-center electronic repository in Chicago, IL, from 2006-2012 ("multicenter cohort"). From 2007 through 2015, we prospectively identified patients with intracerebral hemorrhage at one site ("prospective cohort"). There were 3,422 patients in the multicenter cohort from four sites. The use of levetiracetam varied with -Americans (OR 1.71, 95% CI 1.43 -2.05, P<0.0001). In the prospective cohort (N=450), hematoma location, older age, depressed consciousness, larger hematoma volume and no alcohol abuse were -Americans to receive levetiracetam (OR 1.9, 95% CI 1.25 -2.89, P=0.002). African-Americans were more location was independently associated with levetiracetam administration (P<0.00001) disparities in the use of levetiracetam and they are confounded by hematoma location, a datapoint not typically available in administrative datasets. Deviation from guidelines for the use of seizure medications is common and rational, and any impact on outcomes is uncertain Greater gains in late recovery for intracerebral hemorrhage patients with more debilitating initial injury. Multiple intracerebral hemorrhage (ICH) studies have examined differences between discharge and 3month outcome. However, few studies have examined late recovery specifically between 3-and 12month endpoints. The aim of this study was to identify potential factors predicting late recovery in ICH patients. Twenty-four patients diagnosed with primary ICH at Yale-New Haven Hospital were prospectively enrolled between July 1, 2014 and July 1, 2016. Outcomes were assessed using Barthel Index (BI) at discharge, 3 months, and 12 months. Repeated-measures regression analysis was conducted using age, admission Glasgow Coma Scale (GCS), ICH volume, intraventricular extension, and ICH location, to compare outcomes at discharge, 3 months, and 12 months. There was a significant improvement across time points (p = 0.005), with follow-up testing showing improvement between discharge and 3 months (p < 0.001), discharge and 12 months (p = 0.036), and 3 months and 12 months (p = 0.021). Regression testing resulted in a significant relationship of time (p = 0.002), time x GCS (p = 0.003), time x ICH volume (p < 0.001), time x ICH location (p < 0.001). Plots show increased late recovery (i.e., between 3 and 12 months) for patients with lower GCS scores, larger ICH volume, and deep ICH location. Patients with more devastating initial ICHs show greater gains in late recovery between 3 and 12 months. These results suggest initial disability at 3 months may not represent overall recovery and support continual follow-up out to one year. While extensive studies have examined the outcomes of intracerebral hemorrhage (ICH) patients under the age of 80, few studies have examined outcomes in an elderly cohort (>80 years). The aim of this case-control study was to determine the independent effects of age on outcome after ICH. Nineteen ICH patients prospectively identified at Yale-New Haven Hospital above the age of 80 were matched against 19 patients below age 80, based on ICH location (lobar, deep, cerebellar, brainstem), ICH volume (> 30cc), presence of intraventricular hemorrhage (IVH), and admission GCS (0-4, 5-12, 13-15). The matched groups were compared via univariate analysis to examine differences in morbidity While there was no difference in pre-ICH disability (mRS > 1; p > 0.05), at 3 months elderly patients exhibited higher morbidity (mRS > 3, p = 0.023). Despite these differences, there were no differences in overall mortality between groups at discharge or at 3 months and no differences between Barthel Index at 3 months. Despite greater morbidity as defined by mRS, elderly patients with ICH do not significantly differ from younger patients in mortality rates or ability to perform activities of daily living at discharge and 3 months. These results may warrant further studies to provide more accurate prognostication after ICH in elderly populations. Hypertension after intracerebral hemorrhage (ICH) is associated with hematoma expansion, morbidity and mortality. There are currently no recommendations to standardize the critical care approach to acute blood pressure (BP) management in ICH. We performed a large retrospective cohort study to examine practice variability in BP management in acute ICH. care center using local Get With The Guidelines data. We abstracted time-stamped clinical variables including all documented BP measurements, and medications administered, as well as hematoma location from the electronic medical record. All acute ICH patients are admitted to the neuroICU and comanaged by neuro-intensivists and vascular neurologists. We used descriptive statistics to summarize overall population and treatment characteristics. -14) years, and 47% were female. and 1% other. Electronic medical record data were available in 497 subjects, of which 303 (61%) received continuous infusion for BP control within 72 hours of admission (56% nicardipine, 9% clevidipine). A systolic BP goal was charted in 283 (57%) and ranged from 120 to 220mmHg. These goals were modified in 72% during the same admission (64% increased, 8% decreased). Overall, 48 unique oral antihypertensives were administered (31% received a beta-an ACE inhibitor, and 11% a diuretic). Hypertension requiring continuous antihypertensive infusion is common after ICH. There is high practice variability in BP targets and choice of antihypertensive medications. A prospective study of a systematic and protocolized approach to antihypertensive medication use in ICH is necessary to determine if reducing practice variability improves outcomes. Intracranial hemorrhage (ICH) has long been thought to be a devastating consequence in the setting of end-stage liver disease. Due to its association with abnormal coagulation, the prognosis is thought to be poor, and frequently leads to withdrawal of care. Our aim with this study was to assess the true overall mortality rate of ICH in end-stage liver disease and compare it to mortality of ICH in the general population. All patients > 18 years of age admitted to Mayo Clinic from 2005 to 2015 with a subsequent diagnosis of non-traumatic ICH and end-stage liver disease were identified. Patients presenting with primary epidural, subdural, intraventricular, or subarachnoid hemorrhage were excluded. Using actuarial methods, 7 day, 30 day, and 1 year mortality rates were calculated. 112 patients with simultaneous diagnosis of ICH and end-stage liver disease were identified. 46 of the patients were female (41%) and 66 patients were male (58%). The mean age at diagnosis was 62 years. The mean systolic blood pressure in the mortality cohort was 162 mmHg, compared to 160 mmHg in the survival cohort. The 7 day, 30 day, and 1 year mortality rates were 23%, 32%, and 49% respectively. ICH associated mortality in end-stage liver disease does not significantly differ when compared to ICH mortality in the general population. The mean systolic blood pressure at presentation did not specifically correlate with an increased incidence of ICH associated mortality. These findings are important in the practice of neurocritical care, confirming that poor prognostication and expedited withdrawal of care should be reconsidered. Future directions will include mortality adjusted by hemorrhage location, size, and presence of intraventricular extension. Kcentra® (human prothrombin complex concentrate) is the first FDA-approved non-activated 4-factor prothrombin complex concentrate for the urgent reversal of vitamin K antagonist agents in adults with acute major bleeding. The recommended dosing is based upon the units of Factor IX, which can vary within each vial (range from 20vial size to reduce waste and some based upon the exact Factor IX content. This variation in dosing may be associated with complications in care and has never been evaluated. Underdosing of the medication can lead to suboptimal response and overdosing the medication can cause thromboembolic events. The purpose of this observational trial is to assess the current prescribing practices of Kcentra® in neurocritical care unit patients across the United States and evaluate the impact on patient response and safety. Additionally we plan to characterize current approaches to repeat dosing. This was a retrospective observational study of adult patients across 39 centers who received Kcentra for reversal of warfarin-related bleeding between January 1, 2014 and December 31, 2015. Descriptive statistics and tests for comparison will be utilized to evaluate differences in dosing, outcome, and the occurrence of adverse events. Statistical significance will be defined as p-value < 0.05. Data collection is ongoing but the full results will be presented during the meeting. Result of this study will document the real-world use of Kcentra® in patients with severe life threatening bleeding and assess the impact of variations in prescribing practices on patient responses. There is a paucity of data regarding this topic and combining data from multiple neurocritical care units will be timely in identify optimal dosing strategies. Perihematomal edema (PHE) associated with intracerebral hemorrhage (ICH) has been suggested to have an impact on both mortality and functional outcomes in spontaneous, supratentorial ICH. There have been no studies examining the impact of PHE in infratentorial hemorrhage. The aim of this study was to evaluate the impact of absolute PHE volume as well as PHE expansion rate in cerebellar hemorrhages at the time of discharge and at 3 months. 9 patients diagnosed with primary cerebellar ICH at Yale New Haven Hospital were prospectively enrolled between July 1, 2014 and July 1, 2016. Patients were evaluated using mRS and BI at discharge and 3 months. ICH and edema volumes on CT were measured using a semi-automated threshold based approach and PHE expansion rate was the difference between initial and follow-up PHE volumes divided Scale (GCS), ICH volume, intraventricular (IVH) volume, edema (PHE), and PHE expansion rate from baseline to first follow up CT scan before decompression (6h window). At discharge, patients with higher morbidity (mRS > 3) exhibited higher ICH volume (p = 0.012) and PHE volume (p = 0.005) on admission CT and a trend for greater rates of PHE expansion (p = 0.086). These differences were not significant at 3 months and when adjusted for ICH volume. Plots suggest an association between higher PHE rate and lower BI scores at 3 months. PHE may have a clinically significant impact in patients with cerebellar ICH at time of discharge, but these results are limited by a small sample size. These results warrant further studies and suggest therapies to ameliorate edema may be a treatment option for cerebellar ICH. Yahia thrombolysis and adjunctive stent retriever thrombectomy (SRT) is associated with better perfusion and outcomes. Despite benefit, 39% to 68% of patients had poor outcomes. Thrombectomy in AIS with LAO within 3 hours is performed as secondary after IV thrombolysis, which may be associated with delay. The purpose of our study is to evaluate the safety, feasibility, recanalization rate and outcome of primary STR within 3 hours without intravenous thrombolytic in AIS from LAO. SRT as an alternative to IV rtPA. Consecutive patients who underwent primary SRT for LAO within 3 18 patients with LAO; mean age 62.83±15.32 years and mean NIHSS 16±4; chose primary SRT after informed consent. Near complete (TICI2b in 1) complete (TICI3 in 17) was observed in all (100%) patients. Recanalization from symptoms and groin puncture was 188.5±82.7 and 64.61±40.14 minutes respectively. Immediate post-thrombectomy, 24 hour and 30 day NIHSS score was 4.44±3.75, 1.9±3.2 and 0.28±0.9 respectively. Asymptomatic perfusion related hemorrhage developed in 4 patients (22%). 90 days outcomes; mRS0 50%, mRS1 44.4%, and mRS2 5.6%. Our study demonstrates that primary SRT in AIS from LAO occlusion is not only safe and feasible, but associated with complete recanalization and good outcome. Further study is required. Currently, no reliable predictive tools are available to determine which patients with a large hemispheric infarction (LHI) will progress to cerebral herniation (CH). We sought to determine whether continuous measurements of blood pressure variance (BPV), heart rate variance (HRV), and entropy within 3 hours of admission would enhance the ability to predict future CH in LHI patients. Patients presenting within 24 hours of onset from an internal carotid artery (ICA) or middle cerebral artery 2013 and November 2015. Patients with CH were matched 1:1 by age and NIHSS with patients who did not have CH. Shannon entropy and standard deviation were used to measure the instability of HR and between groups. A model predicting CH utilizing the admission factors of age, gender, NIHSS, intraarterial (IA) therapy, and thrombolysis was compared to an admission model enhanced with BPV, HRV and entropy 3 hours after admission. Data from 40 patients were analyzed (median age 60 years old, median NIHSS 19, 43% women). There were no differences in the proportion of patients with a left hemisphere syndrome (45%), undergoing IA therapy (25%) or thrombolysis (43%) between groups. CH was observed a median of 3 days after the ictus. At 3 hours after admission, median measures of BPV, HRV and entropy were significantly higher in the CH group (P<0.001). A model of admission factors enhanced with physiologic data was better able to predict CH than a model with admission factors alone (ROC:0.77 v. 0.63, P=0.04) Poster 88 Early recognition of which patients with large hemispheric infarction (LHI) will develop malignant cerebral However, the early time-course of edema has not been adequately studied. We applied volumetric in the first 24 hours, prior to development of midline shift (MLS). We identified 38 LHI patients with scans within six hours of onset and subsequent scans early (6- While all 6 recently published endovascular stent-retriever randomized clinical trials (RCTs) were positive, their designs differed considerably particularly with regard to the extent of intravenous rt-PA use prior to thrombectomy. Here, we assessed whether rt- We adapted a method previously published by us to develop a pooled outcome model relating percent utilization of rt-presenting > 11,000 subjects and a range of 0-100% utilization of rt-PA. We correlated percent rt-PA and baseline -2) and mortality. This model includes ± p<.05 statistical interval surfaces to assess whether a trial's outcomes surpasses the variability of the pooled sample (Neurology 85:274-83, 2015) . Stent retriever RCTs were compared against the model. The mRS model showed excellent fit: R-square=0.83, p<0.001. Each stent retriever trial's outcomes exceeded mRS 0improvement varied dramatically according to %rt-PA, with the greatest improvement in those with 100% rt-PA use. When we included all case series and testing their outcomes at their baseline NIHSS and percent rt-PA use, shortest overall treatment times also related to improved outcomes (p=.011). Mortality trends were similar in that lowest mortality was seen in those trials with highest rt-PA use and the trial with higher than expected mortality had the lowest rt-PA use. An outcome model including % IV rt-PA permitted analysis of stent-retriever therapy compared to a large sample. Best outcomes were related to higher rt-PA utilization and shorter treatment times, suggesting a strong treatment interaction between modalities. Based on these results, until studied prospectively, IV rt-PA, if administered rapidly, should not be bypassed prior to thrombectomy. Intra-arterial mechanical thrombectomy (IAMT) is currently considered the standard therapy for acute thrombectomies performed under general anesthesia (GA) may adversely affect functional outcomes. We report our experiences with IAMT performed routinely under GA at the Montreal Neurological Hospital (MNH). This is a retrospective analysis of adult patients admitted to the MNH from January 2015 to April 2016 with AIS from proximal intracranial occlusions of the anterior cerebral circulation. All patients were assessed both clinically and radiologically. Forty-two patients underwent IAMT for AIS. Twenty-eight patients were included for analyses (4 had procedures under conscious sedation, 10 had missing 3 months assessment). Clinical outcomes were (mRS) 3outcomes (mRS 0--good and poor outcome groups respectively. Sixty-eight percent of the good outcome group received IV tPA, as compared to 50% in the poor outcome group. Patients in the good outcome group were also more 58% in the poor outcome group]. Intraprocedural MAP drops below 60 mmhg observed in 3 patients in the good outcome group versus non in the poor outcome group. In our small retrospective single centre study about thrombectomy for AIS under GA, our 3-month outcomes are comparable to larger studies where thrombectomies were done under conscious sedation. Pending the results of ongoing prospective trials about the use of GA during IAMT for AIS, our results do not support the reservations derived from recently published retrospective data on the use of GA in this context. Conservative initial management of young adults with severe hemispheric stroke in a comprehensive stroke center reduces decompressive craniectomy rates Pooled European trial results of early decompressive craniectomy (DC) did not require radiographic mass effect at the time of DC. Early surgery for supratentorial cerebral hemorrhage does not improve recovery or survival compared to initial conservative medical management. Early vs delayed DC for hemispheric A prospective inpatient neurosurgical database from October 2007 to March 2015 was queried for neurocritical care admissions for hemispheric -60 under IRB approval. A retrospective chart review was conducted using a structured questionnaire using the electronic medical record. We identified 30 patients who met the inclusion criteria for the pool were managed with medical treatment only (MTO) with average maximal septal shift of 5.2 mm and pineal shift of 3.1 mm. Twelve patients (40%) underwent DC with average maximal septal shift of 6.8 mm and group, MTO, and DC were respectively: MR 0-3 60% vs 67% vs 50%; MR 4-5 27% vs 17% vs 42%, and death 13% vs 17% vs 8%. Four patients in the MTO group declined DC; 3 died and one survived with MR of 4. No patients developed brainstem herniation prior to referral for decompressive craniectomy. Surgical complications death or survival with severe disabilities. Time of recanalization since symptoms is a strong predictor of outcome in patients who underwent stent retriever thrombectomy from middle cerebral artery occlusion. 3 hours since symptoms have not been clearly investigated especially, those with large artery occlusion (LAO) and underwent stent retriever thrombectomy (SRT) . Objective: To identify the predictors of outcome in AIS patients from middle cerebral artery (MCA) occlusion with large clot burden (LCB >8 mm) and underwent SRT, who recanalize less than 3 hours versus more than 3 hours since symptoms. software was used to analyze the data. 21 AIS patients who underwent SRT in MCA; age 70.62±13.94 years and mean admission NIHSS 17±5. Complete (TICI3) and partial (TICI2b) recanalization was observed in 90.50% and 9.50% respectively onset was 230±160 minutes. Presenting NIHSS of 16.76 dropped to 7, 5 and 2 at immediate, 24 hours and 30 days post SRT respectively. Good univariate analysis, recanalization time, immediate and 24 hours post SRT NIHSS were predictors of outcome (p-value= 0.0039, 0.003 and 0.043 respectively). In multivariate analysis, time of recanalization since symptoms (p-value=0.01) and baseline mRS (p-value=0.01) continued to be the predictors of good outcome. Our study demonstrates that patients with LAO from MCA who recanalize less than 3 hours of symptoms onset have good chance of good outcome compared to those who recanilize more than 3 hours. Therefore, all AIS patients with LAO should offer early SRT to achieve a good functional outcome. Further studies are required. Jennifer A. Frontera. Neurological Institute, Cleveland Clinic, Cleveland, OH, USA. Prospectively collected data of Heartmate II (N=332) and Heartware (N=70) LVAD patients from a single blood stream infection [BSI]), specific pathogens Mann-Whitney U, Chi--wise logistic regression analyses. Of 402 patients, LVAD infection occurred in 158 (39%) including: BSI in 107 (27%), wound infection in 31 infect P<0.05). Driveline and wound infection were not a -4.5, associated with BSI (aOR 2.6, 95%CI 1.3-5.1, P=0.005). There was no association with any specific infectious pathogen. Precautions to mitigate i demonstrate a causal relationship. the frequency of dysphagia is greater than 50%. The early clinical evaluation of swallowing disorders can help define approaches and avoid oral feeding, which may be detrimental to the patient. The aim of this study was to identify predictive clinical factors associated with enteral tube feeding in acute ischemic our database were reviewed. Clinical Early CT score (ASP association. Of the 326 patients, 84 used enteral feeding tubes (25.8%). The mean age (70.2 years -SD 13.1), mean GCS (12.7 -SD 2.1), mean NIHSS (12.6 -SD 5.6), and Aspect score (8.8 -SD 1.8) were significantly higher in the tube group. Logistic regression showed that only age (odds ratio [OR], 1.03; 95% confidence interval [CI], 1.00-1.6. P=0,025), NIHSS score (OR, 1.15; 95% CI, 1.05-1.25, P= 0,001) and NIHSS 10 (dysarthria) subscore (OR, 2.2; 95% CI, 1.2-4.05, P=0,011) were independent predictors of enteral tube feeding. A 3- In conclusion, combining information about age, NIHSS, NIHSS 10 subscore, may be a useful predictor Kyushu university, Fukuoka, Japan. Tissue plasminogen activator (te designed and developed an Information and Communication iPads, PCs, and bigdepartments before the patient's arriva number of calls is reduced. We compared the number of times that emergency room (ER) nurses called for computed tomography (CT) or magnetic resonance imaging (MRI) between before and after the system introduction. Before this system, ER nurses called for CT or MRI an average 3.8 and 4.2 times, respectively; after system introduction, the average number of times decreased to 1.2 and 1.8, respectively. Therefore, this system -PA. Also, it automatically records the transitions between could improve treatment times for IV tclinical trial to confirm the tool's efficacy. An important and controversial issue of peri-(EST) is the management of sedation and airway. According to retrospective data the widely favored intubation and general anesthesia (GA) appears associated with worse functional outcome compared to "conscious sedation" (GS) in the non-intubated state. SIESTA is a prospective, monocentric, outcome assessor-blinded, 1:1 randomized, parallel-group interventional study comparing non-intubated vs. intubated patients receiving EST for acute ischemic endpoint is the improvement of the National Institute of Health S Enrolment of the intended 150 patients has been completed. Sixty of the recruited patients are female these patients, 64% received pre-procedural rtPA. Seventy-eight patients were randomized to CS, 11 (14%) of these had to be converted to GA during the procedure. We will present preliminary results of the study, including the primary endpoint improvement in NIHSS after 24 hours and selected secondary endpoints. The aim of this study is the prospective randomized investigation of potential advantages of the non- In Young Chung. Departments of Neurology1, Seoul National University Bundang Hospital, Seongnam, Korea, Republic of. cerebral edema during therapeutic hypothermia us hypothermia. The authors retrospectively reviewed 21 patients with large hemispheric infarction who were treated with therapeutic hypothermia and hyperosmolar therapy from 2011 to 2014. Patients who were dead or underwent hemicraniectomy was defined as failure of therapeutic hypothermia. Infarction size was measured as sum of restricted area in diffusion weighted imaging which were performed on admission. B -CT was carried out regularly after onset of therapeutic hypothermia. Shift of septum pellucidum, pineal gland and choroid plexus calcification were measured in B-CT. Seventeen patients were enrolled after exclusion of 4 patients whose B-CT was inadequate to evaluate. Ten patients were successfully treated with therapeutic hypothermia (group success, n=10). Six patients were dead and 1 patient had hemicraniectomy (group failure, n=7). Initial infarction size between two groups was not significantly different. Both septum pellucidum shift (SDS) and pineal gland shift (PGS) were significantly different in groups on 2.5±0.5 days after onset of therapeutic hypothermia (mean SDS 5.0 vs. 14.9mm ; mean PGS 2.3 vs. 7.9mm). Specificity and positive predictive values for the failure calcification shift was not significantly different in groups during therapeutic hypothermia. Degree of progression of cerebral edema on 2.5±0.5 days after onset of therapeutic hypothermia helps to excellent in predicting fatal outcome. The main limitation of this study include its retrospective singlecenter nature, which may limit generalizablility of the study. Aminocaproic Acid for Reversal of Tissue Plasminogen Activator (tPA) Related Hemorrhagic Transformation in Acute Ischemic Stroke. -thrombolytic ICH. Aminocaproic acid (ACA) inhibits binding of plasminogen to fibrin, hence inhibiting fibrinolytic property of tPA. There is limited report a case series of ACA use for reversal of post-tPA HT. We reviewed the and identified patients treated with IV tPA. Patients with post-tPA HT who received IV ACA were identified. Data on demographics, clinical characteristics, NIHSS, ICH score, new thrombotic events during hospitalization, and hospital and intensive care unit (ICU) length of stay (LOS) were collected. A total of 36 patients developed post-tPA ICH, 12 of which received ACA. 58% of patients were male, mean age of 70.5±9.8 years. Patients received tPA within a mean time of 153±43 minutes from symptom onset, pre-tPA mean NIHSS was 11.6 ±5.7. Mean time for HT after tPA administration was 2.6±0.7 hours, with a hematoma volume of 45.5±58.3 mm3. In addition to ACA, 92% received cryoprecipitate, 67% platelets, and 17% fresh frozen plasma transfusions. 50% of patients had no hematoma expansion and 33% developed a new thrombotic event. Mean hospital LOS was 18±15 days and mean ICU LOS was 14±9 days. At the time of discharge 25% had an mRS of 4, 33% mRS 5 and 42% mRS 6. In this retrospective case series 50% of patients had hematoma expansion despite receiving ACA, while 33% had a new thrombotic event. Further research is warranted to determine the utility of ACA for the treatment of post-tPA HT. large vessel occlusion (LVO). While studies have analyzed difference in blood clot constructs, limited data is available understanding the effect of prior anti-platelet use on endovascular therapy (EVT) for IAS patients with LVO in the middle cerebral artery (MCA). We aimed to determine if prior anti-platelet use had effect on EVT procedure time, recanalization rate, and functional outcome measured by the modified A retrospective chart review was conducted of 63 consecutive AIS patients who underwent EVT of LVO of MCA at the University of Kansas Medical Center from 2008-2014. Outcomes were measured using total procedural and fluoroscopy time, procedural recanalization score using the modified thrombolysis in cerebral infarction score (TICI), time to recanalization, incidence of sICH, and 3-month mRS. Univariable and multivariable analysis were performed. .05 for all) were similar between antiplatelet use versus those without. In separate multivariable models (adjusting for all significant variables), antiplatelet use was not associated with tandem ICA occlusion, total procedure time, fluoroscopy time, good recanalization, or 3-month mRS. No impact was found of prior antioutcomes in AIS patients undergoing EVT. These findings should be further confirmed in a larger database and prospective cohort study. The study evaluated the effect of a neurology-specific heparin infusion protocol with more frequent PTT monitoring and a narrower goal PTT range (1.5x normal) on rate of hemorrhagic or thromboembolic events. This is a retrospective cohort study evaluating patients before (October 2011-September 2013) and after (October 2013-September 2014) implementation of a neurology-specific heparin infusion protocol. All patients >18 years old receiving intravenous heparin with a diagnosis of acute ischemic st evaluated for inclusion. Primary outcomes are time to first therapeutic PTT and time to therapeutic PTT range. Secondary outcomes include rate of intracranial hemorrhage, rate of thromboembolic events, protocol compliance, number of subtherapeutic and supratherapeutic PTT values, time to initiation of oral anticoagulation, duration of heparin infusion, and number of heparin titrations. Time to therapeutic PTT range was 18.7 hours in the pre-protocol group (n=108) and 17.9 hours in the post-protocol group (n=22) (p=0.049). Number of PTT values per patient was 20.2 in the pre-protocol group and 33.8 in the post-protocol group, of which 27.5% and 44.4% were therapeutic, respectively. Percentage of supratherapeutic PTT values was 22.2% and 16%, respectively (p=0.025). Time to first PTT, time to first therapeutic PTT, and percentage of subtherapeutic PTT values were not significantly different. Assessment of secondary clinical outcomes is ongoing. Our neurology-specific heparin protocol resulted in a faster time to therapeutic PTT range with a higher percentage of therapeutic PTT values and fewer supratherapeutic PTT values. Investigation regarding change in incidence of hemorrhagic and thromboembolic complications is ongoing. Hospital Moyses Deutsch, in the southern city of São Paulo is indicated for the use of rtPA intravenously ctive to demonstrate experience the use of intravenous Retrospective study, in all cases of isch 2011 to February 2016. Protocol indicates the use of alteplase patients with inclusion criteria, the period between the onset of symptoms and hospital admission up to 4.5 hours and no contraindication to the use of thrombolytics, NIHSS calculated on admission and 24 hours after thrombolysis. Computed tomography (CT) on admission and after 24 hours.Evaluation required by neurological Telemedicine Hospital Israelita Albert Einstein shortly after the conclusion of the TC Cranio. trough has been triggered in 51 cases. Time between onset of symptoms and drug administration, 19 patients less than 90 minutes, 37 patients between 90 and 180 minutes, 8 patients 180-360 minutes .The average NIHSS at admission was 18, with 42 patients showed a reduction of 5 or more the points NIHSS score within the first 24 hours. 8 patient non-symptomatic intracranial hemorrhage and 6 symptomatic intracranial hemorrhage and 9 deaths during the period. All patients receiving alteplase in the recommended time interval and underwent CT Cranio control. Some cases were not triggered by the evaluation of Neurology Telemedicine. There was improvement in NIHSS score similar percentage observed in reference studies. The protocol implementation has been adequate excellent support of Telemdicina Neurology team. Good profitability of time and therapeutic efficacy. The mortality that correlated with the severity of patients and the NIHSS admission. This data ratifies the intracerebral hemorrhage (ICH) is approximately 6% and non-ICH bleeding over 5%. Hypofibrinogenemia occurs in approximately 5% of tPA-treated patients and i Cryoprecipitate is often used to restore fibrinogen levels, despite limited published evidence. Cryoprecipitate has several limitations, including the need for ABO matching, thawing, and concerns regarding potential transmission of viral pathogens. Riastap, a purified fibrinogen concentrate, is a promising alternative to cryoprecipitate for the reversal of hemorrhage post-tPA. The objective of this study was to evaluate the safety and efficacy of Riastap for the treatment of post-tPA hemorrhage. A single-center retrospective observational analysis was conducted to evaluate patients who received Riastap for the treatment of postmeasure was reversal of hypofibrinogenemia. Hypofibrinogenemia was defined as a fibrinogen level < hospital mortality. Eleven patients were included in our analysis. The average dose of Riastap administered was 3,443 units. Five patients had hypofibrinogenemia prior to Riastap administration, with a mean fibrinogen level (IQR 123 to 150). The six patients who were not hypofibrinogenemic at baseline had minimal effect on fibrinogen levels post--14.8 to 57). One patient was diagnosed with a deep vein thrombosis 10 days post-Riastap administration and no infusion reactions were reported. In-hospital mortality occurred in 45.5% of our patient population. Riastap administration successfully and safely treated hypofibrinogenemia in patients with post-tPA hemorrhage. Casey Catheter-associated urinary tract infections (CAUTI) are the most prevalent hospital-acquired infections (HAI), and account for more than 560,000 cases annually and 13,000 deaths per year. CAUTI is the most common HAI in Neuroscience Intensive Care Units ( immobility and urinary retention. We implemented a team-driven multimodal quality improvement initiative to reduce CAUTI and catheter-utilization rates in the NSICU. We convened a multidisciplinary CAUTI prevention team including nurses, advance practice providers, physicians, and infection control specialists. We developed a CAUTI surveillance program that involved review CAUTI and catheter utilization rates. We applied root cause analysis to target improvement opportunities, and implemented interventions including best-practice catheter insertion techniques, modification of bowel regimen, and guidelines for timing of catheter removal. We also implemented a daily goals tool to prompt standardized team communication surrounding catheter removal on morning interdisciplinary rounds. We performed Poisson generalized linear model analyses, controlling for linear time trends and testing with sandwich errors. We analyzed data before and after implementation of interventions, spanning a time period of 28 months. --0.58), as did mean catheter days per --33.2). We observed a 68% reduction in CAUTI rate adjusted by catheter days (95% CI 6%, 89%), p = 0.035. There was an 11% reduction in catheter utilization rate adjusted by patient days (95% CI 3%, 19%), p = 0.011. A team-driven multimodal approach to CAUTI reduction resulted in significant decreases in CAUTI and catheter utilization rates in the NSICU. Team-driven interventions enhance communication and shared -up is ongoing to evaluate sustainability. Non-neurological complications involving a single or multiple organ systems during intensive care in critically ill patients of traumatic brain and spine injuries is significant cause of poor prognosis but often not well managed. The aim of this study was to assess the frequency of such complications in neuro ICU and assess their impact on morbidity and mortality. A prospective observational study on 200 patients of varied demographic profile admitted in neuro ICU over a period of 6 months for injury and associated multisystem involvement was conducted. Significant predefined parameters addressing the non-neurological complications occurring during their ICU stay were recorded including disturbances and bleeding complications. The study period was from admission to the ICU till the discharge from the ICU or demise. 63% of patients developed respiratory complications in the form of chest infiltrate (50%) and atelectasis (10.31%). 35.5% of patients suffered from cardiovascular complications. 37% of patients had dyselectrolytemia, commonest being hypernatremia due to hypovolemia (70%). Sepsis was observed in 23.5%. Bleeding diathesis and acute renal injury were observed in 22% & 3.5% of patients respectively. 17% of the patients succumbed to injury out of which 64.7% was due to non -neurological cause. Further results will be discussed in detail with inferences at the meeting. Intensivists in neuro ICU must consistently assess and treat the non-neurological complications in traumatic brain and spine-injured patients and deliver appropriate care to bring down the mortality and morbidity and improve outcome. Neurocrit Care (2016) 25:S1-S310 Transcranial Doppler (TCD) is a useful ancillary test in neurologic critical care for monitoring patients at with elevated intracranial pressure and cerebral vascular resistance. The normal values of cerebral blood flow velocity and PIs are significantly distorted by nonpulsatile blood flow, as in patients on Venoarterial Extracorporeal Membrane Oxygenation (VA-ECMO) circulation. This analysis evaluates changes in PI measurements in patients on VA-ECMO following cerebral vasodilation, vasoconstriction, increased intracranial pressures, or cerebral circulatory arrest. Data from 11 TCDs in 8 patients on VA-ECMO in the Cedars Sinai Medical Center Cardiac Surgical ICU were reviewed. Mean PIs were calculated for each patient using Gosling's PI formula. The values obtained were compared with Ejection Fractions (EF) obtained within 24 hours of TCD. PIs were globally low or absent in all 11 TCDs. The non-demonstrable PI seen in one patient is from severely diminished cardiac function, resulting in TCDs were performed at the initiation and conclusion of VA-ECMO cannulation. The PI values for these TCDs correlated directly with changes in EFs. Also, an abrupt rise in PI to normal value was seen with placement of a Total Artificial Heart and return of pulsatile circulation. We demonstrate that patients on mechanical circulatory support demonstrate low-absent PIs on TCDs. ion or cerebral circulatory arrest. Moreover, rising PIs in patients with improving cardiac function should not be confused with elevated intracranial pressures. Venous thromboembolism (VTE) prophylaxis in underweight patients with neurologic injury remains unaddressed by recent guidelines and primary literature. This study aimed to describe VTE prophylaxis strategies employed in this population and compare the impact of underweight and non-obese patients on thrombotic and bleeding events. Underweight and non--care unit from September 1, 2009 to July 31, 2015 were retrospectively identified. Underweight was defined as a body -obese as a BMI 18.6excluded if they received >1 VTE prophylaxis regimen, had an ICU length of stay <72 hours, or received VTE prophylaxis for <48 hours. Patients were stratified to non-obese and underweight groups and subsequently matched 2:1, on age and diagnosis. Prophylaxis regimen, prevalence and type of The most common regimen in the underweight (n=107) and non-obese (n=214) groups was unfractionated (UFH) 5000 units subcutaneously every12hrs (69.1% vs. 83.6%; p=0.003). Only underweight patients received UFH 2500 units subcutaneously every12hrs (17.8% vs. 0.0%; p<0.0001). Non-obese and underweight patients had no difference in the proportion of overall bleeding (5.6% vs. 10.3% p=0.204) and thrombotic events (6.1% vs. 2.8% p=0.204) while receiving VTE prophylaxis. Further analyses revealed a statistically significant difference in the proportion of underweight patients that developed intracranial hematoma expansion while receiving prophylaxis versus non-obese patients arge dispositions were seen between groups. Current practice does not reflect a consistent dose reduction for neurologically-injured, underweight patients. Caution should be considered when using increased doses of UFH in neurologically-injured patients. Continued assessment of VTE prophylaxis is needed to confirm these findings. Patients in the neuro intensive care unit (NICU) commonly need vasopressor infusions for various reasons. The traditional approach is to insert central venous catheters (CVC) for this purpose. CVCs carry among others. Phenylephrine is a commonly used vasopressor in the NICU. The purpose of this study was to evaluate the safety of phenylephrine infusion through peripheral intravenous catheter (iv). Retrospective review of consecutive patients admitted to the Neuro ICU and administered phenylephrine infusion through peripheral iv line. One hundred patients, mean age 66 years (SD ± 14) were included in the analysis. Fifty-four (54%) were men. Eightydisease. The most common indications of phenylephrine were hemodynamic augmentation (37%), multifactorial transient post-operative hypotension (28%) and hypotension due to other causes (22%). Most common location of IV line was proximal upper extremity (28 % antecubital, 28% forearm) with gauge of the IV line between 18 (45%) and 20 (44%). Average maximum rate of phenylephrine infusion duration of 19 hours (SD ± 18, range 1 to 114). Central line was eventually placed in 16% due to physician preference and in another 5% due to a change of vasopressor to norepinephrine. There were 2 any complications. Infusion of phenylephrine through peripheral iv appears safe when used in moderate doses for the short term and can be considered in lieu of placing a central line solely for this purpose. This may reduce the complications associated with central lines. Osmotic therapy continues to be standard care in the medical management of cytotoxic cerebral edema. The long term use of monotherapy is often limited by side effect profile. The combination of low dose mannitol and hypertonic saline may provide synergistic effect by combining mechanisms of action, while limiting dose-related toxicities of either agent. We investigated safety and efficacy endpoints for combination therapy. A single-center retrospective cohort study from August 2014 to December 2015. Identified patients were administered combination mannitol and hypertonic saline for >48 hour duration. The primary outcome criteria, sodium fluctuation, and central pontine myelinolysis. 31 patients (mean age 55±14, 52% male) were identified. Underlying neurological injury included 44% brain injury. 52% had neurosurgical management. The average number of mannitol doses given was 22, and the average duration of hypertonic saline was 110 hours. The range of mannitol dose was 0.14 -0.6 percentage of osmotic therapy doses were held for pred mannitol, 17% held hypertonic saline). AKI occurred in 10 (32%) patients (9 -stage 1 AKI, 1 -stage 3 central pontine myelinolysis. Low dose combination osmotic therapy was tolerated with no central pontine myelinolysis and rare sodium fluctuations; however transient low grade AKI was common. Further study is needed to evaluate the relative efficacy of single and combination osmotic therapy in the neurocritical care population. Clinical characteristics of nonconvulsive status epilepticus diagnosed by simplified continuous EEG monitoring at an emergency intensive care unit. The clinical characteristics of Nonconvulsive status epilepticus (NCSE) presenting in ICU in Japan is limited. Our institute provides a noninvasive monitoring system of two-channel simplified continuous EEG (sEEG) for the bedside monitoring of cerebral activities. The present study aimed to elucidate the clinical characteristics of NCSE in patients with altered mental status (AMS). This single-center retrospective study comprised 149 patients who were hospitalized between March 1, 2015 and September 30, 2015 at the emergency intensive care unit (ICU) of the Kagawa University Hospital. Primary outcome was the NCSE incidence. The secondary outcome was the comparison of duration of ICU stay, hospital stay, and a favorable neurological outcome (FO), as assessed using the tal between the groups with and without NCSE. FO and poor neurological outcomes (PO) were defined as mRS scores of 0-2 and 3-6, respectively. Simplified continuous electroencephalogram (sEEG) was monitored in 36 patients (median age, 68 years; 69.4% males) with acute AMS. NCSE was observed in 11 (30.1%) of the 36 patients with AMS. Rates of FO, duration of ICU stay, and hospital stay were not significantly different between the NCSE and non-NCSE groups (p = 0.45, p = 0.30, and p = 0.26, respectively). Approximately 30% of the patients with AMS admitted to emergency ICUs developed NCSE. The outcomes of AMS patients with and without NCSE did not differ significantly when appropriate medical attention and antiepileptic drugs were initiated. sEEG monitoring may be recommended in patients with AMS in emergency ICU to obtain early detection of NCSE followed by appropriate intervention. Approximately 200,000 people per year will need mechanical ventilation secondary to neurological injury resulting in significant mortality. Delaying liberation in neurologically impaired patients otherwise ready for liberation is a source for significant hospital charges. There is no clear guideline to suggest one spontaneous breathing trial (SBT) over another in predicting the liberation success. Zero pressure support and zero positive end expiratory pressure (PEEP) or ZEEP is a traditional method assessing patient's readiness for mechanical ventilation liberation. However, neurologically injured patients with was to assess mechanical ventilation liberation in patients who failed ZEEP and subsequently passed pressure support trial. Retrospective analysis of liberation in 134 intubated patients in a neurosciences intensive care unit. All patients were initially challenged with ZEEP. If passed, patients were liberated from mechanical McNemar's exact test. P value <0.05 was considered significant. 134 adult (>18 years old) patients were included. The majority of patients were successfully liberated from mechanical ventilation using 30 minute ZEEP trial alone (n=123, 91.7%). Eleven (8.21%) patients failed .74%) required reintubation. Ten (7.46%) This study shows that the majority of patients can be successfully liberated from mechanical ventilation successful liberation from mechanical ventilation. Neurocrit Care (2016) 25:S1-S310 vancomycin establishey using NONMEM software by the department of pharmacy of Nanjing Drum Tower Hospital in neurosurgical intensive care unit patients. According to the patient's gender, age, body weight, serum creatinine (SCr), serum albumin (ALB), the actual measured value. During the period from March 2013 to March 2014, 42 patients including 25 male and 17 female, whose age is 53±14 years old (22-78 years old), were grouped and 53 copies of blood concentration of vancomycin were measured. The average concentration was 10.9m the actual measured value (r=0.857, P<0.001), the mean absolute percentage error (MAPE) was 0.4079. neurosurgical intensive care unit patients for drug value prediction and drug dosage guidance. But because of coma, the body weight estimation has errors (about 30%). The renal function sometimes changed by contrast agent and diuretic drug has an impact on predictive results. By adjusting methods, accurate prediction rate increased to nearly 70%. Xi Liu-DeRyke, Sindhuri S. Avula, Jason J. Vilar. Florida Hospital Orlando/Pharmacy department, Orlando, FL, USA. little data exists concerning clevidipine in this population. Large variations in BP during the first 24 hours is an independent predictor for poor outc aneurysmal subarachnoid hemorrhage (aSAH) admitted to neuroscience intensive care unit from January 2012 through December 2015 were identified retrospectively. Patients were included if they received clevidipine or nicardipine for initial acute BP management, and BP goal was defined by the prescribers. BP variability was measured by standard deviation (SD) of mean arterial pressure (MAP) over the first 24 hour of therapy. Seventy three patients were included in the analysis (clevidipine n=18; nicardipine n=55). Admission and 25% aSAH. Baseline MAP between clevidipine and nicardipine group was comparable (102 vs. 109 mmHg). The number of BP recordings was similar between groups (clevidipine 22 vs. nicardipine 24; p=0.51) and the average time to goal was 116 minutes and 47 minutes, respectively (p=0.15). The average MAP during the first 24 hours was similar (clevidipine 86 vs. nicardipine 84 mmHg; p=0.26). Although not statistically significant, clevidipine group had a higher percentage of BP above goal compared to nicardipine group (24.6% vs 9.8%; p=0.09). There was no significant difference in BP variability between clevidipine and nicardipine group (SD 11.6 vs. 13.5 mmHg; p=0.09). Our study did not find a difference in BP variability between clevidipine and nicardipine following acute Long corrected QT interval (QTc) has been associated with malignant ventricular arrhythmias specifically present in neuro intensive care unit (NICU) patients. In addition to medical causes, acute neurologic insult has been shown to cause multiple neuro-cardiac manifestations including QTc prolongation. Prevalence ge and surgical ICU patients which have different disease processes compared to NICU. Retrospective review of consecutive patients admitted to the Neuro ICU and having abnormal QTc interval. Ninety-five patients, mean age 66 years (SD ± 14) were included in the analysis. Fifty four (54%) were men. Average duration of hospitali (47%), subdural hemorrhage (13%), and cerebral hemorrhage (12%). Fiftycardiovascular disease, 30% had abnormal ejection fraction. Thirty-seven patients (39%) needed abnormalities were observed throughout the hospitalization and patients frequently received QTc prolonging drugs. Mean QTc was 495 ms (SD ± 37, range 450-659). There were 6 episodes (6%) of nonsustained ventricular tachycardia which did not lead to any immediate consequences. One patient had cardiac arrest following anesthesia for hemicraniectomy. Initial rhythm was asystole followed by fine ventricular fibrillation and therefore could not be clearly attributed to prolonged QTc. There were no episodes of TdP. -sustained ventricular tachycardia was observed without leading to cardiac arrest. No episodes of TdP were observed in these patients. Deep venous thrombosis (DVT) of the lower extremities is a common cause of morbidity and mortality among neurologically injured patients. The data on incidence and prevalence rates of DVT among high medical or surgical intensive care unit with very limited information on patients in neuro-intensive care units (NICU). The aim of the present study is to assess the incidence and prevalence of deep vein thrombosis among patients admitted with acute neurologic injury. Our institution routinely conducts ultrasound screening within 24 hours of admission and -month period. Data was abstracted and analyzed to assess the prevalence of DVT in this period. We excluded patients presenting with superficial vein thrombosis, hematoma and chronic venous scarring. Over a period of one year; the prevalence of DVT was 6.1% (n=161). Of the 161 cases that were diagnosed with DVT; more than one-half (54.7%) presented with DVT at the time of admission. 73 patients (45.3%) acquired DVT during hospitalization. Majority of the patients with DVT at the time of admission are Caucasian males with mean age 72 and mean SAPS II score of 34.2, ranging between 13 and 71. Prevalence of DVT at the time of presentation to the neuro ICU is relatively high. Further research is S148 Neurocrit Care (2016) 25:S1-S310 Practice guidelines recommend that practitioners should not prescribe prolonged prophylactic systemic antibiotics (PPSA) after neurosurgical procedures, even if drains are left in place. We sought to evaluate 1) current practice patterns related to PPSA administration to neurosurgical patients with drains and devices and 2) practitioner perception about the need for PPSA in this population. We surveyed members of the Neurocritical Care Society on use of PPSA (defined as maintenance antibiotics after the time of insertion) and personal perception about the need for PPSA in patients with intraparenchymal monitors, subdural drains, subgaleal dra -Pratt spinal drains, and lumbar drains. Of 52 respondents, routine institutional use of PPSA was reported by 29fewest respondents reported use with subgaleal drains and the most respondents reported use with -Pratt spinal drains with instrumentation. Respondents had varying personal opinions on the need while the lowest (19%) was for patients with subgaleal drains. It is on the use of PPSA in patients with neurosurgical drains is necessary to optimize patient care. Transition to comfort measures only (CMO) in an intensive care unit (ICU) is a common but delicate process that requires a well-organized multi-disciplinary and multi-professional care model. The goal of this survey was to understand potential deficiencies and inconsistencies in the transition to CMO in order to develop a process to improve the quality of care provided to patients at their end-of-life. After obtaining IRB approval, a web-based questionnaire was distributed to attending physicians, residents, fellow trainees, bedside nurses, respiratory therapists and spiritual care team members, who deliver care to patients in neurological, medical, trauma-surgical, and burn intensive care units at the University of Washington's Harborview Medical Center. Overall survey response was 46.2% (297 out of 725). The concept that transition to CMO is a multidisciplinary process was not universal with only 9.5% of all bedside nursing and respiratory therapists feeling invited and actively engaged in the discussion about CMO. The majority of respondents (75%) encountered at least one 'less than ideal' transition to CMO. Deficiencies identified included gaps d interprofessional conflict (6.5%). Most participants (75%) agreed that a formalized process might reduce round the transition to CMO. We identified several barriers towards an optimal, collaborative transition to CMO in ICUs at a large academic medical center, highlighting the need for a formalized process. Such a process would ensure communication between various disciplines and professions, and offer healthcare providers opportunities for dialogue to address all the issues resulting in a smooth transition to CMO. Dexmedetomidine's propensity to cause bradycardia is well documented in non-neurocritically ill patients. e units (ICU) when defined as heart rate (HR) <60 bpm. Neurocritically ill patients have been excluded from all randomized trials. The aim of this study is to assess the development of bradycardia in patients with neurologic injury who have received dexmedetomidine for sedation in the ICU. was done via nursing driven protocol with no loading doses. Primary outcome was the incidence of bradycardia (HR <60 bpm) during first administration. Secondary outcomes were percent decrease in HR from baseline and time to event analysis using Cox regression. Mortality in the ICU was collected. A total of 202 patients were included (61% male, mean age 55 years, mean SAPS II 42). The most Bradycardia occurred in 80 patients (39.6%). The average maximum dose was higher in patients who infusion duration did not vary. Baseline HR was lower in bradycardic patients (87±21 bpm vs. 96±17 bpm, P=0.001) and a larger mean percent decrease in baseline HR was observed (32.8% ±15.7 vs. 3.5% ±20.8). Median time to first bradycardic event was 11 hours [7.5-15.0] which was significantly impacted by baseline HR (hazard ratio 0.83; 95% CI, 0.72-0.95; P=0.006). Mortality was significantly lower in patients who developed bradycardia, 3.75 vs. 14.7% (P=0.017). These data indicate that bradycardia associated with dexmedetomidine occurs considerably among the neurocritically ill. Future assessment of clinicall development would further contribute to the limited data of dexmedetomidine use within this patient population. Medical complications after subarachnoid hemorrhage. Medical complications occur frequently after subarachnoid hemorrhage (SAH). Their impact on outcome has been previously described, but was not validated in international series of SAH. We evaluated 139 consecutive patients admitted to a tertiary hospital in Brazil with SAH from January -6) at discharge. We calculated the frequency of medical complications according to prespecified criteria and eva of poor outcome. Thirty-six% had a poor outcome; mortality was 12.6%. The most frequent complications were hyperglycemia (37%), fever (14%), pneumonia (1.4%), hypotension (<90 mm Hg systolic) treated with vasopressors (3.6%) and venous thromboembolism (1.5%). Hyperglycemia (odds ratio [OR], 9.0; 95% confidence interval [CI], 1.7-47.8; p=0.01) significantly predicted poor outcome after adjustment for age and Hunt-Hess grade. Hyperglycemia affected more than one third of patients with SAH and was significantly associated with poor functional outcome. Critical care strategies directed at maintaining normoglycemia may improve outcome after SAH. S152 Neurocrit Care (2016) 25:S1-S310 Blood pressure (BP) can be measured in critically-ill patients using non-invasive (oscillometric) blood pressure (NIBP) and intra-arterial blood pressure (IABP) monitoring. The accuracy of NIBP compared to the "gold standard," AIBP, has been questioned. NIBP monitors generally tend to over-read at low values and under-read at high values compared to IABP. Previous studies exploring NIBP-IABP correlations have generally been performed on patients not receiving continuous infusions of vasoactive medications. Since many critically-ill patients receive vasopressors and antihypertensive agents, we wanted to study the relationship between simultaneously-measured NIBP and IABP recordings in this patient population. We prospectively identified patients (N=25, target N=70) admitted to a Neurosciences ICU, who had simultaneous IABP and NIBP monitoring while receiving intravenous infusions of manually abstracted via retrospective chart audit. Covariate and demographic variables were also abstracted and entered into an electronic spreadsheet. Statistical analysis performed using SAS v9.4. Initial results from 25 subjects (60% Caucasian, 56% male, mean age 60.3 years, mean BMI 30.3), observations. Independent-samples t-tests showed a significant difference between NIBP vs IABP readings: ([SBP: m=125 vs 130mmHg respectively; p50mmHg (1.7%)]. Bland-Altman plots demonstrated good inter-method agreement between NIBP-IABP measures (when visually excluding outliers) and -AIBP SBP differences at higher blood pressures. Preliminary analysis indicates a statistically significant difference between NIBP-IABP readings for patients on vasoactive medications. Yet when visually excluding outliers, there is good inter-method agreement. Data from the entire cohort will be available for presentation at the NCS annual meeting and will be helpful in choosing appropriate BP monitoring methods for patients on vasoactive infusions. New-onset refractory status epilepticus (NORSE) is an important syndrome often associated with a poor outcome. The aim of the present study was to review NORSE cases in our hospital and to determine the main factor that may improve patient outcomes. We retrospectively reviewed our hospital medical records and database of electroencephalograms (EEGs) over a 3 years period (May 2013-May 2016). In our facility, we performed 24-h continuous EEG monitoring using the international 10-20 system. Of the monitored patients, we excluded those who were meningitis, herpes encephalitis, and history of epilepsy. We discussed their causes and neurological outc (GO) was defined as a mRS score of 0-2, whereas a poor neurological outcome (PO) was defined as a score of 3-6. Moreover, we attempted to determine the main factor that influenced the neurological outcomes. We identified 785 patients who had undergone EEG, and identified six NORSE patients among them. The on arrival was 9. All patients were diagnosed with limbic encephalitis and all had nonconvulsive status treatments, such as steroids, were delayed in all PO patients. On the basis of our data, the cause of all NORSE cases was limbic encephalitis. In PO patients, definitive treatments, such as pulse steroid therapy, were delayed. This is a relatively small study. Further research is needed to identify the factors which could improve outcomes. Multi-drug resistant organisms (MDRO) are an increasing concern in health systems. Pathogens such as Pseudomonas aeruginosa, Acinetobacter baumanii, and carbapenamase-producing Enterobacteriaceae hold highest mortality rates especially when the central nervous system is involved. When MDROs are cultured treatment options are becoming limited and reliance on medications such as colistin and aminoglycosides is becoming more prevalent. However, penetration of these therapies into the central nervous system is concerning therefore local administration is a potential concomitant therapy. This study was a retrospective chart review from 2009 to 2015 for all patients with documented MDROs who received intraventricular colistin. Seven patients from 2009 to 2015 met inclusion criteria. The average age of the patients included was 49 years old, 4 were males, and the median length of intensive care unit stay was 30 days. The dose of colistin used for each patient was 10 mg via intraventricular route. The duration of therapy ranged from 2 -14 days and all cerebrospinal fluid cultures were sterile at 7 days after administration of colistin. Each patient received concomitant systemic antibiotics while receiving intraventricular colistin. Six of the seven The use of intraventricular colistin was not associated with any reported adverse events. The use of intraventricular colistin was associated with positive clinical outcomes with no reported adverse effects. Myasthenic Crisis: epidemiology, economics and opportunities for change -a single center retrospective analysis. Avinash B. Kumar, Vikram V. Tiwari, Kevin K. Scharfman, Justin J. Calabrace. Vanderbilt University Medical Center, Nashville, TN, USA. Myasthenia gravis (MG) patients are admitted to the ICU for myasthenic crises characterized by immunoglobulin (IVIg) or plasmapheresis and supp and care flow maps of patients admitted to our institution This is an IRB-approved, retrospective cohort study of patients admitted to a tertiary Neuro ICU. We included adult (age > 18 years), with a diagnosis of MG who received plasmapheresis or IVIg therapy. The demographics and clinical data were summarized for patients in the IVIg and Plasmapheresis cohorts. We also compared the ICU and hospital LOS and in addition the hospital cost data for patients in both cohorts. The final cohort included 153 hospital encounters for 82 individual patients (46 Female) admitted between 2006-14. The mean age on admission was 56.9±19.7 Y. There was no significant difference between c ventilation; the median duration of MV was 7.5d (range 1-34). The median readmission rate was 1.5± 0.5. ). 13 patients had multiple crisis readmissions (>2). This cohort was socially challenged ( 5 divorced,8 analysis included 33 patients (16 in IVIg cohort and 17 in plasmapheresis cohort). The mean hospital costs (variable direct-technical) in IVIg cohort was approx. $5000 more than the plasmapheresis cohort. There was no statistically significant difference between in the limited financial analysis. 56.25% of patients were either medicare or Medicaid patients, 21.25% wer The disease burden on patients and hospitals of this orphan condition are significant and continues beyond the ICU. Evidence based care pathways need to be explored for the management of this high resource utility disease. Botulism is a rare potentially fatal and treatable disorder caused by a bacterial-produced toxin that affects the presynaptic synaptic membrane resulting in a characteristic neuromuscular dysfunction. It is caused by either the ingestion of the toxin or the bacteria, inhalation, or wound infection. We present our We report 15 consecutive cases of botulism presenting to University Medical Center of El Paso. Medical records where reviewed to obtain demographic information, clinical presentation, treatment and outcome. in popping and had abscesses in the administration areas. By history the most common 3%, ophthalmoplegia 53%, ptosis 46%. Interestingly enough, in those patients with the documentation the pupils were reactive in 46%. All patients required mechanical ventilation and all were treated with the trivalent antitoxin. Thirteen patients were disc derivatives (mostly 6-monoacetylmorphine and 3-monoacetylmorphine) was associated with the development of botulism. Its presence in the US-Mexican border is not surprising since is frequently produced in Latin America. Its association with the development of botulism should be recognized early to allow a prompt diagnosis and treatment with the antitoxin. A clinical feature worth noting is the presence of normal pupillary light reflex in nearly half of patients thus a normal pupillary response should not be used as a finding to exclude botulism. in clinical trials limits the discovery of effects that may be particularly relevant to underrepresented populations. CLEAR III, a 500presented an opportunity to evaluate African American (AA) enrollment. Investigators across 61 U.S. hospitals screened 8,587 patients over a 5-year period: 25% AA; 4.3% Asian; 0.5% Native American; 0.3% Pacific Islander; 69.7% White; 0.2% mixed race; and 10.8% not reporting. The mean age for AAs was younger at 57.9 (SD: 0.6) vs. 65.7 (SD: 0.4) for Whites (p=0.001). The randomized-to-screened ratio for AAs was 8.7 vs. 3.5% for other racial groups (p<0.001). Higher 3.6%, p=0.26); Northeast (7.8% vs. 2.9%, p<0.001); South (8.2% vs. 4%, p=0.003); and West (8.9% vs. 3.8%, p=nonsignificant). African Americans were less frequently excluded due to non-hypertensive etiology (16.9% vs. 26%, p<0.001), not having ventricular drainage (22.1% vs. 25.4%, p=0.031), DNR status (18.4% vs. 25.2%, p=0.027) and unstable bleeding (13.9% vs. 25.1%, p=0.024); and more frequently excluded for prior disability (32% vs. 24.9%, p=0.23), larger hemorrhages (26.7% vs. 24.8%, p=0.44), and by investigator decision (46.5% vs. 24.9%, p<0.001). Of the 119 patients who refused consent, AAs accounted for 44.1% vs. 49.5% of Whites. In an unadjusted logistic model, the odds ratio for successful enrollment of AAs was 2.74 (p<0.001) vs. Whites, and 2.45 (p<0.001) after adjustment for age and Hispanic ethnicity. The age <49, 50-59 and 60-69 subgroups maintained higher adjusted odds ratios than Whites at 2.60 (p<0.001), 2.20 (p<0.001) and 2.43 (p<0.001) respectively; the above 70 subgroup was not significantly different. Others have reported difficulty enrolling AAs into clinical trials. CLEAR III suggests this may be a misperception S158 Neurocrit Care (2016) 25:S1-S310 pipeline that utilizes machine-learning algorithms to integrate clinical data and quantitative EEG (QEEG) trends, providing continuous estimation of prognosis. A collaboration involving two academic centers in the U.S. assembled a retrospective clinical and EEG database of adult subjects with cardiac arrest and return of spontaneous circulation who underwent continuous EEG monitoring. Four QEEG features were included in the model: regularity, Tsalis entropy, alpha-to-delta ratio, and voltage <20 uV. Only the first 48 hours of EEG data were evaluated in this analysis. Poor outcome was defined as Cerebral Performance Category of 3-5 at discharge. Ten fold cross validation resampling method was utilized, and model performance evaluation metrics were area under ROC curve (AUC), sensitivity, and specificity. The algorithm provided an hourly estimation of poor Clinical and EEG data was available for a total 173 subjects. Mean age was 55.3 years and overall mortality was 55.7%. One hundred and twenty subjects (69.4%) had poor outcome. Our multiparametric QEEG method achieved optimal performance for mortality prediction at 24 hours (AUC 0.75), with a sensitivity of 86% and specificity of 66%. Optimal poor outcome prediction performance was achieved at 26 hours (AUC 0.72), with a sensitivity of 88% and specificity of 70%. At a false-positive rate of 0%, the sensitivity for poor outcome was 53%. Alpha-delta ratio and voltage <20uV were independently associated with mortality and poor discharge outcome at 24 hours (p<0.05). Employment of machine-learning methods in QEEG analysis allows early and robust outcome prediction in cardiac arrest. This approach has potential to facilitate real-time individualized prognostication in cardiac arrest. Secondary brain injury may be a significant barrier to survival following extracorporeal membrane oxygenation (ECMO) for, otherwise reversible, cardiorespiratory failure. Prevalence of brain injury phenotypes on neuroimaging were described in our prior wor neurological injury on outcomes in adult patients on ECMO. A retrospective cohort of ECMO-treated adults. Clinical and outcome data was obtained from electronic chart abstraction of clinical and physi of decannulation. 27% (N=18) of CT scans and 93% (N=13) of MRIs had abnormal findings. Intracranial hemorrhage was seen in 20% patients with neuroimaging. In addition, 86% of MRIs revealed diffuse significant difference in survival to hospital discharge and mean modified with or without neuroimaging during ECMO (64% vs.54%, p=0.2; mRS, 4.8±1.5 vs. 4.8±1.3, p=0.9). However, in the group undergoing neuroimaging, normal scans were associated with better survival to hospital discharge (79% vs. 21% p<0.001) and lower mRS (4.4±1.3 vs. 5.5±1.0, p=0.002). 12 (20.6%) of survivors who did not get neuroimaging and 3 (8%) of those who got neuroimaging achieved abilty to perform independent ADL at discharge (p=0.08). All 3 patients with ability to perform independent ADL in neuroimaging group had normal scans. 20 (34%) of survivors who did not get neuroimaging and 8 (21%) of those who got neuroimaging were discharged home (p=0.1). All 8 surviving patients in the neuroimaging group who were discharged home had normal scans. Secondary brain injury in ECMOsurvival and functional outcomes. A prospective study has been planned to better understand mechanisms mediating this effect. Neurocrit Care (2016) 25:S1-S310 RE-VERSE AD is an ongoing, phase 3, cohort study evaluating the extent to which idarucizumab, a humanized Fab fragment specifically reverses dabigatran's anticoagulation effect in patients with serious bleeding or requiring urgent interventions. Dabigatran is a direct acting oral anticoagulant approved for -valvular atrial fibrillation and venous thromboembolism treatment and prevention. In RE-LY, dabigatran 150 and 110 mg BID were associated with significantly lower annualized rates of intracranial hemorrhage (ICH) than warfarin (0.32%, 0.23% and 0.76%, respectively). Nonetheless, the mortality rate with ICH in the context of any anticoagulation remains high, probably reflecting the effect of hematoma expansion. Whether idarucizumab improves clinical outcome in dabigatran-treated patients This analysis of the first 90 patients enrolled in RE-VERSE AD focuses on patients with ICH. Patients presenting with ICH were given intravenous idarucizumab 5 g as two 2.5 g bolus infusions administered effect, based on central laboratory determination of dilute thrombin time (dTT) or ecarin clotting time (ECT). We compared the clinical outcome of this RE-VERSE AD interim analysis with dabigatran-treated ICH patients in RE-LY. In this interim analysis, 18 patients with dabigatran-associated ICH were enrolled in RE-VERSE AD. Complete reversal of anticoagulation was observed by dTT and ECT within minutes of idarucizumab administration. Preliminary results from this interim analysis indicate that the mortality rate of ICH patients RE-LY. Idarucizumab reversed anticoagulation in ICH patients and appears to improve mortality rates in dabigatran-treated patients with ICH versus historical controls from RE-LY. Results from additional patients in RE-VERSE AD will provide further information on the effects of idarucizumab reversal in patients with ICH. ****Permission was not granted to print this abstract**** S162 Neurocrit Care (2016) 25:S1-S310 Callie Electroencephalography (EEG) has clinical and prognostic importance for comatose survivors of cardiac arrest. Recent interest in quantitative EEG (qEEG) analysis has grown. The qualitative effects of sedation ing effects of sedatives on qEEG are poorly characterized in anoxic injury. We hypothesize that sedation would decrease amplitude-integrated EEG would predict neurological recovery. We routinely monitor comatose post-arrest patients with EEG for 48this prospective study, we included consecutive EEG-monitored patients who had protocolized sedation interruptions, excluding those with contraindications to interruption such as seizure or hemodynamic instability. We used Persyst v12 to quantify SR, aEEG, and ADR and calculated medians for 10min immediately prior to sedation interruption and the last 5min of interruption. We used nonparametric tests to determine if the qEEG signal changed pre-to post-and whether this differed by outcome (Cerebral Performance Category 1-2 at hospital discharge vs 3-5). Of 101 screened subjects, 22 met inclusion criteria (median age 58 years, 73% male). Sedation regimens varied (18 propofol; 13 fentanyl; 5 midazolam). Median duration of sedation interruption was 35min, and did not differ by sedative type. Pre-interruption, higher ADR and aEEG and lower SR predicted favorable outcome. Post-interruption, SR decreased (median change -0.85, IQR: -12.0 to 0), aEEG increased (0.38, (P=0.01), but aEEG and ADR changes did not differ by outcome. In acute anoxic brain injury, sedation increases SR and decreases aEEG. Larger decreases in SR with sedation interruption predict worse outcomes, which may reflect a susceptibility of deafferentated cortex to suppress in response to sedation. Ashley R. Hedges, Gary G. Davis, Brianne B. Wolfe, Erin E. Lingenfelter, Gregory G. Hawryluk, Safdar S. Ansari. University of Utah Hospital and Clinics, Salt Lake City, UT, USA. For patients presenting with subarachnoid hemorrhage (SAH) or traumatic brain injury (TBI), levetiracetam has begun to emerge as a preferred alternative to phenytoin for seizure prophylaxis following initial presentation. However, the optimal dose of levetiracetam has not been determined. A retrospective review of electronic medical records identified patients that received levetiracetam for seizure prophylaxis for SAH or TBI in a level one trauma center from May 12, 2014 to October 30, 2015. The goal of this research was to quantify the combined seizure incidence (including both clinically observed seizures and those confirmed by electroencephalogram) in patients receiving levetiracetam 500 mg twice daily compared to levetiracetam > 1000 mg total daily dose. Among 139 patients captured, 50% were male, with a mean age of 53 years old. For patients receiving seizure incidence was observed, 11.5% vs 1.3%, in patients receiving levetiracetam 500 mg twice daily This was observed despite no difference in potential confounders, includ trend towards increased levetiracetam failure rates was observed in the lower dosing scheme of 500 mg twice daily. T electroencephalograms ordered (22 vs 16, p=0.18). No difference in adverse effects were observed Our project suggests that patients may benefit from a standardized levetiracetam dosing scheme of 1000 mg twice daily. Treating intraventricular hemorrhage (IVH) with a fibrinolytic (IVF) therapy such alteplase via a catheter is becoming an increasingly popular intervention. With the conclusion of the CLEAR III trial a larger cohort of patient data is available to update past meta analyses. Mortality and good functional outcome after thrombolytic treatment was reviewed in patients with hypertensive IVH. A literature search was performed from 1970 to March 2016 to collect all literature on IVH treatment with IVF. Seventeen papers meeting our inclusion and exclusion criteria were collected for further analysis. IVH patients with vascular abnormalities and traumatic injuries were excluded. Mortality and functional outcome was assessed to compare IVF treated and control groups in all publications. There was a significant difference in mortality favoring the treatment group at 30 days, 180 days and 365 days (95%CI, 0.43-0.75; P=0.000; 95%CI, 0.45-0.82; P=0.001; 95%CI, 0.50-0.90; P=0.008). Pooling mRS and GOS good functional outcomes, there was a significant difference favoring the treatment group at 30, 90, and 180 days (95%CI, 0.80-0.96; P=0.003; 95%CI, 0.41-0.68; P=0.000; 95%CI, 0.73-0.96; P=0.012). There was a non-significant trend in mRS scores favoring the treatment group. A significant difference in GOS score favoring the treatment group was found at 30, 90, 180, and 365 days (95%CI, 0.61-0.95; P=0.016; 95%CI, 0.36-0.70; P=0.000; 95%CI, 0.52-0.88; P=0.0003; 95%CI, 0.54-0.95; P=0.019). Treatment of hypertensive IVH with thrombolytic may improve functional outcome and reduce mortality compared to control groups as early as 30 days, a trend that continues to 365 days for mortality and 180 days for functional outcome. Different effect sizes are generated when different functional outcome tools, such as mRS and GOS, are used. An aging population and increasing use of anticoagulants and antiplatelet agents to prevent ischemic hematoma (sSDH) and SDH related to trauma (tSDH). We sought to study the association of antiplatelet agent and warfarin use in SDH patients admitted to our neurosurgical ICU. warfarin were studied. Neurosurgical methods of SDH evacuation and re-evacuation were studied as well as hemostatic factors such as international normalized ratio (INR) within the first 24hrs, blood products and hemostatic agents given to reverse coagulopathy or antithrombotic effects. Demographic information such as age, gender and comorbidities were noted and indication for antithrombotic agent. We excluded major trauma (Level 1) associated with SDH at our center. From January 2006 to May 2016, we admitted 188 SDH patients, of which 46.8% required evacuation (88 ->12, mean = 2.9), and 58 on (33%) underwent redo evacuation (range in days 1-(date range 1-10 days). 24 hour INR rates in Warfarin related SDH were all <1.5 except for 6. Of these 6, only one patient required re-operation. Mort In our patient population, SDH was associated with a need for evacuation in 46.8% of all patients, of which a higher rate of reagents. Mortality was also higher in the warfarin associated SDH patients. Consecutive patie complications were prospectively enrolled. Medical critical care attendings (MICU), neurocritical care attendings (NICU), residents (RES), and nurses (RN) predicted the following: 1) 6-month functional -month quality of life (QOL). Patients were followed up at 6 months and their functional status and QOL were compared to the predicted values. Functional outcomes were dichotomized to good (mRS 0-3) vs. poor (mRS 4-6). 179 (of 365) patients had 6-month mRS predicted by all 4 provider groups. Fifty-four (30%) patients had good outcome and 125 (70%) had poor outcome. The MICU, NICU, RES, and RN providers had similar predictive values (95% CI) for accurately predicting good outcome (65% (52-76), 62% (51-72), 70% (55-83), and 64% (50-76), respectively). NICU was most accurate in identifying poor patient outcome, 97% (91-99), followed by MICU 89% (74-86), RN 86% (78-91), and RES 83% (75-89) (p=0.025, 0.005, and <0.001, respectively). When patients who transitioned to comfort measures only (n=65) were excluded from the analysis, the NICU team was more accurate at predicting poor outcome. Fifty-three survivors had QOL predicted by all 4 provider groups. The accuracy of QOL predi Neurocritical care attendings are better than healthcare providers without neurological training at predicting poor 6-month functional outcome in neurocritical care patients. However, the overall predictive accuracy for 6-month mRS and QOL was similar between healthcare provider teams. There are significant limitations in providers' ability to predict long-term functional outcomes. Patients with severe acute brain injury (SABI) raise important palliative care considerations associated with sudden, devastating injury and uncertain prognosis. The goal of this study was to explore how family members, nurses and physicians experience the palliative and supportive care needs of patients with SABI receiving care in the neurosciences intensive care unit (neuro-ICU). Design: Semi-structured in-person interviews were audiotaped, transcribed, and analyzed using thematic analysis. Setting: Thirty-bed neuro-Subjects: Forty-seven interviews were completed regarding 15 patients receiving care in the neuro-ICU with family members (n=16), nurses (n=15) and physicians (n=16). hope varied depending on the par away, generally in the process of conveying prognosis, while families expressed hope as an action that supported coping with their loved one's acute illness and its prognostic uncertainty. (2) Participants described the loss of personhood through brain injury, the need to recognize and treat the brain-injured patient as a person, and the importance of relatedness and connection, including personal support of families by clinicians. In their pursuit to recognize and preserve personhood, physicians used stories from patients and families to inform them about patient identities, while nurses focused on providing supportive, empathetic care to patients and families. Support for hope and preservation of personhood challenge care in the neuro-ICU as identified by families and clinicians of patients with severe acute brain injury. Specific practical approaches can address these challenges and improve care to meet the needs of patients and families in the neuro-ICU. Despite increasing evidence that early mobilization strategies are effective, we showed patients were not adequately mobilized in two Argentinean hospitals. We implemented a progressive-mobility protocol and examined its feasibility, safety and applicability in our neurocritical patients. Prospective observational implementation study of a progressive-mobility protocol for neurocritical patients admitted to ICUs of two university hospitals in Argentina. All patients were evaluated twice daily for level of movement and clinical stability. Patients progressed as tolerated from passive movement implementation baseline to ICU patients admitted in months 3-4 after implementation began (two month start-up phase not analyzed). There were 33 pre-implementation patients with 348 assessments (40% post-operative, 28% traumatic -implementation patients with 444 measurements (55% postpopulations: median age 53 years (95% CI 44-57), nearly 60% were men. Mobilization was 16.4 times -27.4) after protocol implementation. Two thirds of pre-implementation patients (65.2%) were not mobilized compared with only 9.2% post-implementation (p<0.001). Among mechanically ventilated patients, 82.2% of pre-implementation assessments showed no mobilization vs. 8.6% post. Post-implementation patients with an endotracheal tube had a lower rate of mobilization (78.9%) than ventilated patients with a tracheostomy (94.3%). Passive movement, turns and full assistance to sit up and transition out of bed to chair was achieved for 64.2% and 26.6% achieved higher levels (3, 4, 5) . Mobility sessions with the physical therapist were < 20 minutes in 90% of the cases. No mobility-related adverse events occurred. mobilized following protocol implementation. This prospective study demonstrated that early and progressive mobility among neurocritical care patients in Argentina is feasible and safe. Psychiatry, geriatrics, and oncology have adopted comprehensive approaches to predict outcomes accounting for important constructs such as spirituality and resilience. Critical illness often occurs as a sudden catastrophic event leaving patients with significant long-term cognitive, behavioral and neurological disturbances. Impact of resilience and spirituality on recovery in this setting has not been investigated. We have designed a study to validate two important scales, Connor Davidson Resilience Scale-10 and Brief RCOPE Spirituality scales for surrogate responders. hours with one or two surrogate responders will be included. This prospective cohort study will collect demographic, laboratory and radiographic data in a RedCap database. For every patient enrolled, the CD-RISC-and a behalf, themselves, and for each other. Each patient will complete two resilience and two spirituality scale CD-RISC and three Spirituality scales . The scales will be administered to the patient ,if possible, prior to discharge; at 3 months and at 12 months. If patient ing followup. CARS study has screened 44 patients in 35 days, enrolling 30 patients. Common diagnoses include subarachnoid hemorrhage (8), CNS malignancy (5), intracranial hemorrhage (3), unruptured aneurysm (3), subdural hematoma (2). Based on current enrollment, this unique methodology for surrogate validation of scales is feasible. By August 2016, an anticipated 150 subject will be recruited. Surrogate validation of quantitative measurements of resilience, spirituality can provide new insight into prognostication and patient centered critical care. Active family engagement in the intensive care unit (ICU) could improve patients' and families' experience with care, interactions with the healthcare team, and outcomes. This study examined the perceptions and attitudes of family members regarding increased engagement with passive mobilization of neurocritical care patients. An educational video on passive mobilization of ICU patients was developed to engage family members to participate as valued members of the healthcare team. An anthropologist and a nurse or physician (study team) invited family of neurocritical care patients in an academic medical center ICU in Argentina their perceptions of engagement with care. A multidisciplinary team (3 MDs, 2 RNs, 2 anthropologists) completed the analysis. Thirty-two family members (53% female) of 16 ICU patients participated. The study team observed and the family participants reported to be positively surprised by: the format of learning by video that was different from how they usually received daily information; the information in the video was more detailed than anything previously taught; the invitation to watch the video and engage in patient's care was new and unexpected as was the opportunity to provide opinions on how to improve the video instruction. A newly authorized them to touch the patient and participate in care. They reported increased hope about prognosis and perceived an improved relationship with the healthcare team. Only one family member considered the video irrelevant. Family members' attitudes and perceptions toward this low cost approach to engagement were positive. This approach to teaching and engagement may help humanize the complex ICU environment. 70-80% of Americans believe in the concept of miracles. We hypothesize that a belief in miracles leads to consultation rate of palliative care. Addressing a patient or family's belief in miracles and understanding what a miracle signifies early in the Neuro- The Critical Illness Resilience and Spirituality (CARS) study is a prospective cohort study currently -ICU at Mount Sinai Hospital with an expected length of stay of at least 48 hours and surrogate responders. As part of the study, all recruited patients and their families primary outcome is tracheostomy and PEG tube placement. Secondary outcomes include length of stay, full code status, and palliative care consultation. These groups will be matched with regards to the disease specific sever regression will be used to compare rates of the primary and secondary outcomes. The CARS study has screened 44 patients screened and enrolled 30 patients so far. Of those enrolled, the most common diagnoses include subarachnoid hemorrhage (8), CNS malignancy (5), intracranial hemorrhage (3), unruptured aneurysm (3), subdural hematoma (2). This is an ongoing study, we anticipate recruiting 150 patients by the end of August, 2016. Belief in miracles could potentially influence continuation of aggressive measures in a shared decisionparadigm in the Neuro-ICU. Mary M. Barden, Teddy T. Youn, Carolina C. Maciel, Sonya S. Zhou, David D. Greer. Department of Neurology, Yale-New Haven Hospital, Yale School of Medicine, New Haven, CT, USA. Withdrawal of life-sustaining therapy (WLST) for predicted poor neurological outcome is a common cause of death among post-cardiac arrest patients. Recent guidelines recommend against WLST before 72 hours post-arrest. Early WLST perpetuates a self-fulfilling prophecy that may contribute to premature death in some patients who otherwise would have survived with good neurological recovery. A retrospective cohort of resuscitated cardiac arrest patients from January 2011 to March 2016 at a single tertiary academic medical center was reviewed. Patients were evaluated for outcomes at hospital discharge and (when applicable) the timing of and reason for WLST. Prognostic indicators including clinical examination, electrophysiology, and neuroimaging were analyzed and findings were compared to day of WLST. Of 144 patients, 64 (44%) had WLST due to perceived poor neurological prognosis. Median day of WLST for this reason was post-arrest day 6. When stratified according to treatment with targeted temperature management (TTM), the median day of WLST remained day 6 for both TTM-treated and non TTM-treated groups. Of patients with WLST, the phrase "no chance for meaningful recovery" was used in documentation for 23 (36%), MRI results were cited as indicative of poor neurological prognosis for 16 (25%), and pupillary light reflex was present day 3 post-arrest (or day 3 post-complete rewarming) in 35 (55%). In a retrospective cohort of resuscitated cardiac arrest patients, WLST for predicted poor neurological outcome was the most common cause of death. The median day of WLST was post-arrest day 6. Many patients with WLST had present pupillary reflexes on day 3 post-arrest (or day 3 post-complete WLST in the setting of indeterminate prognostic indicators undermines accurate neurological prognostication of post-cardiac arrest patients and perpetuates a self-fulfilling prophecy of poor outcome. Brain Injury Global Hypoxic Ischemic Brain Injury (HIBI) is a major cause of death and disability worldwide. Invasive monitoring of brain function enables goal-directed treatment strategies that optimize cerebral physiology, reduce secondary brain injury (SBI), and potentially improve outcomes. We report a series of patients with HIBI where intracranial monitors were placed to guide clinical management. Retrospective analysis of 7 patients with HIBI cared for at a large academic center over a 3 year period. All patients received therapeutic hypothermia (TH) to 33°, continuous EEG monitoring, and had a bundle of invasive monitors placed through a multi-lumen cranial bolt. The full bundle consisted of an ICP monitor, brain oxygen (PbtO2) monitor, cerebral blood flow (CBF) probe, and cerebral microdialysis probe. 4 patients received the full bundle, while the others received a partial bundle. Patients were treated using a tiered algorithm designed to optimize cerebral physiological parameters. Precipitants of HIBI included cardiac arrest (5 patients), airway occlusion during anesthesia induction (1 patient), and hanging (1 patient). Mean patient age was 45 years. Average time between initial injury and probe placement was 37 hours. Average duration of monitoring was 6.7 days. No adverse events occurred after monitor placement. Episodes of deranged cerebral physiology-including intracranial hypertension, brain hypoxia, cerebral glycopenia, metabolic crisis, and reduced perfusion leading to treatment changes occurred in 6 of 7 patients. They occurred up to 10 days after initial injury, and in all cases would have otherwise been clinically silent. 4 of 7 patients died in the hospital. The surviving patients all regained consciousness and were discharged to Acute Rehabilitation facilities. We did not find Invasive intracranial monitoring after HIBI appears safe and identifies physiological states associated with SBI. Goal directed treatment utilizing multi-modality monitoring in HIBI merit further study. is associated with worse patient outcomes; however, it can be difficult to reliably detect. Delirium prevention is therefore a potentially beneficial strategy and is most effective in patients who are at high to evaluate whether the Advanced Practice Providers (APP's) would both use the DPS and also find the DPS easy to use. During a 6--bed Neurocritical Care Unit at a large -stratify consecutive admissions of patients with AIS a descriptive statistics. The APPs completed a 14-item questionnaire that included the System Usability Scale (SUS) and open-ended questions to determine the usability of the DPS, as well as to assess for facilitators and barriers for the use of the DPS. No individual patient data was collected. Patients admitted with AIS and ICH (n=20) were assessed by the APP's using the DPS. Compliance with DPS use was 90 of APPs (n=2). The SUS score (76.7) was mid-point between "acceptable" and "excellent." Facilitators and barriers for use of the DPS were identified. The DPS was easy to use and was consistently used by the APP's. Adoption of the DPS with this patient population can be a first step to identify the most atthis vulnerable population. The intensive care unit is a complex learning environment with variability in a number of external factors. Prior studies of neurology residency training in the neurological intensive care unit have focused on general exposure. This study aims to evaluate resident perception of neurocritical care training. An online survey was sent to program directors and neurocritical care members for distribution to neurology residents. The survey consisted of 20 free-text or selection style questions that focus on resident perception of neurocritical training. Statical analysis for group differences was completed with t or Fisher exact tests A total of 95 responses (4.7% response rate) was obtained. Of those responders, 54 completed a freetext question regarding needed improvements to neurointensive care training. 53% responded with needs for educational changes, and these responders did not differ from other responder in average required practitioners (61% vs 81% p=0.15), and neurocritical care attendings (97% vs 93% p=0.59). This is the first study to examine neurology residents' concerns with neurocritical care rotations. There is little neurocritical care educational materials focused to neurology residents, but the emergency assess the neurocritical care educational training priorities during neurology residency are warranted. Our institute had several cases of conflict come to light in the evaluation of patients being evaluated for death by neurological criteria. Provider understanding and awareness of clinical guidelines was found to be low across all sub-specialties. It was deemed important to follow appropriate procedures based on published guidelines and a standardized process to provide appropriate care for each patient, optimize ICU resource utilization and strengthen provider and public trust. Due to medical, legal and ethical issues involved , an institutional standard was called for . -specialties about discrepancies, we revised the institutional policy to reflect emphasis of educational gaps and reflect the latest published guidelines and practice updates . We created an education module , a standardized template in electronic medical to allow escalation in case of conflicts . The project led to increased participation and satisfaction amongst the clinical providers in the ICUs when n education source to evaluate patients with a consistent approach based on published practiced parameters. We observed a trend in decrease in length of stay and variance for brain dead patients since donation referrals as well timely initiation and effectiveness of family discussions in irreversibly confirming this is in a survey model. A standard care pathway towards evaluation of patients with death by neurological criteria can be successfully implemented at an institutional level in a tertiary care academic medical center. Establishing and maintaining optimal brain perfusion is a crucial endpoint for resuscitation and postcardiac arrest care. A recently FDA-approved device that employs laser and pulsed Doppler now provides clinicians with the Cerebral Flow Index (CFI), a non-invasive measure of brain perfusion. We sought to determine if CFI provided by the ORNIM C-Flow device can be used as a simple and valid measurement of brain perfusion after resuscitation in cardiac arrest patients. We performed a single-center prospective observational inception cohort study of adult patients with cardiac arrest starting in October 2015. Comatose patients with sustained return of spontaneous circulation (ROSC) within 60 minutes of maneuvers were included. The ORNIM CFlow was connected as soon as feasible after ROSC. Clinicians were blinded to CFI values. Primary outcome was survival at discharge and secondary outcome was neurological assessment using the Cerebral Performance Categories (CPC) scale at discharge. A total of 24 patients have been enrolled as of May 2016. Half (46.1%) were out-of-hospital arrests and neurological outcome (CPC 1 or 2). Mean interval between arrest and start of monitoring was 15 hours with a mean duration of 47 hours. Adequate signal was available 83.4% of the monitoring time. Mean CFI in survivors was 62.3, compared to 54.6 in non-survivors (P value 0.07). Patients with good neurological outcome at discharge also had a higher mean CFI, although the small sample size precludes any conclusion. Our results demonstrate that cerebral perfusion monitoring using the ORNIM CFlow after cardiac arrest is feasible. It also suggests that higher CFI might be associated with survival at discharge. As enrollment progresses and more data are collected, further insight on the potential role of CFI as a neuromonitoring tool might emerge. Tuberculous meningitis (TBM) is the most devastating form of tuberculosis, yet rates of neurological complications and mortality are uncertain in high-income countries. We used administrative claims data on all admissions at nonfederal hospitals to identify adult patients with TBM in California between 2005 --2013 , and Florida between 2005 -2013 . Our outcomes of interest were mortality and the fo seizure, hydrocephalus requiring a ventriculoperitoneal shunt, vision impairment, and hearing impairment. Kaplan-Meier survival statistics were used to assess the cumulative rates of neurological complications and death. We identified 806 patients with TBM, of whom 54.8% (95% CI, 50.9-58.7%) developed at least one neurological complication or died. More than two-thirds of these complications occurred during the initial hospitalization for TBM. Individual neurological complications were not uncommon: the cumulative rate of -16.9%), the rate of seizure was 18.8% (95% CI, 15.4-22.8%), and the rate of ventriculoperitoneal shunting was 8.4% (95% CI, 6.4-10.9%). Vision impairment occurred in 21.6% (95% CI, 18.5-25.1%) of patients and hearing impairment occurred in 6.8% (95% CI, 4.9-9.4%). The mortality rate was 21.5% (95% CI, 18.4-24.9%). nd death even in high-income countries such as the United States. Neuropalliative Care in Peru: Emergence from the Conspiracy of Silence The development of palliative care in Peru remains limited, particularly for non-oncologic services such as palliative and end-of-life care in patients, families, nurses and physicians in a specialized neurological institute in Lima, Peru. We used a mixed methods approach consisting of 70 surveys and 21 qualitative, semi-structured interviews that were recorded, transcribed and analyzed using thematic analysis. Surveys identified a substantial need for palliative care in the neurological institute (63% of doctors and of do emerged from qualitative interviews evolved around communication about end-of-life choices in neurologic disease. Knowledge about advance directives was limited among both clinicians and families, and prognosis, and who should tell them. However, the perception that a physician should be honest, and that suffering and pain should be avoided at all times was unanimous. Barriers to transparency in patientphysician communication included (1) expectation of cure with medical treatment; (2) families' trust in God training in communication, symptom management and end-of-life care; and (4) a paternalistic culture. Participants identified several challenges specific to palliative care in neurologic disease. In a country without a palliative care training program and no legal basis for advance directives, families and clinicians are emerging from a culture of silence about serious diagnoses and end-of-life care choices. Our findings emphasize the need for palliative care education for neurology providers and the public in Peru. A trained physician must perform the brain death examination in a systematic fashion in order to recognize and prevent potential sources of error. Given the infrequency at which brain death presents in a hospital setting, clinicians may not always have the opportunity to observe a brain death examination during their training. In this study, we plan to evaluate the effect of medical specialty and expertise on documentation errors. We performed a retrospective chart review of 118 brain death examinations between Jan. 1 2014 to July 31st 2015 at the University of Pittsburgh Medical Center Presbyterian. Physician specialty and training level, documentation errors, and confirmatory tests such as cerebral blood flow (CBF), electroencephalography (EEG), and computed tomography angiography (CTA) were collected from medical records. exams. ams carried out. The most common completion of documentation by at least one examiner. Attending physicians, residents and fellows were responsible for 47% ( Neurology and Neurosurgery residents have limited exposure to the brain death examination. Regardless Communication with patients and their families is of central concern in healthcare. However, evidence shows that it is often poorly addressed, especially at times of rapid health status changes and periods of clinical uncertainty. Acute neurological emergencies pose an inherently unique challenge in communication. While emerging studies have addressed communication gaps and strategies to improve them in various critical care settings, none have assessed this issue in acute neurological emergencies Ongoing IRB approved prospective observational study in a 24 bed Neurocritical care unit in tertiary care academic medical center All patients admitted to the unit and all clinical providers participating in their care screened for inclusion. Direct observation of discussions between clinical providers and families by a questions addressing the satisfaction, understanding of treatment options, impact on health care decisions and ways to improve communication Five patients have been enrolled in the study, so far. The results are analyzed for concordance between tween answers is considered between all participants for general satisfaction with the communication (as well as family's understanding of treatment options explained to them by the physician. Some disagreement on the impact of the discussion on health care decisions. Qualitative domains identified by families as areas of good ical radiographic images. Domains identified as needing improvement included explanation of medical circumstances and need for private room for discussions. We observed general satisfaction with communication. Further enrollment will help elucidate any definitive areas of improvement and impact of communication on health care decisions. Laith Maali, Sheema S. Khan, Mahmoud M. Ismail, Rhys R. Brooks, Vishnumurthy V. Shushrutha Hedna. The University of New Mexico, Albuquerque, NM, USA. Cerebral venous thrombosis (CVT) usually accounts for < in their demographics, etiology, clinical features, radiological presentation, and mortality have not been previously explored. A systematic search was performed for publications in PubMed usi thrombosis", "Cerebral vein thrombosis" and "Cortical vein Thrombosis". A total of 600 relevant studies were abstracted with strict selection criteria and a total of 7048 patients' data were used for the final analysis. Linear correlation was used for our descriptive analysis. Cases reported were Europe-3152, Asia-2722, North America-852, Africa-122, Australia-121 and South America-79. Overall Male to female ratio was 1:2.2, among clinical characteristics headache was the most common symptom and hematological factors were the most common etiology. Location of the thrombosis was described mostly in the transverse sinus. Intercontinental differences in relation to demographics, etiology, clinical features, radiological presentation, and mortality were identified. CVT can have significant disparity in their demographics, etiology, clinical features, radiological presentation, and mortality when compared from one continent to another. It is important for the worldwide physicians to recognize these differences and to follow the most recent guidelines, diagnostic methods and treatment to insure the best outcome and prognosis. Timely communication is critical for high quality care in the intensive care unit(ICU). Published literature in surgical ICUs quotes up to 81% of patient caregivers receive prognostic information with mean prognostic interval 1.7 ± 2.8 days since ICU admission Prognostication in acute neurological injuries is challenging and uncertainty may delay communication. We assess occurrence and timeliness of goals of care communication in a Neurocritical Care unit Prospective observational study by surveying nurses in a 22 bed Neurocritical care unit in tertiary level eriod. Data was also collected during daily morning multidisciplinary huddle and verified by verbally surveying the nurses. Survey results were analyzed for 97 patients and 185 patient encounters. In 99.4% encounters , the nurses felt the patient's treatment plan matched patient-centered goals of care In 79.3% encounters, a provider family discussion had occurred in the last 24 hours. Within the 1st 24 hours of ICU admission, 40.2% patients were identified to need goals of care discussion in the multidisciplinary huddle, only 61.5% had such a discussion. For patients needing goals of care addressed, a discussion occurred on an average 1.3 ± 0.6 days since ICU admission. Dichotomized by age, 73.7% patients younger than 65 years old had a discussion , if one was needed, while only 50% older than 65 years had one. When dichotomized by gender, 66.7% of males and 81.4% of females had a provider discussion. 68.4 % females compared to 55% of males received a discussion on goals of care if identified as needed within 1st 24 hours of ICU admission. Our data shows timely communication of goals of care in the Neurocritical care unit with a mean time comparable to published literature. However, there appear to be demographic disparities that warrant further research. Cerebral Vasomotor reactivity reflect prognosis after cardiac arrest SungEun Lee. Ajou University School of Medicine / Department of Neurology, Suwon, Korea, Republic of. Neurological prognostication after cardiac arrest is a difficult problem. Since several studies reported good effect of target temperature management (TTM), prognostication after cardiac arrest was delayed and became complex. Recently, some reports presented that impaired cerebral autoregulation was correlated with neurologically poor outcome. The aim of this study was to determine whether vasomotor reactivity (VMR) test by transcranial doppler (TCD), reflecting cerebral hemodynamic status, affected accuracy of neurological prognostication in post cardiac arrest patients. Since January 2016, 8 patients were enrolled after cardiac arrest. Patients who performed VMR test during TTM period were included and patient with unstable vital sign or malignant findings in brain CT, such as massive subarachnoid hemorrhage or severe brain edema, or poor temporal windows. Primary outcome was cerebral performance category scale (CPC) at discharge. VMR test used breath-holding method during 40 seconds. Carbon d analysis. Other conventional prognostication test, such as EEG, SEP, et al., was performed after 72 hours from rewarming time. We divided patients between good (CPC 1-2) and poor (CPC 3-5) outcome group and compared results from prognostic test between two groups. potential, and electroencephalography after 72 hours from rewarming time were presented favorable results in good outcome group. (p<0.001) VMR during breath-holding technique during TTM period also was more increased in good outcome group at right (41.10 ± 1.15 % vs. 12.17 ± 3.01%, p<0.001) and left (36.47 ± 6.48% vs. 7.23 ± 10.65%, p<0.001) middle cerebral arteries. The present study shows that vasomotor reactivity is preserved in patients with neurological good outcome. To evaluating cerebral hemodynamic status by VMR test seems to be useful tool for early prognostication after cardiac arrest. Michelle L. Lozano, Susan S. Yeager. The Ohio State University Wexner Medical Center, Columbus, OH, USA. As the numbers and opportunities for Advanced Practice Providers (APPS) in Neurocritical Care Units (NCCUs) has increased, the integration of these providers into the health care setting has become a greater challenge. Currently no data exists to support h comfort levels before and after completion orientation. This prospective, pre and post observational study was sent to 10 newly hired NCCU APPs within an Academic Medical Center. A one hundred-item survey was created to evaluate self-reported experience CUs. Baseline data was collected from each APP. Next, APPs were integrated into the NCCU utilizing a three month orientation program which fused a series of didactic, simulated, and precepted experiences. After e survey tool. Student's t test statistics were utilized to compare before and after experience and comfort levels with items identified as necessary to perform in the NCCU APP role. As utilization of APPs in the NCCU becomes more prevalent, integration processes need developed to practice. A structured approach enables identification of high priority areas to assist with initial and and comfort levels. Results indicate that further education and exposure to items such as neurologic imaging may be helpful. Limitations of this study include subjective data from a small, self-reported, single institutional sample. Further research of larger, more diversified sample representation is needed to validate whether these results can be generalized to other NCCUs. Yasuhiro Kuroda, Kenya K. Kawakita, Toru T. Hifumi. Department of Emergency Medicine, Kagawa University, Miki, Japan. Brain damage after return of spontaneous circulation (ROSC) varies among studies and patients despite an established modality enabling proper evaluation. Evaluation of brain injury after ROSC is needed for the determination of the inclusion criteria of neurocritical care, especially of targeted temperature management. Literatures are reviewed and summarized. The association between admission Glasgow Coma Score (GCS) motor score and neurologic outcome after ROSC (day 0) is an independent predictor of good neurologic outcome at 90 days in patients sustaining out-of-hospital cardiac arrest who receive therapeutic hypothermia: GCS motor score 1, n=130 (52.2%); score 2-3, n=23 (76.7%); score 4-5, n=20 (87.0%), P<0.01 (Hifumi 2015 2201). Recently no significant differences of neurologic outcome at 30 days after hospital admission was observed between mild therapeutic hypothermia and control in the subgroup of GCS Motor score 5 or 6. These data show that initial GCS motor score examination immediately after ROSC can at least provide baseline objective prognostic data for decisions by healthcare professionals. Neurological signs such as GCS, brain stem reflex, respiratory status, and degree of shivering are potential variables that can be incorporated into a predictive model for a more precise evaluation of brain injury in cardiac arrest survivors undergoing TTM. Effect of targeted temperature management should be evaluated depending on the brain injury in PCAS. Cydni N. Williams, Jennifer J. Wilson. Oregon Health and Science University, Department of Pediatrics, Portland, OR, USA. -level estimates of ET utilization in pediatric AIS, and explore demographic and clinical characteristics, associated interventions, and outcomes. Retrospective cohort analysis of the Kids' Inpatient Database evaluated ET utilization in children with AIS and age >28days, identified by diagnosis and procedure codes. Analyses were weighted for national estimates and compared with Chi-square and t-tests. Among 3184 pediatric AIS patients, 38 (1%) received ET. Anterior circulation occlusions were seen in 89% of ET patients. ET patient age ranged 2versus 30%, p<.001) was more common and seizure was less common (11% ET versus 31%, p=.004) in ET patients. Average age was higher with ET (14 versus 10 years, p<.05). Other patient demographics, hospital characteristics, and critical care procedures were similar. Thrombolytic agents (tPA) were common with ET (51% ET versus 2% overall). Intracranial hemorrhage was similar (14% ET versus 8%, p= .12), and varied by tPA (22% ET with tPA, 10% tPA only, 7% ET only, 8% neither). There was a nonsignificant trend toward poor outcome (death, discharge to nursing facility, tracheostomy, or gastrostomy) was seen between poor outcome and ET (68% ET versus 39%, p=.004). ET in pediatric AIS is uncommon, utilized mostly in older children and those with paresis. Though hemorrhage was uncommon, this data suggests caution with ET and tPA combination. Associations between ET and poor outcome may reflect disease severity bias. More research on outcomes with ET in pediatric AIS is needed. Luis P. Lee, Michael M. Leoncio, Balagangadhar B. Totapally. Nicklaus Children's Hospital / Pediatric Critical Care Department, Miami, FL, USA. and cerebral edema is the most serious complication leading to morbidity and mortality. We queried a nationally representative database to determine epidemiologic data of cerebral edema in children with DKA. An analysis of the Healthcare Cost and Utilization Project's Kids Inpatient Database for the year 2012 was performed. The database was filtered using ICD-9 diagnosis codes for DKA (250.10, 250.11, 250.12, 250.13 ) and cerebral edema (348.5) from the age of 1 month to 20 years. We examined these procedures, outcome and mortality rates. Sample weighing was employed to produce national estimates. Chi-square test, Mann Whitney U test and binary regression analysis were performed using SPSS to analyze the data. A total of 38,972 patients with DKA were discharged during 2012. Females were 55%. Racial distribution -19) years. Cerebral edema was present in 188 (0.5%) children. The overall mortality rate was 0.08%, but the mortality rate in children who developed cerebral edema was higher at 6.7% (OR: 151; 95% CI: 73-310). Mortality was higher in children who had a major operative procedure (1.13% vs 0.07%; OR 16, 95% CI: 6-45) and in those with Medicaid compared to private insurance (0.11% vs 0.04% p=0. 031) and lower in number of chronic conditions, and hospital charges were significantly higher among non-survivors but there was no difference in the age. The overall mortality rate in children admitted with DKA is 0.08%. Cerebral edema prevalence is 0.5% and it increases mortality significantly. Mullai Baalaaji, Sunit S. Singhi, Muralidharan M. Jayashree, Arun A. Bansal. Pediatric Intensive Care Unit, Department of Pediatrics, PGIMER, Chandigarh, India. Near-Infrared Spectroscopy (NIRS), a non-invasive modality to measure regional cerebral oxygenation (rSO2), is being increasingly used to monitor cerebral tissue oxygenation. We studied relationship of rSO2 with cerebral perfusion pressure (CPP) and intracranial pressure (ICP) in children with acute CNS infections to determine if rSO2 could be used as non-invasive surrogate for CPP. In a prospective observational study we enrolled 31 children, aged < 12 years, with raised ICP due to acute CNS infections after approval by Institutional Ethics Committee. They were monitored simultaneously for rSO2 of both frontal--5100C, Covidien-IIc), invasive blood pressure, and ICP using intraparenchymal fibre-optic catheter (Codman). Linear trends and correlation coefficients were used to define relation of rSO2 with ICP and CPP. A total of 3378 paired values of rSO2, ICP and CPP were analysed. The linear trends during the first 72 hours revealed no significant correlation between changes in rSO2 and changes in ICP and CPP from baseline (R2=0.007,0.012 for ICP and CPP respectively). However, the trend was not uniform -48% patients had no correlation between rSO2 and CPP, 29% showed a positive correlation and 23% showed a negative correlation. Subgroup analysis revealed that strength of correlation between rSO2 and -0.421,p20 mmHg and normal CPP were 6.1(3.4-11.1,p50% respectively. rSO2 has complex interaction with ICP and CPP; the changes in ICP and CPP could not predict changes in rSO2. However, the odds for normal CPP was significantly higher when rSO2>50% and this cut-off could be used as a non-invasive target for age appropriate CPP. Refractory status epilepticus is persistent seizure activity despite treatment with one first-line and one second-line anti-epileptic medication, while seizure activity >24 hours is considered super-refractory. Functional outcome for children with these conditions is not well defined. This study describes functional outcome for children with refractory and super-refractory status epilepticus proposing that prognosis will be variable with high mortality. Survivors will be and technology dependence. This retrospective chart review evaluated children age 0-18 years who received pentobarbital infusion at Texas Children's Hospital pediatric intensive care unit from 2004-2015 for status epilepticus. Outcome was defined using pediatric cerebral performance category score (PCPC) at time of discharge and at the most recent clinical evaluation per the medical record. Additional measures included mortality, need for medical technology (tracheostomy or gastrostomy tube), seizure burden, and number of seizure medications at discharge. 42 children met inclusion criteria. In-hospital mortality was 31%, secondary to withdrawal of support (54%), brain death (31%), or cardiac arrest (15%). Highest mortality occurred in acute hypoxic ischemic injury (p= 0.04). Of survivors, 39% returned to baseline PCPC at discharge while 46% demonstrated tracheostomy and 7 children underwent gastrostomy tube placement. Seizures persisted at discharge for most patients with no prior frequency. Most children required additional home seizure medications. Long-term follow-up was documented for 22 survivors up to 9 years after discharge. 32% demonstrated improved PCPC and 18% showed decline including 3 additional deaths. Mortality in this population was high. Functional outcome in survivors was variable. Some children returned to neurologic baseline by time of discharge and for those who did not, continued functional improvement was possible over time. their s -injury. In addition to standard anatomic imaging, MR sequences obtained "often or always" included: diffusion--perfusionpediatric TBI subjects received an acute MRI within 30 days post-injury. Fifteen ADAPT sites, accounting for over 50% of ADAPT enrollment, committed to recruit ADAPT subjects for a non-sedated MRI scan at one year post -TBI. Conclusion: Collection of 400-600 acute MRI scans from the 1000 subjects enrolled in ADAPT to study associations between acute MRI findings and functional outcome is potentially feasible. Allowing for 20% mortality and 50% recruitment rate, recruitment of 100-150 ADAPT subjects from 15 ADAPT sites for a follow-up MRI to study relationships between advanced MRI measures and neurocognitive function is potentially feasible and would represent the largest such study conducted to date. Antimicrobial prescribing practices and antibiotic resistance following neurosurgical drain placement: A single-center observational study Andrea J. Passarelli, Hasan H. Alhasani. Christiana Care Health System Department of Pharmacy, Newark, DE, USA. The use of systemic antibiotics for the duration of neurosurgical drain placement has not been associated with reduced rates of drain related infection (DRI) and may contribute to the development of antimicrobial resistance and Clostridium difficile infection (CDI). We sought to describe antimicrobial prescribing practices, incidence of DRI, and development of antimicrobial resistance and CDI after neurosurgical drain placement at our institution. This was a single center study including adult patients status post ventriculostomy or ventriculoperitoneal shunt or Ommaya reservoir, and use of an antibiotic impregnated drain. Bacterial cultures and C. difficile PCR during the index admission and 90 days post-discharge were collected. Antibiotic resistance was defined as an organism resistant to the prophylactic agent. Prolonged prophylaxis was defined as antibiotics continued for >48 hours after drain placement. Eighty-one patients with 107 drains were included. The median duration of prophylaxis was 3.3 days and cefazolin was most commonly prescribed agent (85%). Three of 48 patients with EVDs developed DRI. prolonged vs. perioperative prophylaxis. Of non-DRIs 60% were resistant to the prophylactic agent used. E. coli, K. oxytoca, and S. aureus had higher rates of resistance to cefazolin compared to our institutional antibiogram, although not statistically significant. No patients developed CDI. The use of prolonged prophylaxis was not associated with a reduced reduction in DRI. Most bacterial isolates were resistant to the prophylactic agent used. We suggest that antibiotic prophylaxis for neurosurgical drain placement be limited to one preoperative dose within 60 minutes of the procedure The objectives of this study are to evaluate our institution's practice for initiating seizure prophylaxis postoperatively and establish a standard of care. Adult patients who underwent cerebrovascular surgery from August 2013 to July 2015 were screened for study inclusion. Patients who received LEV postoperatively were compared to those who did not receive LEV. Clinical seizures and data were obtained from retrospective review of electronic medical records. The primary outcome was seizure occurrence in the first 7 days after surgery. Secondary outcomes Of the 160 patients included in the study, there were 115 in the no LEV group and 45 in the LEV group. Two seizures occurred in the no LEV group while no seizures occurred in the LEV group (2 vs 0, p=0.373). There were no differences between surgery type, intraoperative blood loss or proportion of aSAH. Of the 25 patients with aSAH, 64% were not on LEV and 1 seizure occurred. Of 19 patients with intraparenchymal or intraventricular extension, 63% were not on LEV. Average length of stay was prolonged for the LEV group (12 vs 7 days, p<0.001). The majority of patients did not receive LEV postoperatively and there was no difference in seizure Developing a standardized approach for initiating LEV may decrease variability in practices and streamline postoperative care. Post-operative pain control after craniotomy: A meta-narrative review Craniotomy is commonly performed for the treatment of a variety of conditions including brain tumors, aneurysms, and vascular malformations. Despite significant advances in the quality and efficacy of neuroanesthetic care, there are no evidence-based guidelines for the management of post-operative pain after craniotomy. Uncontrolled poststay, increased hospital care costs, and poor health-dencebased clinical decision rules, clinicians often rely on institutional or expert-based opinions to guide their decisionon opioid use, there an urgent need to evaluate existing pain management protocols. Hence, we conducted a meta-narrative to evaluate heterogeneity in current practices regarding management of postoperative pain after craniotomy. A meta-narrative review was performed utilizing th terms "pain" and "craniotomy". A total of 344 articles and 16 systematic reviews were resulted. Inclusion criteria were studies from 2004-2016, randomized controlled trials, retrospective studies, systematic reviews, case reports, case series published in English were included. Of these, 230 articles and 7 systematic reviews were included in the final analysis. There is a paucity of randomized controlled trials to develop evidence based peri-operative pain management protocols in craniotomy patients. There is evidence to suggest that scalp infiltration with local anesthetic may improve post-operative pain scores immediately after surgery. The perioperative use of nonsteroidal anti-inflammatory medications may improve pain scores without a subsequent increase in Management of post-operative pain after craniotomy remains a challenging problem for clinicians and patients. There is an urgent need to conduct well designed randomized controlled trials to guide perioperative pain management in craniotomy patients and to use opioid sparing techniques for improving patient outcomes. Symptomatic plateau waves are characterized by paroxysmal neurological symptoms suggestive of elevated intracranial pressure such as depressed level of consciousness, pupillary dilatation, and dysautonomia in a patient with an intracranial mass lesion. Cli seizures, syncope, or new brain injury. Noninvasive cerebral blood flow can be measured using ultrasound-tagged infrared spectroscopy; continuous EEG is sensitive to changes in blood flow. We report two patients without invasive intracranial pressure monitoring who demonstrated changes in blood flow and EEG during symptomatic plateau waves. Case Series. Case 1 was a 24 year-old man with fungal ventriculomeningitis. After a prolonged hospital course, he developed an entrapped 4th ventricle and began to experience periods of complete unresponsiveness with anisocoria, clonus, and tachy-or bradycardia lasting between 10 and 20 minutes. Episodes resolved after decompression and ventricular stent placement. Case 2 was a 49 year-old woman with intraventricular meningioma who underwent partial resection with entrapment of the right lateral ventricle. On post-operative day 7 she developed multiple episodes of unresponsiveness, diaphoresis, clonus, tachy-or bradycardia lasting 10 to 20 minutes, culminating in a persistent episode requiring urgent craniotomy. In both patients, cEEG was started to assess for seizures and ORNIM device was used to characterize blood flow. In each, symptomatic plateau waves were accompanied by decreased blood flow, followed by attenuation of faster frequencies on the cEEG. Symptomatic plateau waves may be characterized noninvasively by using surface measurements of blood flow and cEEG. This case series demonstrates that decreases in bifrontal blood flow lead to depressions in cEEG during these symptomatic plateau waves. Noninvasive measurement of blood flow in conjunction with cEEG provides an adjunct to invasive ICP monitoring in patients with mass lesions at An open--SSEefficacy of SAGE-547, a proprietary formulation of allopregnanolone, in patients with super-refractory status epilepti line agents (TLAs) while SAGE-(none attributed by the Safety Committee to SAGE-547). Post-hoc analyses evaluated the pharmacological effects of SAGE-547 and the effect of SAGE-547 administration in the context of multiple antiepileptic drugs (AEDs), pressors, and TLAs. Here we examine the hemodynamic properties of SAGE-547 in the study patients, with the goal of further understanding the clinical context of SAGE-547 administration in this critically ill population. burst suppression we maintenance of the TLA. Key exclusion criteria were anoxic brain injury and very short life expectancy. At enrollment, patients received an average of 3.0 AEDs and 1.4 TLAs with an average status epilepticus duration of 9.2 days. Hemodynamic measurements (heart rate, systolic and diastolic blood pressure) were collected at screening, pre-dose, during SAGE-547 treatment (15, 30, 45, 60 minutes; 2, 4, 8, 24, 48, 72, 96 , 120 hours) and followwas examined. Twenty-five patients received treatment with SAGE-547. During the study, mean changes in hemodynamic parameters from baseline were limited, both for patients receiving the standard (n=19 patients) and high (n=6 patients) SAGE-547 dose. Regarding hemodynamic parameters, SAGE-547 was well tolerated in the SRSE patients studied, suggesting for further study that SAGE-547 may not elicit immediate or sustained hemodynamic changes in SRSE patients. Real-world studies regarding use of benzodiazepines in pre-hospital and emergency department (ED) was to analyze benzodiazepine usage patterns in SE by emergency medical services (EMS) and the ED of an inner-city hospital. and September 2015 to EMS and hospital ED were reviewed. The associated outcomes of interest were endotracheal intubation, hospital admission, and seizure recurrence. Data was analyzed via descriptive statistics. Of 44 patients analyzed, 43 (97.7%) had a history of epilepsy. Benzodiazepine utilization varied; EMS preferred midazolam (69.2%) while the ED used lorazepam most often (91.3%). Benzodiazepine dosages used were lower than recommended; median dose of midazolam administered by EMS was only 5 mg and median dose of lorazepam in the ED was 2 mg. Patients received 2.7 ± 2.1 benzodiazepine doses on average. Seizure activity was aborted with benzodiazepines alone in 22 (50.0%) patients and recurred in 12 (27.3%). Twenty-three (52.3%) patients were intubated, all post-arrival. There was no observed correlation between number of benzodiazepine doses given and baseline characteristics, decision to intubate, or incidence of seizure recurrence. All 44 patients were admitted and 30 (68.2%) were admitted to the ICU. We observed consistent underdosing of benzodiazepines used for the treatment of SE by both EMS and the ED. There was lower than expected achievement of seizure cessation and intubation rates were higher than reported in previous studies. Further investigation is needed to identify the barriers to optimal benzodiazepine selection and dosing for SE patients at our institution. Super-refractory status epilepticus (SRSE) refers to a condition of persistent seizures that have failed treatment with first-, second-and third-line treatments. SAGE-547, a proprietary formulation of the endogenous neuroactive steroid allopregnanolone (a potent positive allosteric modulator of synaptic and extrasynaptic GABAA receptors in animal models), is being developed for the treatment of patients with SRSE who have not responded to standard treatment regimens. 547-SSE-201 was an open-label, phase -547 in patients with SRSE. The present analysis explores the PK properties of SAGE-547 over the course of the trial. SAGE-547was administered as a 5-day continuous intravenous infusion to 25 patients with SRSE and receiving third line agents (TLAs) for seizure or burst suppression. Patients received either a standard dosing regimen (n=19) or a high dose regimen (n=6) and were subsequently weaned off TLAs and SAGE-547. The standard dose was chosen, based on a modeling approach, to achieve a mean plasma exposure roughly equivalent to the highest endogenous concentrations measured in the third trimester of pregnancy (~157 nM). Since women tolerate this endogenous level without apparent adverse effects, 150 Mean (SD) steadyand for the high dose regimen wa -state concentrations were approximately dose proportional between the standard and high doses, indicating that clearance was dose-independent of infusion to allow determination of half-life or volume of distribution. In this study of patients with SRSE, SAGE-547 clearance was not dose-dependent and plasma concentrations were in line with target exposures. Jennifer A. Creed, Christa C. Swisher. Duke University Medical Center / Department of Neurology, Durham, NC, USA. ****Permission was not granted to print this abstract**** Seizures after resuscitation from cardiac arrest predict worse outcomes, but there is no evidence that treating seizures improves outcomes. We leveraged existing practice variation to compare the effectiveness of aggressive electroencephalography (EEG) and antiepileptic drug (AED) use to infrequent spot EEG and AED use. We performed a retrospective cohort study including comatose post-arrest patients at two academic centers from 2011-2014. The same critical care group staffs both, but Center 1 uses continuous EEG (cEEG) monitoring and aggressively treats malignant EEG patterns while Center 2 uses infrequent spot EEGs and rarely treats with AEDs. We classified each patient's daily EEGs from admission until death, malignant," or "not performed." We abstracted covariates and outcomes from our prospective registry, e rhythm, arrest location, survival to discharge and functionally favorable survival. We used multi-level mixed-effects logistic models to test for an association of Center with outcomes after adjusting for EEG and clinical covariates. We included 824 subjects (Center 1:513, Center 2:311). Center 1 subjects were younger, arrested more often out-of-hospital and had higher illness severity (all P<0.01). Overall, 366 (71%) Center 1 subjects were EEG-monitored (median 2 days (IQR 1-4d)), 196 (38%) had a malignant pattern observed and 227 median of 1d (IQR 1-1d), 62 (20%) had malignant patterns observed (less frequent myoclonic status epilepticu center comparisons). In multilevel modeling, there was no significant Center effect on outcomes. After cardiac arrest, treatment at a center using aggressive cEEG monitoring and AED treatment is not associated with better outcomes at discharge. Phenytoin Dosing Adjustment for Obesity may not be Necessary Effective loading with weight-based dosing of phenytoin for therapeutic levels is necessary in several emergent settings. Practices for dosing obese patients, those >1.3 x Ideal Body Weight (IBW) vary, including using Total Body Weight (TBW), Adjusted Body Weight (AdjBW; correction factor 0.4), and the Abernathy formula (correction factor 1.33). Our objective was to determine whether dose adjustments were necessary for obese patients. Charts were reviewed retrospectively from two tertiary medical centers from September 2014 to August 2015. We included all admitted patients older than 18 years of age, initiated on IV fosphenytoin for any reason, with therapeutic post-load level (total phenytoin of 10-in pre-Mean weight-based loading doses were compared for obese and non-obese patients who achieved postload levels in the therapeutic and high therapeutic range (total level 15-20, free 1.5-2.0), using Welch's two-sample t-tests. A total of 56 patients, 39 non-obese and 17 obese, met inclusion criteria, including 20 non-obese and 7 obese patients who achieved high therapeutic levels, desired for ongoing status epilepticus. The mean -obese and 16.4 (95% CI: 14.5, 18.2) for obese patients (t29=2.4, p = 0.02). Dose to achieve high therapeutic levels was 20.2 (95% CI: 19.3, 21.0) for non-obese and 18.1 (95% CI: 15.9, 20.3) for obese patients (t9=2.1, p = 0.06). Our results suggest that adjustment calculations of fosphenytoin loading dose for obese patients may not be necessary, thus can all patients into high therapeutic range, especially desirable in status epilepticus, while not harmful in other patients. Stephen SAGE-547 is a proprietary formulation of allopregnanolone. SAGE-547 was studied in an open-label nical study of patients with super-refractory status epilepticus (SRSE). The primary -547. -547 in resolving SRSE in these patients. To further understand the specific patient and treatment-related parameters which may affect outcomes in this study, we performed a post-hoc analysis on the completed data set. In this open-label, single-treatment with SAGE-547. Key efficacy outcome measures were: 1) successful wean off of TLA(s) after hour 48; and 2) subsequent successful taper off SAGE-547 after hour 96, without recurrence in the 24hour period following treatment. A total of 25 patients received open-label treatment with SAGE-SAGE-547. Response rate appeared consistent across varying patient demographics (gender, age, ethnicity) and baseline treatment regimens. Overall, 64% of patients experienced at least 1 serious adverse event (SAE) and 6 patients died during the trial. No SAEs and no deaths were attributed by the Safety Committee to SAGE-547 administration. There was little evidence of a relationship between response rate and patient demographics (gender, -trial supports further investigation of SAGE-547 in SRSE, and can inform inclusion criteria for future trials. The clinical efficacy and safety of SAGE-547 in the treatment of SRSE is being evaluated further in an ongoing phase 3, randomized, placebo-controlled trial. Jocelyn Y. Cheng. Drexel University College of Medicine, Philadelphia, PA, USA. In catastrophic neurologic injury, withdrawal of care (WOC) is often considered. While WOC is based on the perception of poor prognosis, the question of whether it creates a self-fulfilling prophecy has been raised. Though studied in traumatic brain injury epilepticus (SE) is unclear. The goal of this study was to describe the final cause of death in adults with SE, and determine the impact and associated clinical characteristics of WOC on mortality rates. A single-center retrospective study at an urban academic medical center was conducted between age;gender;Glasgow Coma Scale (GCS);Acute Physiology and Chronic Health Evaluation-II (APACHE-II);history of epilepsy;etiology of SE;refractory SE (RSE);in-hospital mortality; and cause of death. -tests were used as appropriate. Binary logistic regression analysis adjusted for covariates, and p <0.05 was considered significant. Of 151 subjects, 68 male, mean age 59 years, there were 61(40.4%) in-hospital deaths, 45(74%) due to WOC. The remaining causes were cardiogenic(N=11,18%) and respiratory(N=2,3.2%), with sepsis, brain death and seizures individually comprising 1.6%(N=1 each). Excluding WOC, in-hospital mortality fell to -group without cardiac arrest (CA), inof which 14%(N=15) was due to WOC; mortality decreased t the total cohort, there was no significant difference in baseline characteristics excepting metabolic seizure were more common in WOC subjects. Metabolic etiology (OR:6.50,p=0.005) and CA (OR:4.40,p=0.018) remained significantly associated with WOC after univariate but not multivariate adjustment. Withdrawal of care is a major determinant of mortality in SE, and is associated with metabolic dysfunction clinical decision- An open--SSE-201) evaluated the safety and efficacy of SAGE-547, a proprietary formulation of allopregnanolone, in patients with super-refractory status epilepticus (SRSE). Entry criteria were designed to maximize patient treatment opportunities. The objective of this post-hoc analysis was to demonstrate SAGE-547 activity despite heterogeneity of SRSE causes and high comorbidity burden. d-line agent (TLA; with anoxic brain injury or very short life expectancy were excluded. -76 years) and mean status epilepticus duration was 9.2 days (range 3-20 days). Probable SRSE causes included infection, hemorrhage, worsening CGI-S score. All patients required 1-2 TLAs and 1-5 AEDs at baseline. Up to 8 weans from TLAs were SAGE-547 at the end of 5 da Six patients (24%) died from underlying SRSE cause or associated comorbid conditions. No SAEs were attributed by the Safety Committee to SAGE-547. Mean numbers of baseline wean attempts, comorbid conditions, and SRSE episode duration were comparable between responders and non-responders. activity despite the heterogeneity of cause of SRSE and comorbidity burden. Delayed neurologic deterioration (DND) from vasospasm is associated with poor outcome after subarachnoid hemorrhage. Continuous EEG (CEEG) monitoring has lead to detection of EEG patterns of uncertain clinical significance -Ictal-interictal continuum (IICA). These have been associated with acute brain injury but variably predict outcome. We describe the types and prevalence of IICA EEG patterns in patients who develop angiographic vasospasm and discharge outcomes Retrospective study of adult patients with non-traumatic subarachnoid hemorrhage admitted at Emory University Hospital Neuro-ICU from January 2013-February 2016 who underwent CEEG. CEEG were recorded using 10-20 electrode placement and interpretation of the IICA EEG patterns using the American Clinical Neurophysiology Society research terminology. In 131 SAH patients, 95(73%) were female and hypertensive 88 (67%). Majority had poor grade SAH (40% HH Grade 3, 24% grade 4 and 17% grade 5). Moderate to severe angiographic vasospasm were detected in 74 (56%) patients 5.7 days after admission. CEEG was initiated 4.17 days after admission. Periodic discharges (PDs) occurred in 42(32%) patients, 14(33.3%) of which were generalized and 19 (45.2%) lateralized. Rhythmic Delta Activity (RDA) occurred in 33(25.10%) with 26(78.7%) generalized. Stimulus induced Rhythmic Discharges (SIRPIDS) were seen in 9 (6.8%) and electrographic seizures in 5(3.80%) patients. Vasospasm was common in patients with any IICA patterns (77.5% vs. 22.4% p=0.36), PDs (77.5% vs. 22.4% p=0.56) and RDAs (84.3% vs. 15.6% p=0.12). RDAs were common in patients with discharge MRS 0-4 (59.3% vs. 40.7% p=0.12) and PDs were equally seen across all outcomes (50% vs. 50% p=0.56). The observed trends were not statistically significant. IICAs such as PDs and RDAs were common in patients who developed vasospasm but seen equally with a larger sample size is needed to support these findings. NCS and (NCSE) have been reported in 8-25% of critically ill patients. Whether patients with NCSE should be treated as aggressively as patients with convulsive status remains controversial. This study sing on its correlation with patients' outcome and possible predictors. In this retrospective study n=261 patients underwent cEEG at Mount Sinai Neurological and Neurosurgical ICU (NSICU). cEEG data according to ACNS guidelines was collected. Outcome was evaluated by mortality, Glasgow outcome scale (GOS), Glasgow Coma scale (GCS). We compared 29 Seizures were detected in 41% of subjects and 11% experienced of CG (p=0.000 OR 27.8 CI 5.53of CG (p=0.019, OR 4.93, . Considering clinical predictors, only gaze deviation and subtle facial movements were significant (p=0. 003 OR 4.7, p= 0.000, . Mortality and mean hospitalization length were not different. Outcome was significantly different in NSICU with mean GCS being 8 in NG and 11 in CG (S for p= 0.004), but was not significantly different at discharge, as mean GOS was 2 in NG and 2.4 in CG (p=0.076). Our findings show that patients' history of epil rstand prognostication in these patients. Raquel Farias-Moeller, Archana A. Pasupuleti, Luca L. Bartolini, Amy A. Kao, Brittany B. Cines, Jessica J. Carpenter. Children's National Health System, Washington, DC, USA. Super refractory status epilepticus (SRSE) ensues when there is no improvement of seizure control in response to anesthetic therapy or seizure recurrence after reduction of anesthetic agents. There is no consensus on standard of care for SRSE. Ketogenic diet (KD) has reported success but technical challenges exist including inability to feed patients, concomitant steroid use, persistent acidotic states and ur step-by-step approach to initiation and continuation of KD in the PICU. Patients with SRSE who had KD initiation in the PICU were identified from a prospective Neurocritical Care database with IRB approval. Data from the hospital course was supplemented by review of the electronic medical record. Descriptive analysis was performed. Neurointensivists used our step-by-step guideline to start patients on KD. Nine children with SRSE who had KD initiated in the PICU were identified. The mean age was 5.4 years (SD 2.24). Median number of days to start KD from detection of seizures was 13 . Mean time nine children remained on the KD for 3 months or longer. The median number of AEDs trialed before KD was started was 4 [IQR 3-4] and the median number of continuous infusions was 2 [IQR 2-3]. After initiation of KD most patients were weaned off continuous We demonstrated the feasibility of a practical approach to initiation of KD in the PICU for children with SRSE. These children were successfully weaned off continuous anesthetic infusions. Larger studies, both in children and adults, are needed to determine the effectiveness, safety and tolerability of KD in the management of SRSE as well as its ease of implementation. Although overall mortality of status epilepticus is high, baseline patient characteristics and co-morbidities may help to predict outcomes and shape treatment decisions. Two previously published scores exist to predict outcomes: the Status Epilepticus Severity Score (STESS) and the Epidemiology-based Mortality Score in Status Epilepticus (EMSE). However, a comparison of the two scores has not previously been completed in an American population. We hypothesize that both scores will adequately predict the primary outcome of in-hospital death. We performed a retrospective analysis of all cases of status epilepticus admitted to the Neuro-Critical collected data on age, comorbidities, EEG findings, and seizure history. The primary outcome was inhospital death. A sensitivity and specificity analysis was completed, in addition to a student's t-test for a comparison of the two scores. Forty-six patients were admitted to the NCCU for management of status epilepticus during June 2014 and January 2016, 13 of which experienced in-hospital death. The median age of the sample was 60, with approximately half of the sample (52.6%) having 3 or more comorbidities. The two most common etiologies were cryptogenic (n=9) and acute cerebrovascular events (n=7). While the sensitivity of both EMSE and STESS were very high (100% and 90% respectively), the specificities were very low (28.6% and 42.9% respectively). A student's t-test between those who experienced in-hospital death and those who did not was only significant for EMSE at the p<0.1 level (p=0.055). The specificity of EMSE and STESS for our external validation did not correlate with previous studies; however, both tools are sensitive. The EMSE and STESS may be useful to predict outcomes of status epilepticus in populations with few comorbid conditions, but are less helpful when patients have multiple medical problems. In 2012, the ACNS published critical care EEG terminology in an effort to improve clinical research and management of patients requiring continuous EEG (cEEG) monitoring. We sought to understand the familiarity of providers in our Neurocritical care (NCC) program with this terminology two years after implementation at our institution. We administered a 21 question web-based survey to NCC fellows, advanced practice providers (APPs), tanding of ACNS terminology, and clinical EEG application. There were 5 attending physicians, 4 NCC fellows, and 13 APPs. Attending physicians and APPs had a median of 4 (range: 3, 13) and 2 years (range: 0.25, 15) experience since most recent post graduate training, respectively. All data is reported for APPs and physicians respectively as percentage correct. highest rated component of the cEEG report that influenced patient management was "conversation with lected by 92.3% of APPs and 100% of physicians. set forth by the ACNS and highlight the importance of communication between NCC providers and epileptologists as well as areas of potential education for providers of all training levels. Carbamazepine (CBZ), an oral antiepileptic drug (AED), is a potent inducer of cytochrome P450 (CYP) (eg, phenytoin, fosphenytoin, phenobarbital, valproic acid, levetiracetam, or lacosamide) to reduce the -drug metabolism (reduced efficacy or toxicity). An IV CBZ formulation has been developed; study OV-1015 (NCT01079351) evaluated bioequivalence and 13181A (NCT01128959) evaluated tolerability. Both studies were similarly designed. Eligible adult patients received a stable oral CBZ regimen (400daily dosage (divided doses q6h) during the confinement period (OV-1015: 15-or 30-min infusions q6h for 7 days, patients in the 15-min group were eligible to receive four 2-to 5-min infusions on Day 8; 13181A: 15-min infusions q6h for 4 days, then one 5-min infusion on Day 5). Oral CBZ was resumed for 30 days (13181A: 28 days). Bioequivalence of IV to oral CBZ was evaluated in OV-1015; tolerability data were pooled. In OV-1015, 30-min IV CBZ infusions were within the 80%-125% bioequivalence range vs oral CBZ; 15min infusions exceeded the upper limit for maximum plasma concentration. In both trials, 203 patients switched to IV CBZ (30-min: n=43; 15infusion was dizziness (19%); infusion-site reactions (12%) were the only new AEs experienced by patients vs oral CBZ. Seizure control was maintained during the switch. To avoid potential drug toxicity reactions, it is beneficial to maintain patients on CBZ. IV CBZ administered as multiple 30-min infusions was bioequivalent to oral CBZ. IV CBZ was well tolerated. Treatments for aneurysmal subarachnoid hemorrhage (aSAH) remain inadequate. EG-1962 is a sustained release formulation of nimodipine for intraventricular delivery in order to avoid dose-limiting -label, dose escalation study of a single intraventricular dose of EG-1962 that was designed to determine the maximum tolerated Glasgow outcome scale (www.clinicaltrials.gov Identifier: NCT01893190). Subjects with aSAH repaired by clipping or coiling were randomized within 60 hours of aSAH to EG-1962 or oral nimodipine if they were World Federation of Neurological Surgeons grade 2 to 4 and had a ventricular catheter. Cohorts of 12 subjects received 100, 200, 400, 600, 800 or 1200 mg EG-1962 (9 per cohort) or oral nimodipine (3 per cohort). Plasma nimodipine concentrations were sustained for 21 days. The maximum concentration, steady state concentration and area under the curve for the first 14 days increased with increasing dose of EGbetween males and females. Plasma nimodipine concentrations following EG-1962 administration did not exceed plasma concentrations of oral nimodipine 60 mg every 4 hours at steady state. Cerebrospinal fluid nimodipine concentrations with EG-1962 were orders of magnitude higher than in plasma or with oral nimodipine. Subjects treated with EG-1962 (n=45) had a median intensive care stay 3.5 days less and hospital length of stay 2.5 days less than subjects treated with enteral nimodipine (n=18 , Table) . Intraventricular EG-1962 produced sustained, dose-dependent nimodipine plasma concentrations and shortened intensive care and hospital length of stay. improved clinical outcome support conduct of a pivotal phase 3 study of EG-1962. Increasing exposure to ionizing radiation for medical diagnostics and treatment has raised questions about possible long term effects. This study describes the effective dose of ionizing radiation exposure in patients with aneurysmal subarachnoid hemorrhage (aSAH). Thirty-five aSAH patients admitted to a university hospital between Jan 1, 2015 and December 31, 2015, greater than 18 years old, and discharged alive were included. The primary objective was to calculate the mean total effective dose of ionizing radiation (TEDIR) in aSAH patients. Secondarily, the number of aSAH patients with a calculated TEDIR greater than the annual and cumulative maximal permissible radiation dose (MPRD) as described by the National Council of Radiation Protection and Measurements (NCRP) and the International Commission on Radiological Protection (IRCP) was determined. Factors associated with greater than maximal exposure limits were evaluated. TEDIR exposure ranged from 5.7-89.2 millisievert (mSv), with a mean (SD) of 32.1 (21.79) mSv. Seven , the presence of vasospasm (p<0.01), external ventricular drain (EVD) (p <0.01), or ventriculo-peritoneal Shunt (VPS) (p< 0.05) were statistically significant factors for increased TEDIR in aSAH. Demographic data, previous medical history, and location of aneurysm were not statistically significant. Univariate analysis representing the degree to which TEDIR increases for each of these factors revealed HH class (3 or 4) 20.2 (p<0.01), vasospasm 24.4 (p<0.01), EVD 22.3 (p<0.01), VPS 21.5 (p<0.05). In multivariate analysis representing the degree in which the TEDIR increases, only vasospasm 26.1 (p<0.01) and EVD 14.4 (p<0.01) were statistically significant factors. Following aSAH, patients with severe disease requiring vasospasm treatment and shunting are at warranted. Patients with aneurysmal subarachnoid hemorrhage (SAH) have high morbidity and mortality related to cerebral ischemia and infarction. In this study we explored the reversibility of reduction in cerebral blood flow (CBF) after SAH. We conducted a retrospective analysis using prospectively collected CT perfusion (CTP) data from SAH patients. Patients were grouped as good (Hunt Hess 1-2) and poor grades . CTP data were compared at baseline (0-4 days after aneurysm rupture) and follow-up (>5 days). CBF at baseline was comparable between good and poor grade patients (39.06±9.75 vs. 38.16±11.00 -up there was an improvement from baseline in both groups (42.27±9.08 and 40.23±9.53 respectively, p=0.06). However, in hypoperfused areas, rCBF was significantly lower in poor grade patients compared to good grade (24.71±9.06 vs. 31.25±10.60, p=0.003) and significantly lower than global CBF in both groups (p<0.001). At follow-up, only poor grade patients demonstrated an increase in rCBF (29.02±9.55, p=0.01) while in good grade patients, rCBF remains unchanged (31.79±9.86, p=0.41). The absolute improvement in rCBF was significantly greater in poor grade patients ow-up in both groups was not statistically different in the two groups (p=0.09) but significantly lower than global CBF (p<0.001). Regional hypoperfusion occurs after SAH. In good grade patients there is no recovery in rCBF while in poor grade patients there is partial recovery implying a biphasic response with a reversible and an irreversible reduction in rCBF. This has not been previously described in the literature and may implicate two distinct mechanisms responsible for rCBF reduction after SAH. Aneurysmal subarachnoid hemorrhage (aSAH) is an important cause of morbidity and mortality, with patients susceptible to a variety of medical complications. External ventricular drains (EVDs) are commonly used for intracranial pressure monitoring and CSF drainage; however, this puts aSAH patients -associated infections (VAIs). Many preventative strategies have been proposed and implemented over the last 20 years. Using the Nationwide Inpatient Sample (NIS) database, this study examines trends in EVD usage, VAI rates, and mortality over a 12-year period. In this retrospective analysis, data from the NIS was obtained for the period of January 1, 2000 through December 31, 2011 using International Classification of Diseases, 9th revision (ICD-9) codes. Analysis was performed using SAS 9.4 surveymeans. Primary outcomes of interest were rates of EVD use, VAI and in-patient mortality, as well as hospital length of stay. During the study period, there were 422,168 aSAH admissions, with EVD placement reported in 15.8% of cases. There was no change in either EVD use or rate of VAI (mean VAI rate of 5.91% over the 12 years). No change in hospital length of stay was observed. From January 1, 2000 to December 31, 2011, Despite a variety of VAI-preventative strategies advocated for and implemented, the national VAI rate in aSAH patients has not changed; however, mortality has declined by a mean ARR of 0.83% per year over the study period. This may reflect improved neurointensive care provided to this critically ill population. Stable VAI rates may reflect incomplete adoption of effective preventative strategies, or use of ineffective o study has previously examined these trends in aSAH. The purpose of this study was to determine the effect of routine use of dexamethasone on delayed cerebral ischemia and poor outcome (death and severe disability) after aneurysmal subarachnoid hemorrhage (aSAH). This is a single center, observational cohort study comparing patients with aSAH admitted to a tertiary referral center from 2013 to 2015. A variation in practice patterns for the use of dexamethasone 4-6 mg every 6 hours after aSAH exists in our institution depending on neurosurgeon preference. Patients were followed prospectively for the occurrence of complications including delayed cerebral ischemia (DCI), e (bad outcome defined as a mRS of 4-6) and 3 months (bad outcome defined as mRS of 3-6). Out of the 173 patients, 59 (34%) patients received dexamethasone during the first 72 hours of admission. Significant factors associated with steroid use were females (41% v 21%;p=0.01) and aneurysm clipping verses coiling (55% v 27%;p<0.01). There was no difference in HH, Fisher grade, incidence of infections, or incidence of DCI (34% v 23%;p=0.15). Steroid use was significantly associated with bad outcome at discharge (72% v 50%;p=0.01), but no difference at 3 months (34% v 31%;p=0.7). When examined separately for coiled and clipped patients similar trends were found in both subgroups. Steroid use was associated with a longer hospital length of stay (LOS). In multivariable regression analysis steroid use was significantly associated with worse outcome (OR 3.6;p<0.01) when controlled for age, HH grade and type of surgical intervention. The study did not detect any benefit in the use of steroids in reducing the incidence of delayed cerebral ischemia in acute aneurysmal subarachnoid hemorrhage. However, steroid use was significantly associated with longer LOS, and worse functional outcome at discharge. Antiplatelet therapy for the prevention of peri-coiling thromboembolism in high risk patients with ruptured intracranial aneurysms Thromboembolic events (TEE) during or after coiling of intracranial aneurysms is the most frequent procedural complication, resulting in permanent neurologic disability in a subset of patients. In unruptured aneurysm patients, there is evidence supporting the use of periprocedural antiplatelet therapy to prevent TEE. Whether patients with ruptured aneurysms and subarachnoid hemorrhage should also be given peri-coiling antiplatelet therapy is less clear. We reviewed a prospective registry of 169 endovascularly treated aneurysm patients to delineate angiographic features associated with periprocedural TEE. We then performed a controlled before-andangiographic feature associated with TEE) to evaluate whether selective aspirin administration would reduce the rate of periprocedural thromboembolism without increasing major hemorrhagic complications. small parent artery diameter, an incorporated branch, intraprocedural thrombus formation, and parent rate of periprocedural TEE, from 53.8% in the control group to 10.6% in the aspirin-treated group (p = 0.001). TEE reduction in the aspirin-treated group continued to be statistically significant even when ith TEE in other large studies with an adjusted OR of 0.16 (95% CI 0.03 -0.8). There were no major systemic hemorrhagic -bleed, symptomatic intracranial hemorrhage, or major external ventricular drain (EVD)-associated hemorrhage (p = 0.3). significantly reduce the rate of peri-coiling TEE without increasing major systemic or intracranial hemorrhages. Neurocrit Care (2016) 25:S1-S310 Muhammad K. Athar, Umer U. Mukhtar, Umer U. Shoukat, David D. Boorman, Fred F. Rincon, Matthew M. Vibbert, Syed S. Shah, Jacqueline J. Urtecho, Jack J. Jallo. Thomas Jefferson University Hospital, Philadelphia, PA, USA. Fever is frequent in patients with subarachnoid hemorrhage (SAH). Differentiating infectious fever from central fever can be challenging. It is important to diagnose the cause of fever in the neurological intensive care unit (NICU) because of the detrimental effects of fever on brain injured patients. We hypothesized that procalcitonin (PCT) could be useful to distinguish central fever from infectious fever in patients with SAH. Prospective, chart review study conducted in the NICU between December 2012 and September 2015. was clinical infection defined as positive cultures (blood, urine, sputum, mini BAL, CSF, and C. difficle toxin) or infiltrate on chest X-ray within 3 days of onset of fever. Sixty-Twenty--0.5, and 16 had PCT >0.5. Out PCT >0.5. Using multiple logistic regression, PCT between 0.1-0.5 had an odds ratio of 5.34 (95% CI 1.20-23.73), PCT >0.5 had an odds ratio of 64.82 , and a maximum temperature odds ratio of 1.825 (CI 1.095-3.042). Using PCT >0.5 alone had an odds ratio of 10.11 (95% CI 2.51-40.74). -PV: 70.0% with a sample prevalence of 42.4%. ROC Curve area: 78.2%. fever in SAH patients. The test has high specificity and NPV so it can be a valuable toll to rule out infectious fever in NICU. Intraventricular hemorrhage (IVH) due to subarachnoid hemorrhage (SAH) has been associated with fever, hydrocephalus, and shunt dependence. The modified Graeb Score (mGS) as an enhanced measure of intraventricular hemorrhage has been shown to correlate wit intracerebral hemorrhage (ICH) as well as shunt dependency in SAH. We evaluated the mGS's association to complications during hospital stay and impact on functional independence at discharge in SAH patients. Retrospective review was performed of prospectively collected data for consecutive SAH patients enrolled into the University of Maryland Recovery After Cerebral Hemorrhage (REACH) Study. Hunt and Hess (HH) grade, global cerebral edema (GCE), and infarct on admi team of neurointensivists. mGS was calculated from each patient's admission CT scan and dichotomized according to a cutoff value based on the median value for our sample. Clinical complications during each admission were recorded, and independence of performing ADL's was obtained from physical and occupational therapy notes. Statistical analysis was performed using univariate and multivariate logistical regression. Ninety-eight SAH patients from July 2014 to November 2015 were reviewed for this study. mGS was calculated in 83 patients and dichotomized based on a median cutoff value of 5. HH, GCE, and admission infarcts were not found to be significantly associated with high mGS. On univariate analysis, elevated mGS was significantly associated with hospital acquired infections (UTI, pneumonia, and sepsis, p<0.03), fever (p=0.02), hypotension (p=0.002), hypernatremia (p=0.046), symptomatic vasospasm (p<0.03), and new i independence with ADL's (p=0.02). Severity of IVH as measured by mGS is associated with multiple in-hospital complications. The mGS can be used as an independent predictor of loss of independence of ADL's on discharge for patients with SAH. ed cerebral ischemia (DCI) and brain injury following subarachnoid hemorrhage (SAH). While systemic corticosteroids may mitigate inflammation and promote fluid and salt retention following SAH, there is limited evidence on the impact of corticosteroid administration on outcomes following SAH. Corticosteroids are frequently administered in clinical practice following SAH for the management of post-operative cerebral edema and refractory headache. Our goal was to examine the impact of corticosteroid use following SAH on the occurrence of DCI and poor functional outcome at discharge. Retrospective analysis of data from a single center SAH registry on patients admitted between 2010-2015 who survived >24 hours. A logistic regression model was created with multiple potential predictors of outcome and steroid use, and with corticosteroid use as the response variable. Patients were divided into quartiles based on the propensity score. The impact of corticosteroid use on the outcome of interest (DCI then poor functional outcome at discharge) was then determined while controlling for the propensity score quartile. Co-variates in analysis included age, gender, ethnicity, history of diabetes or statin use, aneurysm location, aneurysmal vs non-aneurysmal bleed, treatment modality, Hunt Hess, modified Fisher. A total of 440 patients with aneurysmal and nonaneurysmal SAH were included in this analysis. Corticosteroids were administered in 166 (37.7%). DCI occurred in 122 (27.7%). Poor outcome (mRS>3 at discharge) occurred 174 (39.5%). Following propensity score analysis, corticosteroid use was not associated with DCI (p=0.826) but was associated with a significant reduction in poor outcomes at discharge (p=0.038, OR 0.609, 95% CI 0.381-0.972). Corticosteroid use following SAH was not associated with a reduction in DCI but was associated with an approximately 40% reduction in the odds of poor functional outcome at discharge. A clinical trial of corticosteroids initiated in the early period following SAH may be warranted. Besides the impact of the initial bleeding, CV remains the leading cause for mortality and morbidity after successful CIAN therapy? Data of 30 patients with CIAN were analyzed with regard to onset of multimodal neuromonitoring, if one or both hemispheres were monitored and for the integration of neuromonitoring values in decision 6 months after SAH using the Glasgow Outcome Scale (GOS). CT-scans were reviewed for infarctions at time of discharge. 24 patients were in the favourable outcome group (GOS4-5), 6 patients showed unfavourable outcome (GOS 1-3). In 4 patients of the GOS 1-3 group neuromonitoring was implanted in the hemisphere with the highest transcranial doppler (TCD) values. Additional monitoring was installed contralaterally if TCDs increased. In 3 of those patients, contralateral PbtO2 values were ischemic and angiography revealed severe CV in the non CIAN treated hemisphere. CT scans of those 3 patients revealed significant infarctions in the hemisphere that was not initially monitored. The 24 patients with GOS 4-5 were monitored bilaterally at early timpo small infarctions but no territorial infarct was seen at discharge. A delay in bilateral multimodal neuromonitoring might facilitate delayed cerebral ischemia (DIC). This might be due to a delayed detection of a mismatch between oxygen supply and consumption. In addition severe CV is not always detectable in TCD and might thus be diagnosed too late to initiate a successful CIAN therapy. In contrast, new severe CV or relaspses of CV after stop of CIAN therapy are detected efficiently if bilateral neuromonitoring is in place and the values are acted upon accordingly. The effects of short versus longer duration prophylaxis of levetiracetam on cognitive / functional outcomes in aneurysmal subarachnoid hemorrhage and risk of development of delayed seizures Tamara Majic, Dela D. Amoussou, Chrystal C. Reed, Asma A. Moheet. Cedars Sinai Medical Center, Los Angeles, CA, USA. Chart review performed on 136 patients admitted from January 2010 to December 2015 with aSAH, who received levetiracetam seizure prophylaxis 500 mg BID or greater for <3 days versus 7 days or more. We compared the length of ICU stay, delta GCS at discharge, MRS (30 vs 90days), and incidence of delayed seizures --1.05 ; p < 0.453) Lengths of ICU stay for short-duration levetiracetam therapy was 1 --6.54 for long-duration (P<0.378). Length of ICU stay in low dose --6.6 days (p < 0.1). Preliminary data for early and late onset seizures delayed seizures occurred with longer duration prophylaxis Preliminary data suggests delayed cerebral ischemia was universally present in patients with delayed seizures. The incident rate of DCI was higher in patients with early seizures (40%) vs without seizures (20 %) between low dose and high dose, and between short duration and long-duration levetiracetam therapy. Length of ICU stay is shorter in subjects treated with low dose levetiracetam vs high dose, which may suggest that a low dose levetiracetam may have a lower adverse effect profile. The presence of delayed ischemia may warrant a longer duration prophylaxis. The longer duration prophylaxis does not seem to reduce the incidence of delayed seizures, although a longer duration of study is warranted. In spite of improvements in mortality and physical disability for aneurysmal subarachnoid hemorrhage for 'delayed brain injury' often attributable to the direct neurotoxic and neuroinflammatory influence of the initial hemorrhage burden. These processes can result in global brain atrophy and commonly manifests as new cognitive disability including deficits with memory, executive function, and language. Heparin exerts a wide range of interactions postulated to antagonize multiple pathophysiological mechanisms implicated in aSAH. Here we review low-dose IV heparin (LDIVH) as a promising treatment for preventing 'delayed brain injury' in aSAH survivors and inform on a new multi-center randomized trial. Recent studies evaluating LDIVH in aSAH are reviewed. The ASTROH study is an open-label, blindedadjudication, randomized phase II trial. The primary efficacy outcome is mean Montreal Cognitive Assessment (MoCA, 0-30, Normal 26-30) scores at the 90-day follow-up and 88 patients will be enrolled over 2 years at 9 academic medical centers. The primary safety outcome is any Major Bleeding or Clinically Relevant Non-Major Bleeding. One-year outcomes are also being assessed. LDIVH significantly reduced neuroinflammation, demyelination, and transsynaptic apoptosis in a rat SAH model. In a retrospective study 43 LDIVH patients were compared to 43 well-matched controls. LDIVH subjects had 9% clinical vasospasm and 0% vasospasm related infarction compared to 47% and 21% respectively in controls (P=0.0002 and P=0.003). In another retrospective cohort study LDIVH patients (n=25) had mean MoCA of 26.4 compared to 22.7 in controls(n=22) (P=0.013). Multivariate analysis confirmed LDIVH positively influenced MoCA scores when controlling for factors that negatively influenced cognition. The ASTROH study is active and enrolled its first subject in April, 2016. LDIVH is a promising treatment for aSAH and is currently being investigated in a multi-center randomized trial (ASTROH), NCT02501434. gastrointestinal bleeding (aSAH) patients and to determine the effect of GIB on in-hospital complications and outcomes. GIB in aSAH patients and to determine the effect of this complication on other in-hospital complications and outcomes. The incidence of GIB in aSAH hospitalizations (N=239,808) was 1,363 per 100,000 patients with 24.7% requiring blood transfusions. Multivariate independent predictors of GIB included: age 55-64 GIB in aSAH is uncommon and is influenced by patient demographics and preexisting comorbidities and significantly increases disability and mortality. Twenty-six consecutive aSAH patients undergoing multimodal neuromonitoring including CMD were studied. Interventions of full-strength enteral nutrition (EN) after >4 hours without any feeding preceding EN were identified. Parameters of systemic and cerebral metabolism and insulin dose were timeand analyzed together with continuous variables to study the effect of EN on brain metabolism (glucose, lactate, pyruvate and glutamate). Out of 208 interventions in total, 145 were excluded because of simultaneous parenteral supplementation or missing values, leaving 63 interventions in 17 patients eligible for analysis. The mean EN--glucose significantly increased from perfusion pressure (CPP), baseline serum and brain glucose levels, the baseline metabolic profile [brain metabolic di and independent of the insulin dose given during the intervention. The increase of CMD-glucose was strongly dependent on the delta increase of serum glucose (median during the intervention (p<0.01). Although probe location influenced absolute CMD-glucose-levels (p<0.001), significant increases were even observed in perilesional brain tissue (p<0.001). No change in CMD-lactate, CMD-pyruvate, CMD-LPR or CMD-glutamate levels were observed (p over 0.4). Brain glucose levels increased during enteral feeding independent of CPP, baseline glucose levels, insulin administration, and probe location. Despite this increase, no additional metabolic improvement was observed. The clinical benefit of interventions ta needs to be investigated in a prospective approach. Mean platelet volume (MPV) is a common daily laboratory investigation in subarachnoid hemorrhage diseases and is postulated to signify increased systemic thrombogenicity. Similarly, diabetics have elevated MPV suggestive of associated vasculopathic complications through increased thrombogenicity. with non-aneurysmal SAH (naSAH) as compared to aneurysmal SAH (aSAH). Hence, we investigated if vasculopathy. We reviewed charts of patients admitted with the diagnosis of SAH between January 2011 and December 2015. We compared proportions using Fisher's exact tests, and constructed ROC curves to find threshold values for admission MPV that had the best combination of sensitivity and specificity to predict naSAH versus aSAH. Of the 237 patients who met the inclusion criteria 200 were aSAH. Diabetic patients who presented with diabetic patients, a threshold value for MPV of 8.97fL yielded the best combination of sensitivity and specificity to predict aSAH vs naSAH (AUC=0.6; 95% CI 0.5, 0.7). Using this threshold, SAH is more -Similar MPV association was not observed among diabetic patients presenting with SAH. Mean MPV at admission did not differ between diabetic patients who presented with aSAH (9.1fL) and those who presented with naSAH (8.9fL, p=0.36). Nonng with SAH. Aneurysmal subarachnoid hemorrhage (aSAH) is associated with mortality rates up to 20%, and up to one half of survivors suffer from long term neurologic disability. Though several clinical scores have been developed to predict in-hospital mortality and long term outcomes, there is no universally accepted score. create a new predictive model. We conducted a retrospective chart review of patients admitted with aSAH to a single neurocritical care unit from September 2011 to February 2016. We excluded patients with non-aneurysmal SAH (including trauma, AVMs, and mycotic aneurysms). Demographic and clinical variables collected included age, admission GCS, admission APACHE II score, Hunt and Hess score, presence of delayed cererbral ischemia, and hospital acquired infections. Our outcome measure was Glasgow Outcome Scale at discharge. We created a penalized logistic regression model to determine predictors of outcome. We assessed performance by estimating the area under the ROC curve (AUC). Of 204 patients reviewed, 160 met inclusion criteria. The mean age of the cohort was 58 years. 69.4% (n=111) of patients were female. The mean APACHE II score on admission was 14.4 (median 13.5). Majority of patients (n= 97, 60.63%) had a discharge GOS of 3. A combination of 5 predictors performed optimally: age, admission Apache II, GCS, use of mechanical ventilation and presence of hospital acquired infections. The mean AUC of the model was 89%. At the point of maximum-accuracy on the ROC curve, the sensitivity was 96%, and specificity was 68%. Clinical features at admission and during hospitalization can predict outcomes in patients with aSAH. Clinical characteristics from the first few days of the hospital admission, such as hospital acquired infections, can be added to existing models, to improve outcome prediction scores. Subarachnoid hemorrhage (SAH) patients may experience supply-demand mismatch of cerebral metabolism from seizures, vasospasm, cortical spreading depolarization, hydrocephalus, or cerebral edema. Previous studies have focused on non-neuronal measures of cerebral autoregulation. We examine the impact of various neurocritical interventions by examining anecdotally identified intracranial EEG (iEEG) responses considered clinically impactful as well as systematic examination of repeated interventions within patients. SAH patients of Hunt-Hess grade 3-5 underwent 1) clinical multimodality neuromonitoring utilizing brain tissue oxygen cerebral oximetry, cerebral blood flow, Spencer depth electrode, and fiberoptic ICP through a quadconsent using time-synchronized monitoring (CNS-210, Moberg Research). We reviewed clinician anecdotes of treatment responses to vasopressors, endovascular vasodilators, anti-seizure pharmacotherapy, nimodipine, and ventriculostomy adjustments. We then assessed each patient's response to multiple grouped interventions using spectral features including alpha-to-delta ratio (ADR) normalized to pre-intervention baseline (nADR). Paired t-tests and scatter plots, respectively, demonstrated the impact of interventions and blood pressure on nADR. 12 patients had available post-SAH iEEG data over 22 months. Of 5 patients with post-procedural brain responded with an increase in iEEG alpha activity power. Two patients developed a decline in ADR associated with aSAH-related vasospasm, one who had EEG improvement after endovascular spasmolysis. Two patients developed scalp-negative iEEG seizures, both electroclinically improving with anti-seizure pharmacotherapy. Grouped interventions showed heterogeneous responses to vasopressors and one patient with a significant, repeated response. Nimodipine had no consistent discernible peri-dose impact on nADR. We display scatter plots showing the peri-intervention patient-specific correlation between mean arterial pressure and nADR. Patients with aneurysmal SAH may develop neuronal impairment rescuable by neurocritical care interventions. Our data show these responses are patient-and statemodels of dynamic SAH pathophysiology. Introduction: (DCI), may be an important determinant of outcome following subarachnoid hemorrhage (SAH). Potentially, early treatment measures such as control of intracranial pressure, blood pressure management and initiation of nimodipine may mitigate EBI. Our objective was to study the impact of delayed presentation to medical care on the occurrence of DCI and poor outcomes following SAH. Retrospective analysis from a single center SAH registry. Patients admitted between 2011 and 2015 for nontraumatic SAH, who survived more than 24 hours were eligible for inclusion. [VR1] The explanatory variable of interest was time from symptom onset to diagnostic CT, dichotomized at 72 hours. Covariates included age, gender, ethnicity, Hunt-Hess grade, modified Fisher grade, hypertension, aneurysm location and treatment modality. The primary outcome of interest was poor functional outcome at discharge (defined as models were constructed with the outcomes of interest as the response variables. A total of 315 patients were included. The median time to diagnosis was 9.0 hours (interquartile range 3.9-16.9). Twenty-four patients (7.6%) presented greater than 72 hours from onset. Poor functional outcome at discharge occurred in 138 (43.8%) and DCI in 86 (27.3%). Multivariate analysis revealed no association between delayed presentation and either DCI (P = 0.289) or poor functional outcome at discharge (P = 0.7). Hours from symptom onset to diagnosis as a continuous variable also did not reveal a significant association with DCI or poor functional outcome. Delayed presentation to medical care beyond 72 hours is not associated with either DCI or poor functional outcome at discharge following subarachnoid hemorrhage. A treatment bundle including extracorporeal cardiopulmonary resuscitation (ECPR) combined with targeted temperature management (TTM) may improve outcome of cardiac arrest (CA) patients, however, prognostication for these patients still remains challenging. We sought to examine the prognostic value of amplitude-integrated electroencephalogram (aEEG) for CA patients during ECPR and TTM. This was a single-center, retrospective analysis of adult CA patients treated with ECPR and TTM under aEEG monitoring with subhairline montage. Intra-arrest cooling was immediately initiated with cold fluid infusion and extracorporeal cooling method and maintained at 34 °C for 24 h. Patents underwent intraaortic balloon pumping (IABP) and percutaneous coronary intervention (PCI) if needed. Neurological outcome was assessed with the Cerebral Performance Category (CPC) scale at hospital discharge. ECPR was conducted in 17 patients (age 57.8 [24-84] years, 82% male) amongst 73 CA or post-CA comatose patients since November 2012. The initial cardiac rhythm was refractory ventricular fibrillation in 14, pulseless electrical activity in 2, and asystole in 1. The cause of CA was cardiogenic; 10 underwent PCI and 9 needed IABP support. Collapse-to-ECPR time was 34.3 min. Initial aEEG patterns were; flat trace (n=4); low voltage (n=6); suppression-burst (SB) (n=2); electrographic status epilepticus (ESE) recovery (CPC 1-2). Their aEEG pattern was continuous in 1, low voltage in 3, and ESE in 1. Among 3 rn of spontaneous circulation. Patients with ESE recovered after antiepileptic administration. ECPR was withdrawn in 6 patients based on clinical and prolonged flat aEEG findings. Continuous aEEG adds early prognostic information for CA patients with ECPR under TTM. The Suppression Ratio (SR) is a processed EEG variable estimating the percent of an EEG epoch (0-100) that is suppressed. SR has been associated with neurologic outcome after several types of brain injury and using different technologies including full montage EEG recordings and simplified processed EEG monitors. We compared SR during targeted temperature management (TTM) after cardiac arrest, using two independent blinded assessment tools. A convenience sample of adult patients treated with TTM after cardiac arrest were enrolled to compare and the full montage continuous EEG using Natus equipment with Persyst Magicfor 0.5 seconds). Machine times were recorded to synchronize, and SR results were recorded once for each subject at a time without stimulation or artifact using correlation and Altman-Bland analysis. 20 adults were enrolled in this study with a median age of 56 years, 14 (70%) were male. During SR -46) for Persyst SR 1.7 (0.2-41). Comparing Medtronic and Persyst SR, the Spearman correlation was 0.88 (p<0.0001), and Altman Bland testing revealed a bias of 0.6 with 95% limits of agreement -3.7 to 4.8. Bedside estimation of Suppression Ratio during TTM after cardiac arrest showed excellent agreement when measured with the Medtronic bispectral index monitor and the full montage Natus cEEG monitor Though the impact of therapeutic hypothermia on neurological outcomes remains controversial, there is strong evidence that pyrexia is detrimental. Posthypothermia fever experienced by cardiac arrest patients is of particular concern. This abstract examines the ability of an esophageal heat transfer device (EHTD) to maintain core temperature below 38°C in critical care patients, with a focus on posthypothermia fever in post cardiac arrest (PCA) patients. De-identified data for subjects who received temperature management using an EHTD were collected with a condition appropriate for active temperature management. Core temperature readings for each patient were recorded at least hourly; if measurements were recorded more frequently, temperature over an hour span was averaged. Patient data was analyzed to determine what proportion of measurements were above 38°C. Data from a total of 26 patients was collected, including 23 post-cardiac arrest patients and 3 fever reversal cases. A total of 934 core temperature measurement events (over an average of 35.9h per patient) were included in the analysis. 927 (99.3%) were below 38°C, 7 recorded measurements exceeded 38°C, and no data were recorded for 4 time points. Of the 254 measurements recorded posthypothermia, 252 (99.2%) remained below 38°C. Esophageal temperature modulation using an EHTD appears to be an effective method for fever prevention and reduction. Visual representations of the PCA subset showed an upward trend in temperature after 10-18 hours of maintaining target temperature, but before active cooling ended. This suggests that many of these patients might have become febrile in the absence of active temperature management. Achieving and maintaining normothermia (NT) after subarachnoid hemorrhage (SAH) or intracerebral hemorrhage (ICH) often requires surface or intravascular cooling devices that are associated with a significant burden of shivering. We describe a new, closed loop esophageal cooling device (ECD: --0.50 C) and the shiver burden during the maintenance of NT. We enrolled mechanically-ventilated patients with SAH or ICH with refractory fever (>38.3 C). Temperature and Bedside Shivering Assessment Scale (BSAS) were recorded every 15 minutes for the time above 38 C, median BSAS and cumulative number of anti-shivering interventions per patient was recorded prospectively. All patients received magnesium, buspirone, and acetaminophen as baseline anti -shivering interventions. Ten patients (7 ICH, 3 SAH) were enrolled between October 2015 and April 2016. The median GCS at initiation was 6 (4----0.2 m2, and 70% were women. There was a temperature reduction at 120 minutes (mean 38.7C to 37.9C, p=0.005) and 90% of patients achieved NT (median time = 4.5 hrs.; range: 0.5 -38 hours). NT was maintained for median 91 --28%) time above > e time. The median number of total shiver interventions per patient was 5 (1 -22) throughout the TTM time period. No device related complications were noted. The ECD successfully achieved and maintained NT with a low shiver burden and may be a feasible option for NT in this critically-ill population. We present a case of toxic leukoencephalopathy in a young woman taking a thermogenic dietary supplement. A 30 year old female with unremarkable PMH except being on a diet drug "Remuvik" presented with a 2 day history of severe headache, blurry vision, photophobia, phonophobia, nausea, vomiting and brief intermittent hand spasms. Neurological exam was notable for mild right finger-to-nose ataxia and diffuse hyperreflexia. Initial MRI demonstrated extensive bilaterally symmetric T2 hyperintensities of the corpus callosum and periventricular white matter. CSF profile was unremarkable. Labs were unremarkable except for serum sodium of 127 mEq/L. Approximately 18 hours later, patient became unresponsive with bilaterally fixed-dilated pupils and decerebrate posturing. She was intubated and 100gms of mannitol was emergently administered with concern for cerebral edema. IV lorazepam was also given. A stat CT head showed diffuse cerebral edema. An external ventricular drain was placed emergently and 3% NaCl was started. Continuous EEG was negative for seizures. Next day she started following commands and on day 5 she was discharged home with normal neurological exam. Given her presentation and MRI findings, she was diagnosed with acute toxic leukoencephalopathy due to thermogenic diet pill "Remuvik". Patient had been taking the diet drug for 6 months and had lost 30lbs. The main ingredients in Remuvik are listed as Resveratrol, caffeine-free green tea and L -Carnitine while the remaining ingredients are unknown. Similar presentation with another diet drug "Thermatrim" has been previously reported. These products are not FDA regulated and are easily available to the general public. The acute cerebral edema with decompensation was thought to be due to hyponatremia caused by Remuvik. While the mechanism of leukoencephalopathy is not well understood and further investigation is needed, spreading awareness is the key to prevent serious adverse effects of such unregulated products. Baclofen is a frequently used muscle relaxant. We report a case of, low dose Baclofen causing reversible GPED's (Generalized Periodic Epileptiform Discharges). On review of literature, Baclofen toxicity/overdose has been associated with burst suppression patterns on EEG, with one case report of Baclofen toxicity causing GPED's. To the best of our knowledge there have not been reports of low dose Baclofen induced significant EEG changes. Case reportThe patient is an 87 year old woman, with poor baseline functional status from advanced dementia and limb contractures, on Coumadin for old DVT/PE, with sub-therapeutic INR, was admitted with new onset seizures secondary to venous infarcts over bilateral parieto-occipital areas, due to extensive venous sinus thrombosis. She was monitored on continuous video EEG and initiated on antiepileptic medications, Vimpat and Dilantin. EEG recording initially demonstrated occasional sharp waves, maximal in the left frontal region. However, because of excessive EMG artifact caused by hypertonia, the patient was started on Baclofen 5 mg. Within 24 hours patient's mental status deteriorated and EEG recording demonstrated GPED's with periods of suppression. Due to concern for drug adverse reaction, Baclofen was discontinued. The EEG reverted to pre-baclofen pattern, while her mental status slowly improved. She was provided supportive care and ultimately discharged to a rehabilitation facility. In this elderly dementia patient, with low seizure threshold from the acute cerebral insult, low dose of Baclofen was enough to induce encephalopathy and GPED's. The absence of any metabolic disturbances along with rapid resolution of clinical and electroencephalographic abnormalities after discontinuation of the drug supports the hypothesis that these findings may be the direct cerebral toxic effect of Baclofen. Iatrogenic encephalopathy with Baclofen should be considered in the differential for elderly patients with low cognitive reserve Rotational vertebrobasilar insufficiency, also called Bow Hunter's syndrome after the symptom-inducing head position adopted when aiming a bow, is a rare cause of posterior circulation ischemia. We present a case of an 84-year-old woman who presented to Barnes-Jewish Hospital with several days of episodic vertigo and gait instability. Two weeks prior to presentation she had fallen and struck her head. Imaging revealed a complex C1 Fracture as well as an odontoid fracture with posterior displacement. She began having positional spells characterized by loss of consciousness, gaze deviation, fencer posturing, and sonorous breathing. Review of clinical records and literature review. The spells were initially highly concerning for seizures. The patient was monitored on continuous video EEG, however no seizures were detected during typical spells. A CT angiogram revealed an occluded right vertebral artery at the level of C2 with diminutive vs. absent posterior communicating arteries isolating the posterior circulation. Subsequent MR angiography revealed a patent right vertebral artery with no evidence of stroke. Catheter cerebral angiography demonstrated a patent left vertebral artery. Turning the head 15 degrees during the procedure, however, elicited a typical spell and completely occluded the left vertebral artery. The patient underwent occipitocervical fusion, but unfortunately suffered a multifocal posterior circulation stroke and was discharged with hospice care. We present an unusual case of rotational vertebrobasilar insufficiency that mimicked a classic sezure semiology and presented several diagnostic dilemmas in the ICU. In this case, traumatic injury resulted in likely bilateral positional vertebral artery occlusion with resultantly profound brainstem ischemia. Bow hunter's syndrome should be considered in all cases of positional neurological spells, particularly in the setting of neck injury. Optic nerve sheath diameter (ONSD) measurement using ultrasound has been proposed as a reliable method for non-invasive assessment of intracranial pressure (ICP). We report a case of using ONSD to monitor ICP in a TBI patient with elevated ICP undergoing medical treatment with acetazolamide. We hypothesize that a difference in ONSD could be detected with ultrasound before and after treatment. Patient is a 31 year old man with mild TBI due to assault. His Head CT reveals a long calvarial fracture extending along the superior sagittal suture line and posteriorly into the left parietal bone, as well as a large epidural hematoma overlying the frontoparietal vertex near midline, and causing inferior displacement and extrinsic compression of the superior sagittal sinus. His physical exam reveals a young man with right orbital ecchymosis who is sleepy but easily arousable with a GCS of 4 5 6 and no motor deficits. Dilated fundus exam by ophthalmology reveals grade 2-3 papilledema consistent with elevated ICP. The patient complains of persistent headaches and nausea that is unremitting. Acetazolamide was started to decrease ICP. We measured ONSD with a sonosite ultrasound device prior to start of acetazolamide and 2 days afterwards. Two measurements were taken on each eye, one in the horizontal and vertical orientation each. The average ONSD was 6.4mm on the right eye and 6.4mm on the left eye prior to initiation of treatment. On the day after treatment ONSD was 4.65mm on the right and 5.1mm on the left eye. The patient's headache improved and nausea resolved. The next day ONSD was 3.5 mm on the right and 3.6mm on the left eye. Headache and nausea completely resolved. This case report affirms that ultrasound measurement of ONSD could be used reliably to assess ICP noninvasively during the course of treatment for elevated ICP. Manoj K. Mittal. Kansas University Medical Center/ Neurology, Kansas City, KS, USA. Timing of brain death evaluation could be crucial in maintaining organ perfusion for donation. A new bedside cerebral blood flow monitor (cFLOW monitor from Ornim) has not been previously studied for determining the timing of brain death examination. We present here a case illustrating the role of bedside blood flow monitoring in determining the timing of brain death evaluation. A 73 year-old-woman presented with acute right middle cerebral artery stroke and bilateral internal carotid artery occlusions. She was not a candidate for intravenous thrombolysis or endovascular therapy due to unknown time of symptoms onset. Her initial NIHSS was 19 (right gaze deviation, mild aphasia, mild dysarthria, left facial droop, left hemiparesis, left sided decreased sensation and neglect). Day-2, she got intubated for hypoxic respiratory failure. Day-3, CT head showed cerebral edema with midline shift of 8 mm. Patient was not a decompressive hemicraniectomy candidate. Day-5, patient was comatose. Day-7, patient lost bilateral pupillary reflex. CT head showed worsening midline shift of 17 mm with right uncal herniation, bilateral anterior cerebral artery and left posterior cerebral artery stroke, and brainstem compression. Day-9, bedside cerebral blood blood flow monitoring was started with right sided cerebral blood flow index (CFI) of 16 and left side CFI of 35. Patient met criteria for brain death except that she was still breathing over the ventilator. Patient was extubated for comfort measures. After 60 minutes patients stopped breathing. Her CFI dropped <10 bilaterally. Patient underwent cardiac arrest after 10 minutes and then both CFI were <8. Patient was not a candidate for organ donation. Bedside cerebral blood flow monitoring may assist in determining the timing of brain death evaluation in comatose patients with imminent brain death. Patients with CFI <10 may be considered for brain death evaluation. Our finding needs further confirmation. Aneurysmal Subarachnoid Hemorrhage Patient. ICUs are high cost in the U.S., comprising about 17% of the US GDP. Pressure is being placed on hospitals and intensivists to reduce costs, including earlier palliative care engagement to shorten length of stay.. As the U.S. migrates to a value-based system, further pressure will be made on reducing prolonged and expensive ICU interventions, similar to quality adjusted life year (QALY) cutoff values to justify costs. A young 23 year old man presented with worst headache of his life, and was found comatose by EMS and referred to our NeuroICU. He had a 2.2cm giant basilar aSAH with intraventricular hemorrhage. The aneurysm was coiled endovascularly with external ventricular drain placed. He required therapeutic hypothermia, osmotherapy, induced hypertension and balloon angioplasty and intraarterial verapamil for refractory basilar and bilateral middle cerebral artery vasospasm. He had refractory intracranial pressure from global cerebral edema and around post-operative day #13 required bifrontal craniectomy. Later percutaneous tracheostomy, PEG tube, and ventriculoperitoneal shunting were performed. His total costs exceeded $300,000 U.S. dollars. One year later, his modified Rankin scale was zero, and he went to college. His QALY (quality adjusted life year) for the rest of his year was 0.833 given a utility of 1. His physicians felt he should live to a normal life expectancy of 77 years of age, Q is quality of life weight = 1 (perfect health, utility =1), L is residual life expectancy =54 more years. His QAL-Expectancy 2,3 is about 54 life-years gained which divided over his life span is about $5,555/year and less than the current CMS reported value of $9,500 per year. This case exemplifies how high cost care can be delivered to deliver cost-effective, high quality care and underscore the need for integrated high-complexity neuroICU care. Early mobility in the intensive care unit setting is associated with a number of positive effects including improved quality of life. Though there is a strong body of evidence supporting early mobility in medical intensive care units, the benefits of very early mobilization after acute stroke are yet unclear as early hemodynamic variability in patients with impaired cerebral auto regulation is of concern. Another potential barrier to early mobilization is the presence of an external ventricular drain (EVD) for cerebrospinal fluid diversion and intracranial pressure (ICP) monitoring. This case demonstrates hemodynamic and ICP responses to progressive, device assisted mobility interventions during the acute phase of intracerebral hemorrhage (ICH) in the setting of persistent elevations of ICP requiring two EVDs. A 55 year-old man was admitted to the Neuroscience Critical Care Unit with an acute thalamic ICH and intraventricular hemorrhage requiring placement of two EVDs. Starting on day 3 following ICH onset, the patient underwent progressive mobilization following the Johns Hopkins NCCU Activity and Mobility Algorithm. Range of motion exercises were performed initially, progressing to supine cycle ergometry followed by incremental verticalization using the tilt table (Sara Combilizer®Arjo Huntleigh Inc.,Il). Blood pressure, heart rate, oxygen saturation, and ICP were recorded before, during and after the mobility interventions. No adverse neurologic effects were noted during these mobility interventions. Recorded hemodynamic variables and ICP remained within the set goals throughout. Moreover the patient was able to tolerate 45 degrees of verticalization on the tilt table. Progressive, device assisted early mobilization was feasible when titrated by skilled healthcare professionals in a critically ill hemorrhagic stroke patient with EVDs. Studies on larger patient samples are needed to improve our understanding of the hemodynamic and neurophysiologic responses to establish safety of progressive early mobilization of critically ill patients with acute stroke. Anand Venkatraman, Ayaz A. Khawaja, Angela A. Shapshak. University of Alabama at Birmingham / Department of Neurology, Birmingham, AL, USA. We describe a case of a patient with uncontrolled hypertension (HTN) and prior intracranial hemorrhage (ICH) who developed an intracranial hemorrhage shortly after consuming Redline, a heavily-caffeinated energy drink. A 57-year old Caucasian male with prior history of ICH and chronic untreated HTN was transferred to our service for evaluation of 1.8x1.3cm ICH in the left thalamus. Blood pressure had been elevated in the Emergency room there and he had been started on a nicardipine infusion. NIH Stroke scale was 1. ICH score was 0. Admission labs were normal. Urine drug screen was negative. On questioning, patient revealed that symptoms had started within 2 hours of consumption of 1 bottle of Redline, an energy drink. He was not a regular user but did consume it whenever he needed to get a lot of work done. MRI of the brain did not show any vascular malformation or other lesion. Multiple remote hemorrhages were seen in the subcortical areas. We stared lisinopril and weaned off his nicardipine infusion. He was discharged with minimal deficits. The high caffeine content is the most likely component of the drink that led to the ICH, given that high caffeine consumption is linked to increased risk of hemorrhagic strokes. Caffeine also has effects on platelet aggregation and endothelial function that could raise stroke risk. Generalizability is limited by the fact that the patient had uncontrolled HTN and prior ICHs. However, given that 15% of adults in the USA have uncontrolled HTN, and given that ICH account for 10-15% of all strokes, a large population is at risk. To our knowledge this is the first report of intracranial hemorrhage following consumption of an energy drink. Consumers must exercise caution, especially in the setting of uncontrolled risk factors. Elena Schmidt, Varada V. Nair, Gene G. Latorre. SUNY Upstate University Hospital / Department of Neurology, Syracuse, NY, USA. Often times medications given in emergency cases have unintended consequences, sometimes posing even more harm than the reason for their administration. We report a case of a young lady with history of anaphylactic reaction who received i.m. epinephrine after developing allergic reaction to antibiotic, resulting in bilateral intracerebral hemorrhage (ICH). Our case is of a 42 year old female with history of anaphylactic reaction who had been recently started on Cephalexin for orbital cellulitis. She was sent to ED after having a syncopal event in the doctor's office. While in ED, the patient was administered 0.3 mg of 1:1,000 Epinephrine i.m. because of suspected anaphylactic reaction. Shortly after, she complained of nausea, vomiting and developed right sided weakness and numbness. Immediate CT head revealed two areas of ICH, within the left parietal and right occipital lobes. Extensive work-up ensued, with CTA head and neck (negative for vessel anomalies), cerebral DSA (negative for vasculitis), MRI brain w/wo contrast (negative for malignancy or amyloid angiopathy), CT thorax and abdomen (negative for malignancy). Serum studies for vasculitis work-up were also unrevealing. Echocardiogram did not show evidence for chronic hypertension such as LV hypertrophy. Although there have been reports in the literature of spontaneous intracranial hemorrhage (intraparenchymal or subarachnoid) after various ways of epinephrine administration, in our patient's case, the extensive work-up done to exclude other etiologies stands out. This strengthens the hypothesis that epinephrine, causing an acute spike in blood pressure, ultimately led to spontaneous ICH. In addition, the case of our patient is unique in that she developed two areas of ICH, in a location typical for Posterior Reversible Encephalopathy Syndrome (PRES), a syndrome known to be caused by significant elevation in blood pressures, suggesting a common underlying pathophysiology. Careful consideration of indications should occur before administering such potentially harmful treatments. "Last known normal" (LKN) time remains the standard for determining the onset of acute ischemic stroke and appropriateness of providing acute therapies. As older adults become more familiar with social media platforms, these applications may become a source of recognizing when a patient was LKN. We report an 82 year-old woman who was "found down" at home. The patient lived independently, and was able to crawl to a telephone for help. On arrival to the emergency department, she had a right middle cerebral artery syndrome with an NIH stroke scale of 17. She had a decreased level of arousal and severe dysarthria which precluded assessment of her LKN. The patient's son reported that he had last seen her normal 9 hours prior, placing her outside the time window for acute therapies. However, the patient's granddaughter reported that the patient had been logged into Facebook <1 hour prior to her admission "chatting" and commenting on photos. "Timestamps" of comments left on photos by the patient provided exact times of the patient's activity. The only logical means of being able to perform these relatively high-functioning tasks would have been if the she was normal at the time of posting, thereby establishing her LKN. The patient was treated with systemic t-PA followed by endovascular therapy for a proximal M1 occlusion. The patient had rapid improvement of her stroke symptoms. She was discharged home with an NIH stroke scale of 2. This patient's recent use of the social media was critical in determining the patient's LKN, leading to lifesaving acute stroke therapy. Providers should be aware that social media may serve as a useful source of symptom onset information. In this case, it led to good outcome and discharge home. Bilateral Recurrent Artery of Heubner (RAH) infarctions have been seldomnly reported in the literature. Even more so for those cases that have occurred subsequent to Neurosurgical extensive resections of large invasive Olfactory Groove Meningioma. RAH, a branch of the anterio-inferior cerebral artery, supplies anterior limb of the internal capsule, anterior caudate, putamen and globus pallidus. Infarction typically results in contralateral paresis of the arm and face. Other symptoms can occur i.e. choreiform movements, abulia, attention disorder, impaired memory, apathy, decreased spontaneity, depression, dementia etc. We present a case of Bilateral RAH infarcts as a complication of a large Olfactory Groove Meningioma resection. We did an extensive chart review of our patient during post-operative Neurointensive Care unit stay, rest of the hospital stay and discharge follow up at 1 month. Our patients Brain MRI done as a part of routine post-operative imaging showed bilateral caudate head infarcts in the territory of RAH. Post-operative exam was significant for a left hemianopsia and right super quadrantopsia with color desaturation. Patient did not experience any new weakness or movement related problems. He did have changes in cognition (forgetfulness & Irritability) along with a subjective loss of sense of smell but these were consistent with his pre-op assessment. Olfactory Groove Meningioma's comprise 10% of all intracranial meningiomas, are slow growing and tend to engulf and compress neighboring structures. Most common complications of Olfactory Groove Meningioma resections are post-operative cerebral edema, CSF leak, seizures, CNS infections, hydrocephalus and rarely brain ischemia. Bilateral RAH infarction, although rare has been reported in literature in association with vascular anomalies and other stroke risk factors. Cerebral infarction involving the ACA territories remains a known adverse complication of large olfactory groove meningioma resections, but bilateral infarcts due to these have not been reported before. Angioinvasive Aspergillus Associated Stroke In An Immunocompetent Host. Aspergillus vasculitis is an under-recognized cause of stroke in immunocompetent hosts, especially when other risk factors are present. We present a case of autopsy proven angioinvasive aspergillus causing strokes in an immunocompetent host, and review the characteristic imaging findings to aid diagnosis. 59-year-old female developed cardiogenic shock after three-vessel-Coronary artery bypass grafting (CABG) using saphenous vein grafts requiring intra-aortic balloon pump placement. This was complicated by aortic dissection, and she underwent replacement of the ascending aortic arch. Refractory cardiogenic shock ensued for which she underwent placement of veno-arterial extracorporeal membrane oxygenation. Postoperatively, she was noted to be in coma, and a non-contrast CT of the brain showed small multiple small ischemic strokes bilaterally. With persistent multi-organ failure, she was ultimately transitioned to comfort care and passed. Autopsy revealed multiple perivascular petechial hemorrhagic infarcts involving white matter, deep gray matter and cerebellum on gross specimen. Histopatholgic study showed Aspergillus associated acute and chronic inflammation of blood vessel, and surrounding gliosis. Aspergillus was also found in coronary grafts and kidneys. Aspergillus associated cerebral vasculitis was considered less likely, as CABG, Extracorporeal membrane oxygenation (ECMO) device-related thrombosis and acute MI were the leading differentials for stroke here, and no obvious immunosuppression was evident. Cerebral aspergillosis can occur from direct spread from sinus infections or through hematogenous mode, and seemed to have originated from coronary grafts in this case. Aspergillus has a predilection for posterior circulation arteries, and lacunar-type infarcts or petechial hemorrhages within the midbrain, thalami, or corpus callosum are characteristic. These findings should raise suspicion for aspergillus, especially without objective evidence of other mechanisms of stroke. Early initiation of anti-fungal therapy may improve the likelihood of survival, and confirmatory testing in the form of blood vessel imaging, CSF analysis and fungal blood cultures should be performed in suspected cases. Autoimmune ganglionopathy: a rare cause of cardiac arrest Kelly Braun. Neurological Institute, Cleveland Clinic, Cleveland, OH, USA. Autoimmune autonomic ganglionopathy is a rare disorder characterized by pandysautonomia that occurs as a result of autoantibodies to ganglionic nicotinic acetylcholine receptors. We describe a 27 year old male with autoimmune ganglionopathy previously treated with IVIG who suffered cardiac arrest and anoxic brain injury as a complication of this disorder. The patient had a history of multiple autoimmune diseases (DM1, autoimmune hepatitis, Hashimoto's thyroiditis, celiac disease, antiphospholipid syndrome and ulcerative colitis). To alleviate his pre-syncopal lightheadedness related to dysautonomia, he would typically kneel and place his head on his folded arms. The patient was found unresponsive in this position in PEA arrest. He underwent CPR followed by therapeutic hypothermia (34°C x 24h). The initial exam off sedation showed an obtunded patient with intact pupillary and corneal reflexes, but no tracking or command following. Though he moved all extremities spontaneously, the movements were not purposeful and had a choreiform quality. Notable labs were an elevated AChR ganglionic neuronal antibody (5.44 nmol/L on hospital day 8 and 3.10 nmol/L on hospital day 15; normal < 0.02 nmol/L). MRI brain showed symmetric diffusion restriction and FLAIR changes throughout the brainstem, thalami and cerebellum, however there was no cortical diffusion restriction. EEG showed generalized intermittent rhythmic slowing, which was maximal bifrontally. He was treated with Methylprednisolone 30 mg daily for 23 days and transitioned to Prednisone 40 mg daily. At the time of discharge to an acute rehabilitation facility, the patient followed most simple commands and moved all of his extremities against resistance, though he was noted to have diffuse hypotonia. To our knowledge, this is the first reported case of cardiac arrest attributed to autoimmune autonomic ganglionopathy. While rare, this is a serious complication of this syndrome. Bibhukalyani Das, Shantanu S. Shubham. Institute of Neurosciences Kolkata, Kolkata, India. Global burden of tuberculosis is still high particularly in developing world. India is the largest TB burden country accounting for 1/5th of the global incidence.CNS tuberculosis is the most severe form of infection with Microbacterium tuberculosis.Emergence of MDR(multi drug resistant) tuberculosis has compounded the risk and adverse outcome. Fatality rate of MDR TB meningitis is 57% with significant functional impairment in most of the survivors. Mortality >90% if patient is HIV positive. We report a case of 15 yrs old girl from Eastern India case of MDR -CNS tuberculosis with a protracted clinical course of 2 years. She developed a whole range of complications including Hydrocephalus, optochiasmatic arachnoiditis with secondary optic atrophy, multiple tuberculomas, cerebellar and brainstem tubercular abscesses and SIADH with Hyponatraemia. Our case is notable for few rare complications in the form of Transverse Sinus Thrombosis secondary to chronic meningitis necessitating oral anticoagulation . The patient also developed various side effects of long term ATT such as -(i) ATT induced Hepatitis. (ii) Moxifloxacin induced seizures and re-adjustment of antiepileptics due to interaction with antitubercular drugs. She was managed with 9 antitubercular drugs (5 first line drugs) along with second line drugs (Amikacin, Levofloxacin, Cycloserine, Ethionamide) Her clinical course was complicated by obstructive hydrocephalus requiring EVD, VP Shunt and Shunt Revision . Developed acute SDH possibly secondary to shunt and required surgical drainage . Later she developed posterior fossa tubercular abscess and needed craniectomy. Pus from tubercular abscess grew MTB resistant to Rifampicin and Isoniazid. So the patient was maintained on second line drugs, ultimately succumbed to hospital acquired pneumonia. CNS tuberculosis if associated with multiple medical, surgical complications, impose real critical care challenges compounded by MDR which often encountered in a developing country like India. Megan Lange, Rebecca R. Horrell. University of Maryland Medical Center, Neurocritical Care Unit, Baltimore, MD, USA. Super-refractory status epilepticus, defined as seizures persisting despite anesthetics, is associated with high morbidity and mortality. Here we present two cases of super-refractory status epilepticus intractable to aggressive therapies, including but not limited to anesthetics, electroconvulsive therapy, and immunotherapy. In both cases, the patients developed sepsis and cardiac arrest following prolonged hospitalizations with subsequent termination of seizure activity and improvement in electroencephalogram findings and neurologic exams. A review of the literature revealed a variety of publications describing super-refractory status epilepticus as a result of sepsis or cardiac arrest, but there is limited data describing either complication as therapeutic for status epilepticus. We propose that the systemic effects associated with profound sepsis, or the brief electrographic silence occurring in the setting of cardiac arrest could have played a role in halting seizures in these patients. We describe two theories regarding the potential mechanism by which cardiac arrest or sepsis could play a role in termination of seizures. Exploration into specific mediators involved in these conditions and their relationship to status epilepticus could uncover therapeutic targets. Targeted therapies could demonstrate promise in effectively treating super-refractory status epilepticus, thereby improving morbidity and mortality rates. Ticagrelor is approved for prevention of cardiovascular events in adults with acute coronary syndrome (ACS)1. We present a patient with SAH who developed thrombus during coiling procedure that was treated with abciximab followed by ticagrelor and aspirin, with potentially devastating consequences. A 64-year-old male presented after sudden onset severe headache. Imaging revealed SAH from a ruptured basilar tip aneurysm. The patient was neurologically intact with mild confusion, but declined during transfer and required intubation. An external ventricular drain was placed for hydrocephalus. During cerebral angiogram with coil embolization, a thrombus formed on the coil. Intra-arterial abciximab was used with resolution of thrombus. He was extubated post procedure. Aspirin and ticagrelor (90 mg twice daily) were prescribed. The following day, the patient became increasingly lethargic with an increased respiratory rate (30s). He reported no perception of increased work of breathing. Portable chest radiograph demonstrated only mild pulmonary edema. He did not have an oxygen requirement. Venous blood gas demonstrated a pH 7.52 with a pCO2 of 27, suggesting a respiratory alkalosis. Transcranial dopplers demonstrated normal velocities, but the patient was considered for cerebral angiogram given a high concern for vasospasm with his neurologic exam. Ticagrelor was stopped the following day. The patient's tachypnea and mental status rapidly improved. Ticagrelor reversibly inhibits the platelet P2Y12 adenosine phosphate receptor and is indicated for prevention of cardiovascular events in adults with ACS. In patients with both cerebral hemorrhage and a need for antithrombotic therapy, this reversible agent may become more widely used. Dyspnea is a known side effect of ticagrelor, occurring in 13.8% of patients (p<0.001).1 Dyspnea causes respiratory alkalosis and the resulting hypocapnea results in vasoconstriction. In this case, mental status change after administration of ticagrelor suggests that patients at risk for vasospasm may be particularly vulnerable to its side effects. Over the past 20 years, the Americas have experienced waves of emerging and re-emerging arboviruses that cause neuroinvasive disease, including West Nile virus, Chikungunya virus, Zika virus, and Dengue virus. These viruses pose great challenges for traditional candidate-based infectious disease diagnostics that already fail to identify a causative pathogen in approximately 50% of encephalitis cases. We present the case of a 14 year-old girl with a history of renal transplant managed with mycophenolic acid, tacrolimus and prednisone who presented to an emergency department with two days of high fevers, chills, upper back, neck pain and rash followed by encephalopathy. One month prior to presentation she attended summer camp by a lake in the Angeles National Forest, California. Her hospital course was complicated by status epilepticus. Cerebrospinal fluid (CSF) analysis demonstrated a mixed neutrophilic and lympocytic pleocytosis. MRI of the brain demonstrated symmetric T2 hyperintensities and edema in the bilateral thalami and leptomeningeal enhancement in the thalamus, cerebellum, brainstem, cervical spine and caudal equine. An extensive diagnostic work-up for infectious causes of encephalitis was performed and only identified Epstein-Barr virus. research protocol. Unbiased MDS of RNA extracted from her CSF and processed through a custom bioinformatics pipeline identified West Nile virus. Subsequently, convalescent serum serologies confirmed West Nile virus infection. This case provides a first proof-of-principle that MDS can detect even low level arbovirus burden in the CSF of a patient with acute meningoencephalitis. Given the rapidly changing landscape of viral causes of encephalitis in the Americas, the ability of MDS to comprehensively detect a huge array of microbes with a single assay may make it an optimal method for early identification of emerging causes of viral encephalitis, including in the transplant patient population. Cladophialophora bantiana is a dematiaceous mold with a predilection for causing central nervous system infection, particularly in normal hosts. There is no standard therapy and mortality rates from this disease remain extremely high approaching 70%. Here we describe a case involving a 59 year-old immunocompetent man who presented with new onset seizures. Brain imaging revealed bifrontal ring enhancing lesions concerning for abscess.The patient underwent surgical debridement of the lesions and bilateral intracavitary treatment with Amphotericin B using Ommaya reservoirs for several months. After approximately 16 months of treatment which included surgical debridement , oral voriconazole and intracavitary Amphotericin B, our patient is off all antifungals and no longer receiving intracavitary treatment. He remains fully functional with a nonfocal neurologic exam, being monitored with serial brain MRIs. Due to rare incidence of cerebral phaeohyphomycosis, there are no clinical trials to help formulate standardized treatment guidelines despite its high mortality . This case places emphasis on an early aggressive multimodal approach for treatment of cerebral phaeohyphomycosis using a combination of surgical debridement, intracavitary antifungal injection, and oral antifungal therapy. Does Neurocritical Care Need to Improve Outreach to non-Neuro Specialties? Firas Abdulmajeed, MB. ChB, Bart B. Nathan, MD, FCNS. University of Virginia/ Department of neurology, Charlottesville, VA, USA. The number of Neuro-critical care(NCC) fellowship positions has been increasing yearly. The number of applicants has plateaued over the last 3 years, leaving many programs with unfilled fellowship positions. The demand for neurointensivists is on the rise. Trainees have come from Neurology traditionally, with a limited number from Internal Medicine (IM). We hypothesize that the relative paucity of IM fellows was in part due to insufficient knowledge of NCC. We surveyed IM residency programs in the United States, asking program directors to forward our survey to their residents. On another survey, NCC fellowship directors were asked: 1) how many trainees and how many faculty hires from IM and/ or EM have they had within the last five years Internal Medicine residencies survey results: We obtained 102 individual responses, 98 responses were complete. How long is the NCC fellowship? N=100 12% 1 year 41% 2 years 47% 3 years Residents of what specialty can apply to a NCC fellowship? N= 97. 73% (neurology, neurosurgery, anesthesiology, IM and EM) 11% neurology only 15% neurology, neurosurgery and anesthesia Knowledge of San Francisco matching system? N=92 48% yes. 52% No Do you know about the application cycle for the NCC match? N=93 10% Yes. 90% No Knowledge of Emergency Neurological Life Support? N= 92 7% Yes 93% No NCC Fellowships' directors survey results: 20 of the 52 programs responded Fellows with IM/EM background that were trained within the last 5 years: 12/10 Neuro-Intensivists with IM/EM background hired: 6/7 IM residents appear to have little knowledge of NCC fellowship. A lack of awareness of ENLS could affect the quality of care provided for neurological emergencies. Additionally, for the specialty to grow and fill unmatched fellowship positions,current training and outreach strategies to non-neurology trainees may need to be improved Somatosensory evoked potentials (SSEPs) are a sensitive, minimally invasive technique used to identify injury from the posterior columns of the spinal cord to the somatosensory cortex. The role of SSEPs as a neuromonitoring tool, in the neuroICU has not been well established. We present a case using SSEPs as a neuromonitoring tool illustrating electrical improvement along with clinical and radiographical improvement in a symptomatic Chiari I malformation. 18 year old female who was 2 months postpartum after vaginal delivery with epidural analgesia presented with headaches, diplopia and nausea/vomiting. After arrival to ER, she acutely developed flaccid quadriparesis with ophthalmoplegia and loss of airway while awake and following commands. Given the concern for intracranial hypotension, we administered mannitol, hyperventilated and placed in Trendelenburg position. Head CT showed cisternal effacement in the setting of a likely pre-existing Chiari 1 malformation with cerebellar tonsillar. Decompressive surgery was not an option initially given her dysautonomia and neurologic instability whenever the patient was not in Trendelenburg. During her prolonged course, she had two MRIs of her brain and spinal cord which showed Chiari I malformation with syrinx at C4 and presyrinx down to T2. There was cervical spine venous engorgement and CSF block at the level of the foramen-magnum. She was evaluated with serial SSEPs which initially showed low amplitude N20 response that improved with her clinical improvement. SSEP is a minimally invasive method to electrically assess the somatosensory pathway integrity from the spinal cord, brainstem and cortex. Given its sensitivity to the function of the dorsal columns of the spinal cord and medial lemniscus of the brainstem, SSEPs may be a useful monitoring adjunct to follow the evolution of posterior fossa lesions in patients that may not tolerate other means of monitoring and/or transportation, such as MRI. Review of prospectively maintained patient database identified one case of status epilepticus in a patient with CNS-PTLD. We present a case report with literature review. A 74-year old Hispanic woman with a history of renal transplant 19 years prior, presented with episodic confusion and gait ataxia progressing over two weeks. She was on immunomodulation with Mycophenolate. Patient had witnessed periods of behavioral arrest. Continuous electroencephalography (cEEG) demonstrated right temporal sharps and 8-20 second epochs of bi-frontal 14-18 Hz activity, some of which were associated with non-stereotyped movements of her left shoulder and trunk, suggestive of SE. She received benzodiazepines followed by Levetiracetam (renal dose) and Phenytoin load for seizure control. MRI brain without contrast demonstrated multifocal infiltrative T2-hyperintense white matter lesions, most prominent in right temporal lobe. CSF analysis demonstrated 2 RBC, 63 (L 88%) WBC, 80 protein, 71 glucose, culture and gram stain were negative. There were 4 unmatched CSF bands with an unremarkable cytology. CSF PCR was positive for EBV and viral load was detected at 1458 copies/nL. Other CSF microbial assays including JCV were negative. Stereotactic right temporal brain biopsy demonstrated areas of necrosis, axonal disruption, loss of myelin with polytypic plasma cells, CD20 and CD30 positive B cells and CD3 positive T cells on immunohistochemistry consistent with a diagnosis of polymorphic PTLD. Despite treatment with dexamethasone and rituximab, patient continued to remain critically ill and eventually received palliative measures. Among transplant recipients, PCNS-PTLD is rare but debilitating with varied neurological presentation. High degree of suspicion, early diagnosis and treatment are paramount for survival. Cortical Myoclonus caused by activation of cortical areas subjacent to multiple subdural hematomas is an unusual mechanism of epilepsy. We report the case of a patient with an extra axial bleeding and myoclonic seizures evaluated with ictal FDG-PET. Case Report A 59 year old male was admitted to our hospital because of worsening symptoms of cardiac failure of Chagasic etiology. Due to hemodynamic instability he was treated with the placement of an intraortic balloon pump. Sixteen days after hospital admission, he presented intermittent generalized myoclonic jerks. On initial examination he was alert and oriented to time and place, had preserved strength in all limbs, although presenting with very frequent clusters of myoclonus. Initial investigation with a head CT showed multiple foci of extra axial bleeding, distributed over the frontal and parietal areas. The electroencephalogram (EEG) confirmed the suspected diagnosis of myoclonic seizures, exhibiting generalized polispike-slow wave complex. Due to the multiplicity of bleeding sites, with no obvious reason for spontaneous bleeding other than regular anticoagulation, the patient was submitted to a whole-body FDG-PET in order to exclude the possibility of dural metastatic implants. FDG-PET showed areas of cortical hypermetabolism adjacent to the bleeding foci, probably reflecting an epileptogenic mechanism of cortical activation. There was no evidence of hypermetabolism directly over the extra-axial areas of bleeding, what ruled out the hypothesis of dural metastasis. A diagnosis of spontaneous subdural hematomas associated to anticoagulation was given after all other causes were excluded. The patient was treated with sodium valproate and had sustained improvement of the myoclonic seizures. The bleeding areas were eventually reabsorbed, but the patient died from complications of cardiac failure. Our report is the first to illustrate the mechanism of cortical activation leading to epileptic status in a patient with multiple subdural hematomas detected by ictal FDG-PET. To present a case of early onset myoclonic status epilepticus (MSE) after cardiopulmonary arrest with incomplete resolution of myoclonus and good cognitive outcome. A 20 year-old man presented status post cardiopulmonary arrest and CPR in the field with return of spontaneous circulation (ROSC) after arrival to the hospital and cardioversion. The patient was intubated and treated with therapeutic hypothermia, but developed clinical MSE with normal EEG within 12 hours. He was aggressively managed with propofol, levetiracetam, and fentanyl. Initial MRI demonstrated diffuse hypoxic ischemic injury. MRI on the 9th day of admission demonstrated improvement but a new white matter lesion in the splenium of the corpus callosum. After multiple unsuccessful attempts to discontinue fentanyl and 15 days of treatment, the patient was given a poor prognosis based on the 2006 AAN MSE Practice Parameters and was placed on do not escalate care orders. Propofol was slowly decreased; however the patient improved significantly throughout hospitalization with improved language and cognitive examination and only mild residual reflex myoclonus at the time of discharge. MRI imaging had completely resolved by the 19th day of hospitalization. The patient's final diagnosis is Lance-Adams syndrome of action myoclonus incompletely controlled with levetiracetam. Aggressive and prolonged treatment including therapeutic hypothermia in young patients with early onset MSE was effective despite AAN practice parameters. This patient survived with good cognitive outcome and with relatively modest deficits. Further research is needed to assess whether improvements in intensive care unit capabilities over the past decade may contribute to improved outcome in young patients with cardiac arrest and whether practice parameters should be revised. At the start of this protocol, the serum sodium was 148 and one-hour urine output was 1 liter. This protocol was continued for 48 hours. Endocrinology was consulted and recommended changing to DDAVP. Serum sodium was 149[NP1] and one-hour urine output 320 cc prior to first dose of DDAVP University of Washington, Department of Surgery, Seattle, WA, USA. Earlier feeding results in improved outcomes in adults with severe traumatic brain injury (TBI) and in the overall pediatric intensive care unit (PICU) population. Current practices of nutrition initiation in children with TBI are not well described. This multicenter study evaluated timing and factors associated with nutrition initiation in children admitted to PICUs with TBI. We hypothesize that severely brain injured patients would have a delay in initiation of enteral nutrition. We retrospectively analyzed the multicenter Pediatric Trauma Assessment and Management Database (PTAM) from 2013. Patients with severe TBI were defined as Glasgow Coma Scale (GCS) <9 with n in this group was compared injury, abdominal procedures were compared between the two groups. Chi square and Fisher exact tests were used for dichotomous variables; non-parametric tests were used for continuous variables. Multivariable regression analysis with a stepwise procedure was performed to ascertain the best set of variables associated with delayed initiation of enteral nutrition. Of 423 patients admitted to the five PTAM PICUs with severe TBI, 285 (67%) were fed < 24 hours from admission. Patients with GCS <9 were fed a median 45.1 hours from admission (IQR 22.0-72.6) compared to 10.3 hours (IQR 3.8regimen, higher injury and illness severity scores and lower minimum GCS were significantly associated with feeding initiation > 24 hours. On multivariable analysis, scheduled bowel regimen, higher PRISM score and lower minimum GCS were significantly associated with nutrition initiation > 24 hrs. Lower GCS is independently associated with delayed initiation of enteral nutrition in children with TBI, independent of severity of injury or abdominal injury. All patients that suffered CA within 24 hours of SAH onset were identified from a prospectively collected characteristics, and outcomes of those with and without CA in the setting of SAH using binary logistic regression. Only 2% (N=38) of SAH patients had CA within 24 hours of the bleed. 55% (N=21) of those with CA had f these patients died while in the hospital. Three patients had a ventricular fibrillation (VFIB) arrest, and one of these patients (N=14), and half of these patients survived. 79% of patients were comatose after the arrest, most of which underwent cooling (goal temperatures 32-34). 37% of deaths in our cohort were from withdrawal of life support (N=11). Increased aneurysm size (OR 1.08 for each 1mm, 95% CI 1.01-1.15), amount of SAH (OR 4.97, , and global cerebral edema (OR 3.35, CI 1.06-9.65) were associated with Noncomatose patients and those with VFIB arrests may have a better prognosis. Acute herniation at the time of bleeding as indicated by large volume SAH and global cerebral edema may be the underlying mechanism of most early cardiac arrest in SAH patients. Early identification of delayed cerebral ischemia (DCI) following aneurysmal subarachnoid hemorrhage (SAH) could allow more effective intervention. Statistical methods that predict DCI using variables collected routinely during ICU care such as trends in vital signs and laboratory values have shown promise in recent studies. However, these studies have not all employed methods to guard against model overfitting. In this study we use cross validation to obtain minimally-biased estimates of the value of passively collected ICU variables for predicting DCI. Early identification of delayed cerebral ischemia (DCI) following aneurysmal subarachnoid hemorrhage (SAH) could allow more effective intervention. Statistical methods that predict DCI using variables collected routinely during ICU care such as trends in vital signs and laboratory values have shown promise in recent studies. However, these studies have not all employed methods to guard against model overfitting. In this study we use cross validation to obtain minimally-biased estimates of the value of passively collected ICU variables for predicting DCI. DCI occurred in 34% of patients. Penalized logistic regression selected 9 features for inclusion in the final predictive model, derived from GCS, heart rate, mean aterial blood pressure, respiratory rate, SpO2, ventricular drainage, and sodium data. The mean AUC of the model was 83%. Potentially clinically relevant (sensitivity, specificity) points on the ROC curve included (73,89)% and (90,52)%. DCI occurred in 34% of patients. Penalized logistic regression selected 9 features for inclusion in the final predictive model, derived from GCS, heart rate, mean aterial blood pressure, respiratory rate, SpO2, ventricular drainage, and sodium data. The mean AUC of the model was 83%. Potentially clinically relevant (sensitivity, specificity) points on the ROC curve included (73,89)% and (90,52)%. Subarachnoid hemorrhage (SAH) remains a highly morbid disease leading to > -related year of life lost before age 65. Mechanisms of SAH-related early brain injury and vasospasm remain MicroRNA (miR)-26a is released in response to hypoxia and promotes angiogenesis. We hypothesize that higher levels of miR-26a is associated with outcome in human SAH. functiona -up every 3 months. Good functional outcome is defined as mRS50% reduction in caliber of any vessel on post-SAH day 7 cerebral angiogram. In 56 SAH subjects we compared CSF and plasma miR-26a by quantitative PCR on post-SAH days 1, 3 and 5 between outcome groups. Data are normalized using log-transformation and then compared using student's t- Study population has mean age of 55. 62% has Hunt and Hess (HH) grade >3. Good outcome at 6 months is associated with higher plasma miR-26a levels on post-SAH day 3 (p=0.0007) and day 5 (p=0.04). After adjusting for important predictors of outcome (HH grade; age), plasma miR-26a on post-SAH day 3 remains strongly associated with outcome (p<0.0001). Plasma miR-26a levels were not associated with vasospasm. MiR-26a is present in CSF and is elevated in SAH compared to controls (p<0.0001), but CSF miR-26a showed no association with functional outcome or vasospasm status. Higher plasma miR-26a level at post-SAH day 3 is independently associated with 6-month SAH outcome. Mechanistic experiments are necessary to determine whether miR-26a expression is neuro-protective in SAH. Validation studies in larger, independent cohorts are necessary to validate miRNA-26a as a Accurate assessment of renal function remains a unique challenge in patients with aneurysmal subarachnoid hemorrhage (aSAH). Mathematical estimates of creatinine clearance (CrCl) routinely used are often inaccurate in this setting. Patients with aSAH have been shown to exhibit a hyperdynamic response leading to an enhanced renal clearance. No studies exist evaluating the directly measured creatinine clearance of patients with aSAH over time. This was a single-center prospective observational study of adult patients with aSAH admitted to the NSICU between January 2015 and July 2015. Eight-hour urinary creatinine clearances were performed daily to directly measure CrCl until the patient no longer had a foley catheter or the patient left the NSICU.-Gault equation. Statistical significance was defined as p-value < 0.05. Fifty patients with aSAH were enrolled in the study. The study sample was 68% female with a mean age of 57.2±10.7 years. The median Hunt and Hess grade was 3 (IQR 2-4) and the median modified Fisher grade was 3 (IQR 3-4). Additionally, the median admission GCS was 12.5 (IQR 6-14) and median admission SOFA score was 2 (IQR 2-4). The mean urinary CrCl over the study period was 147.9±50.2 Patients with aSAH consistently experienced urinary CrCl greater than estimated CrCl predicted based on -Gault equation. As renally eliminated medications are routinely dosed based on mathematical estimates of renal function, further study is needed to optimize medication regimens in this patient population to prevent underexposure. Agitated delirium is frequently encountered after acute brain injury, but data is limited in patients with nces of agitation in these patients. via records of antipsychotic or dexmedetomidine administration, and agitation was confirmed via chart study team. Outcome was assessed at 12 months using Interview for Cognitive Status (TICS), and Lawton-IADL score. Agitation developed in 52 of 309 patients (16.8%) and was most common in the first 72 hours after admission, and in patients with Hunt and Hess grades 3 and 4. Agitated patients were significantly more in half of these patients a complication appeared to occur within 24 hours of the onset of agitation. Patients with agitation had increased ICU and hospital lengths of stay, but this was not significant after controlling for other predictors of length of stay. For patients with Hunt and Hess grades 1-4, agitation was not independently associated with functional impaired at 12 months compared to those without agitation after controlling for other predictors (Lawton >8; p = 0.03, OR 2.7, 95% CI 1.1-6.8). Patients with SAH frequently experience agitation requiring medical treatment, especially early in their clinical course, and especially in non-comatose patients with higher clinical grades. Agitation is also associated with the development of multiple hospital complications, and may have an independent impact on long-term outcomes. Seizures after subarachnoid hemorrhage (SAH) are a frequent complication. SAH patients are typically prescribed prophylactic anti-epileptic drugs (AED) for three to seven days. Phenytoin has fallen out of favor as AED prophylaxis due to its association with worsened outcome as well as drug interactions. Newer AEDs including levetiracetam are more commonly used despite an incomplete understanding of their effect on outcome. Retrospective analysis was performed of prospectively collected data for 97 consecutive SAH patients enrolled into the University of Maryland REcovery After Cerebral Hemorrhage (REACH) Study between -Hess (HH) and modified Fisher score (mFS) was adjudi a team of neurointensivists. Retrospective analysis of cumulative dose of levetiracetam was divided into groups of low-dose (=12,000mg) using the median as a cutoff. Concordance and discordance was noted. Pearson Chi-Square was used. Association of levetiracetam dose and quetiapine use as a surrogate of in-hospital delirium was also investigated. Multi-variate logistic regression was used to determine predictors of ability to perform activities of daily living (ADLs) in 78 survivors. 97 aSAH patients from July 2014 to November 2015 were reviewed for this study. Cumulative levetiracetam dose was calculated in 89 patients and dichotomized into high-dose (>=12,000mg) or lowdose groups. Hunt-Hess was found to be significantly associated with high-dose levetiracetam. On multivariate analysis, high-There is a trend towards increased use of quetiapine in the high-dose levetiracetam group. Full analysis will be provided at time of presentation. An extended course of levetiracetam is an independent predictor of loss of independence in activities of daily living after SAH. There is also a trend toward increased delirium. Larger, prospective studies are necessary for a more complete understanding of the impact of seizure prophylaxis on functional outcome after subarachnoid hemorrhage. Myocardium: A Case Series and Review of the Literature. Intra-aortic balloon pump (IABP) counterpulsation has been used to maximize cerebral blood flow in patients with subarachnoid hemorrhage (SAH), refractory vasospasm and evidence of cardiac dysfunction. Neurogenic stunned myocardium (NSM) pr LV dysfunction. We present 9 cases with SAH, vasospasm and IABP placement, including 7 cases with NSM. We also reviewed the literature with the goal of examining the safety of IABP for cardiac dysfunction after SAH, outcomes and selection criteria for its use. We searched for cases of SAH and IABP placement at the University of Kansas Medical Center (KUMC) from 2008 to 2015. 9 patients met criteria and all had a secured aneurysm, refractory vasospasm and echocardiograms prior to IABP placement. We collected demographics, vitals, EKG, troponin, medications, IABP and ICU complications, discharge and follow-up mRS. However, at follow-r outcome. Literature review identified 17 patients -up. Our results indicate that patients that have IABP placement in the setting of SAH, vasospasm and cardiac dysfunction may have a good outcome if they are younger, have evidence of reversible NSM and avoid ICU complications including PE, UTI and Sepsis. The patients level of mobility and independence at discharge may not be indicative of overall functional improvement. A significant complication of non-traumatic sub-arachnoid hemorrhage (NT-SAH) is the development of delayed cerebral ischemia associated with cerebral vasospasm. Milrinone, an inotrope and a phosphodiesterase inhibitor, has been used intravenously, intra-thecally and intra-arterially as a delayed cerebral ischemia treatment and prophylaxis. The purpose of the current study is to systematically review the available evidence on its efficacy for that indication. Articles from MEDLINE, EMBASE, Cochrane Library, clinicaltrials.gov, reference lists of relevant articles, and gray literature were searched. Study selection criteria were used and strength of evidence was graded. Neurological outcomes and side effects were assessed. Of 158 articles identified, 18 studies met the selection criteria and analyzed. The level of evidence varied and was generally low. This systematic review helped determine the current state of evidence for the efficacy and safety of Milrinone in the management of delayed cerebral ischemia in the context of NT-SAH. The available evidence is promising but of generally low quality suggesting the need for a randomized controlled trial. Blood lactate variability: a strong independent predictor of neurological outcomes in patients with aneurysmal subarachnoid hemorrhage Blood lactate levels during intensive care unit (ICU) management of patients with aneurysmal subarachnoid hemorrhage (SAH) can be used as an indicator of not only volume status but also aerobic glycolysis caused by excessive catecholamine levels and impaired lactate clearance. To determine whether blood lactate variability (LV) can predict neurological outcomes in patients with SAH, we assessed the standard deviation (SD) of blood lactate level of each patient during ICU stay. We retrospectively reviewed all patients at the age of 18 years or older who were consecutively hospitalized in Kagawa University Hospital with SAH and at least five arterial lactate measurements between January 1, 2009 and May 31, 2015. Patients were divided into two groups with a mean lactate to identify independent predictors of unfavorable neurological outcome. Unfavorable neurological outcomes occurred in 44.2% of a total of 122 patients. In both groups, there were increases in unfavorable neurological outcomes with increasing SD of lactate (quartile 1, 25%; that SD of la correlated with unfavorable neurological outcomes (p < 0.01). Multiple logistic regression analysis showed that SD of lactate (odds ratio, 9.07; 95% confidence interval, 2.06-47.6, p < 0.01), age, and H&K grade were independent predictors. This study demonstrated that increased LV was an independent predictor of unfavorable neurological outcomes in patients with SAH. The main causes of mortality and morbidity after aneurysmal subarachnoid hemorrhage (SAHa) are rebleeding and delayed cerebral ischemia secondary to cerebral vasospasm. The use of milrinone, an inotropic and vasodilator agent, is described in as one option to treat vasospasm in patients with refractory symptoms. Our objective was to describe the experience of our neurocritical care service with the use of milrinone in accordance with the Montreal Protocol for patients with refractory vasospasm. A retrospective study based on data obtained from medical records of patients suffering from SAHa and refractory vasospasm treated with milrinone from February 2015 to February 2016. From 70 SAHa patients admitted to our hospital during the study period, 8 were identified with refractory were female and 50 % of patients were pre-hypertensive. A total of 62% of the patients had Hunt-Hess scores between 1-3 and 75 % scored 3 or 4 in the modified Fisher scale. Vasosespam was identified after --9.2 days. In 75% of the patients hypertension was induced with norepinephrine as an initial treatment. The mean duration of the treatment -4.6 days. Two cases were treated with intra-arterial milrinone and angioplasty. The most common adverse event during the use of milrinone was hypotension (50%). Death occurred in 2 patients. Favorable functional outcome at the discharge was observed in 37% of the cases. In conclusion, the use of milrinone seems to be a safe option in the treatment of delayed cerebral ischemia secondary to vasospasm, especially in services where the availability of endovascular treatment is not a routine. (SAH). However, pathomechanism and etiology of this elevation leading to poor outcomes remains uncertain. This study investigated the effect of troponin elevation on multi-organ dysfunction and outcomes in patients with SAH. admitted to the Neuroscience Intensive Care Unit from July 2013 to January 2016. Among 307 patients, 209 patients were eligible for inclusion with investigation of serum troponin level at admission. Troponin elevation (> elevation were older (59.1 ± 14.4 vs 53.0 ± 12.8 years; P < 0.01) and more often had a loss of consciousness (31.0% vs 16.7%; P < 0.01), symptomatic hydrocephalus (42.3% vs 23.2%; P = 0.01), and a higher Hunt-Hess score (3.4 ± 1.1 vs 2.6 ± 0.9; P < 0.01) and modified Fisher score (3.4 ± 0.6 vs 3.0 ± 0.7; P < 0.01) at ictal period. During hospitalization, patients with troponin elevation more often had a respiratory failure (31.0% vs 18.1 dysfunction (21.1% vs 8.0%; P = 0.01) and more often treated with vasopressure (25.4% vs 13.8%; P = 0.01) and longer duration of mechanical ventilation (3.7 ± 6.2 vs 1.6 ± 3.3 day; P = 0.04) than those without troponin elevation. Troponin elevation in the acute stage of SAH is associated with multi-organ dysfunction. Thus troponin To mitigate early brain damage in subarachnoid hemorrhage (SAH), we have been treating World Federation of Neurological Surgeons Grade (WFNS) Grade 5 patients with therapeutic hypothermia (TH) for 7 days immediately after onset. Management after rewarming was problematic since fever in SAH is associated with vasospasm and poor outcome. We studied the feasibility and safety of endovascular cooling to maintain prophylactic normothermia following initial TH in patients with severe SAH. TH (core body temperature 34.0 °C) was initiated, using surface cooling, immediately after the diagnosis of WFNS Grade 5 SAH was made. The ruptured aneurysm was surgically clipped as soon as feasible. Around postoperative day 7, after rewarming to 36 °C, an endovascular catheter with 2 cooling balloons jugular vein and connected to XP® Temperature Management System (Asahi Kasei ZOLL Medical Corp.) for 7 days. Prospectively collected data were analyzed. 6.7 days. Nine patients developed shivering with increased temperature and were given acetaminophen and dexmedetomidine. There was no evidence of vasospasm or additional cerebral infarction during endovascular cooling, and no catheter-related sepsis or thromboembolic event. After removal of the cooling catheter, vasospasm-related cerebral infarction and fatal bacterial meningitis related to spinal drainage occurred. Three-month outcomes were good recovery (n=2), moderate disability (n=4), severe disability (n=2); vegetative state (n=1), and death (n=2). Elimination of fever burden in the first 14 days after onset was safe and feasible with combined surface and endovascular cooling in patients with WFNS Grade 5 SAH. disease processes. This study examines: 1) the relationship between admission lactate and the clinical and radiographic severity of aSAH, and 2) whether levels predict outcomes including vasospasm, delayed cerebral ischemia (DCI), and inpatient mortality. This is a retrospective analysis of 75 consecutive aSAH patients with lactate drawn on admission.compared to those with normal levels. Differences between groups were compared using chi-square tests for categorical variables, and independent t-tests for continuous variables. Spearman correlations were calculated between lactate levels and mean values for continuous variables. Elevations in lactate were associated with admission GCS, Hunt & Hess (HH) grade, Fisher score, serum white blood cell count (WBC), troponin I (TN), glucose, and ventilator-free days (VFD). Positive correlation was found between lactate and HH, Fisher score, WBC, TN, and glucose. An inverse correlation existed between lactate and GCS, and VFD. Compared to survivors, non-survivors had significantly higher lactate levels. All results were considered significant with a p-value <0.05. No association between lactate and the development of vasospasm or DCI was found. Higher admission serum lactate is positively correlated with HH grade, Fisher score, serum WBC, TN and glucose, but negatively correlated with admission GCS and VFD. Presence of an elevated lactate was also predictive of inpatient mortality. This is the first report of correlation between early lactate and aSAH severity, and conflicts with prior results suggesting an association between lactic acid and the development of DCI. Further studies are needed to determine whether lactate elevations relate to hypovolemia, acute inflammatory response, elevated sympathetic outflow, or other cause. with malignant cerebral infarction with high osmotic pressure therapy. Some patients need decompressive hemicraniectomy or expire due to cerebral herniation after TTM. Hence this study was performed to determine associated factors in case of failure of TTM in patients with malignant cerebral infarction. From January 2011 to December 2014, a study was performed in patients with malignant cerebral infarction occurred within 24 hours at neurological intensive care unit. All patients were diagnosed hyperosmotic fluid therapy and TTM. We defined failure of TTM to cases treated by decompressive hemicraniectomy or death due to brain herniation. A total of 21 patients, TTM were failed in 12 patients. Failures of TTM were common in patients without recanalization after thro ( Fever occurs in 20-50% of critically ill neurological patients, and small temperature elevations are correlated to increased morbidity. It is therefore crucial to acutely control the temperature of such patients. Systems currently available are resource intensive and not always readily available in units, resulting in delays in treatment. EMCOOLS pads and are composed of multiple cooling units filled with graphite and water, with an adhesive underside that allows for efficient heat transfer. Pads are stored at -18 C, and are available for immediate use. The retrospective analysis of the EMCOOL device included all subjects that had the device applied in the Neuroscience Intensive Care Unit (NSICU), with consistent temperature data recorded. Preliminary subarachnoid hemorrhage (50%), intracerebral hemorrhage (17%), subdural hemorrhage (17%), and pituitary tumor (17%). All subjects were febrile ( treatment period. The bedside shivering assessment scale was recorded at each application. C, T60avg = 37.9 C) drop in temperature at 60 mins C) achieved at 55 mins. Unconscious patients displayed a much higher rate of cooling at T60 as C). Of the total subjects, 33% had shivering events upon application (BSAS 2), device. Preliminary results show the EMCOOLS pads are an effective and safe method to control temperature elevations in neurologically critically ill patients. All even numbered posters will present on Friday All odd numbered posters will present on Saturday All poster sessions are in Prince George Exhibit Hall A from 5:30pm Superior sagittal sinus thrombosis (SSST) accounts for only 0.5-1% of all strokes, with a traumatic etiology representing an uncommon occurrence. Current guidelines advocate treating SSST with anticoagulation regardless of etiology, though efficacy is controversial and not yet studied in the traumatic brain injury (TBI) patient population. We recognize the importance of alternate treatment modalities of post-traumatic SSST, particularly surgical alternatives, and the dilemmas faced with anticoagulation therapy in the trauma population. We report a case of a 22-year-old male admitted with SSST who suffered severe TBI secondary to a pedestrian versus automobile collision. Imaging demonstrated bifrontal and right temporal lobe hemorrhagic contusions, scattered subarachnoid hemorrhage, diffuse cerebral edema, multiple nondepressed skull fractures, and SSST. On post trauma day two, the patient clinically deteriorated; an external ventricular device (EVD) was placed, and therapeutic heparin drip was started, despite the presence of intracranial hemorrhage and risk of EVD-related hemorrhage. The patient developed refractory elevated intracranial pressure (ICP) mandating initiation of pentobarbital to achieve burst suppression on continuous electroencephalography (CEEG) and serial administration of 23.4% hypertonic bolus and mannitol for two weeks. Hemicraniectomy and endovascular treatment were entertained though not pursued due to anticipated complications associated with concomitant anticoagulation therapy. Anticoagulation was briefly interrupted for EVD removal on post trauma day 17. He was extubated on post trauma day 20 and transitioned to warfarin. Repeat imaging showed complete recanalization of the superior sagittal sinus. The patient was discharged to inpatient rehabilitation after a 28-day hospital course. Management of SSST secondary to TBI remains controversial as these patients present with multiple confounding factors, further complicated by the lack of treatment guidelines. Further studies are needed to determine which independent or combined medical and surgical treatment modalities will decrease morbidity and mortality in this patient population. Takotsubo cardiomyopathy (TC) is known to occur in patients with subarachnoid haemorrhage (SAH) but is rarely reported in patients with traumatic brain injury (TBI). Here we present a TBI patient with complicated clinical course developing severe TC and compared to previously published reports. Case report and literature review. A 73 years-old-woman was admitted to our tertiary care hospital because of TBI with admission Glasgow Coma Scale score of 14. Computed tomography (CT) scanning of the brain revealed an acute subdural hematoma and traumatic SAH over left hemisphere and a small left frontal hemorrhagic contusion. Six hours later she deteriorated and head-CT showed significant progression of right frontal hemorrhage with intraventricular expansion and a midline shift. Hematoma evacuation was immediately performed. Postoperatively the patient developed cardiogenic shock necessitating an increasing dose of noradrenaline, neosynephrine and dobutamine to achieve a cerebral perfusion pressure of >65mmHg. Echocardiography demonstrated severe left ventricular myocardial dysfunction suggestive for TC, supported by raised troponin-T and NT-proBNP levels and abnormalities in ECG. Continuous infusion of levosimendan was added and maintained for 28 hours which led to stabilization allowing noradrenalin to be decreased the following days. Repeated echocardiography 8 days later showed normalized cardiac function. The patient's condition gradually improved and was extubated after 11 days fully awake with mild left facial-brachial weakness. Here we present a complicated case of TC with TBI developing cardiogenic shock within 24 hours of admission. We will compare the patient's TC characteristics and clinical course with published cases (N = 16) of TC with TBI. Further studies of TC in patients with TBI and the utility of levosimendan is warranted. Management of post-operative central diabetes insipidus (DI) is focused on replacing urine output with free water. This may not always be sufficient, and desmopressin (DDAVP) is needed. The use of DDAVP, however, is known to cause profound changes in sodium particularly if the triphasic response postpituitary surgery is occurring. Herein, we report a case using a dilute vasopressin bolus protocol in managing hypovolemia in acute, post-operative, central DI. Case report comparing two protocols for DI management. Statistical analysis was performed on serum sodium, urine specific gravity, and urine output using student t-test. P<0.05 was considered significant. Anti-programmed cell death 1 (PD-1) antibodies are an effective treatment option for NSCLC and other cancer entities. Anti PD-1 antibodies including nivolumab can induce immune-related adverse events (irAEs) in a number of organ systems. Neurological irAEs can be life-threatening and necessitate appropriate investigation and management by a neurologist. Myasthenic syndromes have rarely been described. Here we present a case of a man who developed a Lambert Eaton myasthenic syndrome thought to be a nivolumab-related immune adverse event. A 90-year-old man on nivolumab for metastatic NSCLC developed asymmetrical ptosis followed by facial diplegia, dysarthria and dysphagia and fatigable limb weakness. He had clinical features of Lambert Eaton myasthenic syndrome with striking truncal, shoulder and pelvic girdle fatigable weakness that demonstrated a temporary increase in strength during the first few contractions. He developed neuromuscular respiratory failure requiring noninvasive positive pressure ventilation. His muscle reflexes were initially absent and after repeated attempts could be elicited. His MRI brain was normal, CSF was within normal limits and his serum and CSF antibodies against Ach Receptors, MUSK and Voltage gated Calcium Channels were negative. He weakened after an initial 1mg/kg dose of prednisone. Plasma exchange resulted in a significant improvement of his weakness and the ability to wean the patient off noninvasive positive pressure ventilation. It is important to recognize that neurologic immune-related adverse events associated with nivolumab can cause Lambert Eaton myasthenic syndrome. Early recognition and aggressive treatment with plasma exchange can be life-saving. Neuroendocrine tumor (NET) of middle ear is extremely rare. Many names have been ascribed to these seemingly benign neuroendocrine lesions including middle ear adenoma (MEA), adenomatous tumor and carcinoid tumor (CT). 'Neuroendocrine adenoma' has also been used to better describe the histologic nature of these tumors. Here we present the first case of Carcinoid tumor of middle ear (CTME) complicated by dural sinus thrombosis and bony metastasis. A 73 year old man presented with right sided facial palsy for 4 hours and progressive hearing loss for 2 months. Physical examination revealed perforated right tympanic membrane with visible purulent material. CT scan of the head showed a small right cerebellar infarct. The MRI demonstrated right cerebellar hemorrhagic venous infarct and a heterogeneous middle ear mass extending into the internal auditory canal. MRV revealed thrombus in the right sigmoid and transverse sinus.Cervical spine MRI revealed diffuse osseous metastases .The histology and immunohistochemistry (IHC) after surgical resection was consistent with CTME. Proliferation rate of>75% was seen by Ki67 staining. He suffered massive intracranial bleed on heparin therapy and passed away. The NETs of head neck region are divided in 3 categories based on histology and IHC. 1) Well differentiated CT, 10/10 HPFs and Ki-67>20%. Although regional metastasis is not uncommon, only 2 cases of distant metastasis have been reported in the past. Our patient was diagnosed to have atypical carcinoid but the Ki-67 was >70% which is consistent with small cell cancer. The new classification system that takes the IHC and presence of metastasis into consideration to classify these tumors is much more clinically relevant. More research is necessary to find out the biological markers for better prognostication of this rare cancer. Sung M. Cho.Neurological Institute, Cleveland Clinic, Cleveland, OH, USA. Acute transverse myelitis is an acquired inflammatory spinal cord disorder, which can be due to infection, autoimmune disorders, or malignancy, however, the cause is often unknown despite an extensive workup. We describe a rare case of acute transverse myelitis caused by Coxsackie B3/B4 virus. A 36-year-old male with no past medical history presented with a viral prodrome of flu-like symptoms followed by severe headache, neck stiffness, photophobia, encephalopathy and paraplegia. MRI of the brain with contrast was unremarkable, but MRI of the spine showed an extensive longitudinal, nonenhancing T2 cord signal change from C6-T11 without hemorrhagic components. Lumbar puncture revealed 203 RBC, 214 WBC, 143 protein, and 46 glucose (95 mg/dL serum glucose). Extensive serum and CSF work-up was negative for HIV, WNV, EBV, CMV, Lyme, VZV, HSV, NMO antibody, ACE, paraneoplastic panel, cytology, Cryptococcus, and CSF bacterial and fungal cultures. CT chest and blind transbronchial needle lymph node biopsy were negative for sarcoidosis. Serum Coxsackie B3 (1:320) and B4 (> 1:640) antibody titers were significantly elevated. The patient was treated with intravenous methylprednisolone 1000 mg for 5 days along with plasmapheresis for 5 sessions. The patient had improvement in lower extremity strength during his hospital stay and was discharged to a rehab facility on a steroid taper. At one-month follow up, the patient had complete recovery of lower extremity strength as well as bowel and bladder function and was ambulatory. At 3-month follow-up, MRI and CSF studies were markedly improved. Transverse myelitis due to Coxsackie has been reported in serotypes B2, B5, A9, A10, however only three cases of B3 or B4 related transverse myelitis have been reported and this is the first case, to our knowledge, with both B3 and B4 related transverse myelitis. External ventricular drains (EVDs) are necessary for select patients admitted to the Neurointensive Care Unit (NICU). EVDs are critical to the management of diseases such as subarachnoid hemorrhage, traumatic brain injury, and acute hydrocephalus. We report a case of a woman with poor-grade aneurysmal subarachnoid hemorrhage who experienced inadvertent intraventricular non-iodinated contrast injection during vasospasm evaluation with CT angiography. We provide a review and analyses of adverse EVD related injections reported in the literature and summarize management recommendations. A PubMed search was performed for unintended EVD injections from 1927 to 2016. Unique cases were selected and classified by the type of inadvertent injection and location of the event. Acute management was categorized by the use of EVD manipulation, lumbar drain placement, and supportive medical therapies. Cases involving ventriculoperitoneal shunts, Ommaya reservoirs, or other intrathecal device systems were excluded. A total of seven unique cases were identified, four involving contrast administration and three involving medication administration. The sentinel event in five cases occurred outside of the NICU. Acute management with EVD manipulation was employed in six cases; three of these cases also used lumbar drains. Three cases necessitated intubation. Our patient received prophylactic dexamethasone and levetiracetam, underwent immediate EVD manipulation, and placement of a lumbar drain. Continuous EEG monitoring and daily head CTs were performed until intraventricular clearance was noted at 72 hours. Acute management of inadvertent intraventricular injections entails immediate EVD manipulation and possible lumbar drain placement to facilitate rapid CSF clearance of injected substances. Intubation may be required immediately depending on exposure. Inadequate distinction between compatible drain tubing and relative inexperience of providers managing the EVDs likely contributed to the errors in these cases. The increased incidence of these events outside of the NICU suggests additional safety measures may be warranted when patients with EVDs travel off-unit. An unusual presentation of isolated brain abscess in non-traumatic convexal subarachnoid hemorrhage: A case report To present a unique case of isolated brain abscess presenting as non-traumatic convexal subarachnoid hemorrhage (cSAH) six days before radiologic signs could be seen. To our knowledge only one other case of cSAH due to brain abscess has been reported thus far. A 42-year old man with no past medical history or prior trauma presented with acute onset of transient left hemianesthesia lasting ten minutes. Computed tomography (CT) of the head revealed cSAH. We were presented with a diagnostic dilemma when magnetic resonance imaging (MRI) of the brain, magnetic resonance angiography (MRA) of the head and neck, magnetic resonance venography (MRV) and conventional angiography failed to show the cause of cSAH. The patient was discharged in a stable condition, but returned six days later with worsening symptoms, including left hemiparesis. Repeat MRI with contrast revealed a cerebral abscess in the same location as the prior cSAH. cSAH without clear evidence of trauma due to abscess is a rare occurrence. We suggest in cases of cSAH where imaging techniques present no abnormalities, follow-up imaging within seven days should be considered. Jonathan Marehbian, Diane D. Chan, David D. Greer.Yale New Haven Hospital, Department of Neurology, New Haven, CT, USA. Spinally-mediated reflex movements can be present in brain dead patients. However, abnormal movements have long been a challenge in the clinical determination of brain death. In this report, we describe delayed plantar extension with noxious nail bed stimulation that has not been previously described in brain death. A 28-year-old male suffered severe anoxic brain injury following respiratory failure due to heroin overdose. His clinical exam and apnea testing were consistent with brain death with the exception of a reproducible delayed plantar extension with noxious nail bed stimulation. Ancillary testing with technetium 99m nuclear scan (SPECT) demonstrated no cerebral blood flow, confirming that the finding was spinally-mediated. Novel movements are important to document in order to aid in the timely determination of brain death, and to avoid unnecessary and potentially confounding ancillary testing. The mechanism underlying delayed plantar extension is likely spinally mediated. Vascular perforation during a neuroendovascular procedure is an unexpected and feared complication, which can lead to fatal outcomes. A prompt recognition and initiation of treatment are paramount. Endovascular strategies to address this complication have been widely described. However, the goals of therapy in the neurointensive care unit (NeuroICU) remain unclear. We report two cases in which endovascular strategies associated with aggressive intensive care resulted in a good clinical outcome at discharge. Case Report. Vessel perforation occurred in 2 patients during neuroendovascular interventions: a left-carotid stenting and an ICA-aneurysm embolization with balloon-assisted coiling. Once contrast extravasation was demonstrated, heparin was immediately reversed and endovascular strategies were performed to minimize the complications. SAH was present in the initial head CT in both cases. In the NeuroICU, targets of therapy were A) systolic blood pressure (SBP) <110 mmHg, B) mechanical ventilatory support, and C) seizures and vasospasm prophylaxis with phenytoin and nimodipine respectively. One of the patients developed mild hydrocephalus and left upper extremity weakness, with MRI showing tiny right hemispheric strokes, while the other one, did not show any focal deficit. Both patients were discharged home few days later with mRS of 2 and 0, respectively. Iatrogenic vascular perforation is an uncommon complication that occurs secondary to inadequate manipulation of the catheter, guide wire, devices, or forceful contrast injection. Clinical care strategies aim to prevent fatal outcomes. We recommend reversing heparin; provide an adequate ventilatory support, maintain a strict control over the SBP, and seizure and vasospasm prophylaxis. But, larger studies are required to determine the standard of care since endovascular therapy is rapidly becoming first-line of treatment for neurovascular conditions. Cerebral Hyperperfusion Syndrome (CHS) symptoms range from severe unilateral headache to seizures, focal symptoms and intracerebral hemorrhage, usually occurring follow carotid endarterectomy (CEA)/ carotid stenting (CAS). We describe a case of a patient who developed CHS after spontaneous recanalization of carotid intra-stent thrombosis. Case report. 68 years old African American male who had recently placed left cervical stent was transferred to Baptist Medical Center with new evidence of intra-stent thrombosis. He was started on heparin drip and SBP was augmented with levophed for SBP>160. Two days later, the patient developed excruciating headache followed 6 hours later by acute onset of right upper extremity and face twitching associated with severe hypertension. The BP was controlled, protamine was given to reverse heparin and Keppra was loaded to treating seizures. CTA at this time showed recanalization of carotid in-stent stenosis and CTP confirmed hyperperfusion of frontal and parietal lobe with neither evidence of new ischemic area nor bleeding. Fortunately, inspite a delay in diagnosis by 6 hours, patient did not suffer intracranial hemorrhage. Most patients who develop CHS will have complete recovery if it is discovered and treated early.Aggressive prophylactic blood pressure control is the main treatment. For those who are diagnosed late and those progressing to ICH, the prognosis can be devastating with mortality rates up to 50%. Considering the importance of blood pressure control in the cerebral hemodynamic, studies have been done, trying to find a better tool to predict the best BP target in order to prevent CHS. Near Infrared Spectroscopy (NIRS) is a non-invasive and reliable technique that monitor the cerebral hemodynamic. Had NIRS been deployed during anticoagulation, the diagnosis of cerebral hyperperfusion syndrome would have been made at the onset of headaches and would have avoided the development of seizures and potentially a life threatening hemorrhage. The Triphasic Response: Water imbalance after neurosurgery : A case reportBibhukalyani Das, Indranil I. Ghosh.Institute of Neurosciences Kolkata, Kolkata, India. Water balance disorders after neurosurgery are well recognized, but detailed reports of the triphasic response are scarce. We describe a 30-year-old woman, who developed the triphasic response with hyper and hyponatraemia after resection of craniopharyngeoma. A 30-year-old female (no previous medical history, no medication, normal electrolytes and endocrine parameters) with MRI showing s/o craniopharyngeoma underwent neurosurgery using a subfrontal approach to resect the craniopharyngeoma while leaving the pituitary stalk intact (according to the surgical notes). Within a few hours of an uneventful postoperative recovery she started having polyuria and hypoosmolar urine with hypernatremia. Being normoglycaemic and not on any diuretics a diagnosis of central diabetes insipidus was made treated with Desmopressin nasal spray and drinking water ad libitum. Improvement occurred over the next postoperative day when desmopressin was discontinued. On the third day she developed with hypoosmolar hyponatremia along with seizure treated with 3% saline and fuid restriction. Two days later polyuria returned and was ultimately discharged with desmopressin tablets and advice to take fluids ad libitum. NA This case illustrates the dramatic and sudden changes in water balance that may occur after neurosurgery. The pathophysiology of the triphasic response appears to be early hypothalamic dysfunction, subsequent release of vasopressin from the degenerating pituitary and, finally,depletion of vasopressin stores. It has been difficult to identify patients at risk, but predisposing factors appear to relate both to the disease (macroadenoma, microadenoma, craniopharyngioma) and to the surgery (degree of manipulation). Successful prevention probably involves a psychological switch by not waiting until frank dysnatraemia has developed, but to act as soon as urine output and tonicity change. This requires an index of suspicion for treating and consulting physicians and specific instructions to nursing staff, especially in nonintensive care settings. Refractory status epilepticus (RSE) has high mortality and is difficult to treat. When traditional therapies fail ketamine may be considered. Ketamine is associated with limited reports of adverse cardiac events during anesthesia, but not during treatment for RSE. We evaluated occurrences of cardiac arrhythmias associated with ketamine. Retrospective chart review of neurocritical care patients in a tertiary academic medical center who received ketamine infusion for RSE between October 2010 and April 2016. Ten patients were admitted to a neurologic intensive care unit and received ketamine infusion for RSE. Etiology of RSE included autoimmune/infectious process (8), ischemic stroke (1) and subarachnoid hemorrhage (1). Of the ten patients who received ketamine, three had documented cardiac events without prior cardiac history. One patient remained clinically stable and did not require intervention. Another patient required escalating doses of ketamine infusion (maximum 6mg/kg/hr) for RSE secondary to presumed leptomeningeal disease, had an asystolic event and expired. The third patient was on low dose ketamine (0.07mg/kg/hr) for RSE secondary to subarachnoid hemorrhage, and developed multiple arrhythmias including recurrent episodes of asystole. Once ketamine was discontinued the patient stabilized. Arrhythmias are not uncommon in critically ill patients, but this is the first report of cardiac arrest associated with the use of ketamine for RSE. Although sympathomimetic properties of ketamine may provide vasopressor sparing effects, which reduce the need for vasopressors to counteract the hypotension commonly seen with other anesthetics used in RSE, it may put patients at risk for cardiac arrhythmias. In addition, ketamine has direct negative ionotropic effects and may raise pulmonary artery pressures. Caution should be employed when ketamine is used in RSE in patients with other independent risk factors for cardiac events. A 36 year-old veterinary technician with headache and fever for 2 days presented with altered mental status and myoclonic jerking. Initial LP showed 21 white cells and elevated protein (71 mg/dL). Recurrent clinical seizures occurred for 6 days prior to transfer to our institution. His exam demonstrated diffuse hyperreflexia and coma; EEG demonstrated up to 2.5Hz frontally-predominant rhythmic delta but no unequivocal seizures. Extensive workup revealed no evidence of infectious, toxic, or immune-mediated encephalitis. MRI demonstrated bithalamic injury and a region of questionable periventricular nodular heterotopia in the right parietal region. Subsequently, he developed recurrent clinical and unequivocal electrographic seizures from the right parieto-occipital region. Five periods each lasting >48 hours of anesthesia-induced EEG burst suppression failed to stop seizures, qualifying him as super-refractory status epilepticus. After 6 weeks of failure to wean from anesthetia, invasive monitoring for seizure localization was carried out using strips and depth electrodes. Multiple seizures were recorded, localizing to the medial occipital lobe, which was subsequently resected along with the region of pathologyconfirmed heterotopia. Seizures gradually improved, requiring 7 anti-seizure drugs and a ketogenic diet. He regained consciousness with preserved higher cognitive functions (language, memory) and personality 6 months later as his antiseizure drugs were successfully decreased. His motor recovery was limited by critical illness myoneuropathy. New-onset focal super-refractory status epilepticus may respond to surgical resection in extreme cases. Close collaboration with a multidisciplinary team of epileptologists and neurosurgeons can lead to resolution of seizures and eventually recovery. In patients with leptomeningeal metastases (LM) and elevated intracranial pressure (ICP), transient neurologic events secondary to plateau waves -temporary elevations in ICP -may occur. There is a paucity of clinical reports correlating video-EEG with definite or presumed plateau waves in patients with LM. Case report and literature review. Pubmed was queried for 'leptomeningeal metastases and EEG', 'leptomeningeal metastases and plateau waves' and ' leptomeningeal metastases and intracranial pressure'. A 48-year-old woman was transferred to our hospital for further care of a pituitary microadenoma noted on outside imaging. She had a history of Hodgkin lymphoma treated with radio-chemotherapy. During her admission, she developed episodes of confusion and video-EEG monitoring was initiated. While there were no electrographic seizures, several events, between 5 to 10 minutes in duration, of delayed or absent verbal responses, eye rolling, staring and alternating gaze preferences to both sides were noted. All these coincided with abrupt onset of marked background slowing, evident through generalized 2-3 Hz delta, mixed with some theta frequencies. Review of outside and repeat imaging with MRI of the brain with contrast revealed brain and leptomeningeal metastases. Cerebrospinal fluid examination showed raised opening pressures and malignant cells, leading to a subsequent diagnosis of primary signet ring cell cancer. We found one additional report describing video-EEG correlates of presumed plateau waves in patients with LM. Our clinical description of transient neurologic events in this patient adds to the current literature of paroxysmal manifestations owing to raised ICP in patients with LM. Awareness of this clinical phenomenon may serve as a surrogate of raised ICP before clinical signs of the same develop in patients with LM. It may also help delineate the cause of raised ICP due to CNS metastases in a patient with a corresponding history of cancer. Background: Terson's Syndrome is the development of intraocular hemorrhage (IOH) in association with subarachnoid hemorrhage (SAH). We report a case of Terson's syndrome and review the literature. Case Report: A 48yr woman presented with severe neck pain and somnolence. CT imaging showed Fisher Grade 3 SAH with aneurysms in the right internal carotid and posterior communicating arteries. She developed low pressure hydrocephalus treated with ventriculostomy. On hospital day 12 she developed vision loss. Ophthalmologic examination demonstrated bilateral vitreous hemorrhages with near complete fundoscopic resolution by day 19. Over the next 4 months she underwent pars plana vitrectomy (PPV) of the left eye two times. Current vision OD 20/20, OS 20/50. Discussion: The incidence of Terson's syndrome among patients with SAH is documented to be between 8-40%. In prospective studies, IOH was found in up to 28% of patients with SAH compared to only 3% in retrospective studies suggesting under-reporting. This is likely because IOH is found more often in higher severity bleeds where patients cannot self-report visual loss due to decreased LOC. Patients with IOH are more likely to have worse neurologic outcomes and die more often than those with lower grade bleeds without the development of IOH. Along with fundoscopy, hand held ultrasound may be used for bedside diagnosis. ERM development is the most common intraocular complication from Terson's syndrome and occurs in 15-78% of patients. Complete or near complete return of visual acuity is less likely without surgical intervention. Visual acuity has been show to recover better and faster if PPV is performed within 90 days. There is no literature on incidence of visual loss after acute phase of SAH. Conclusions: Routine evaluation with imaging and fundoscopy may help in detecting IOH sooner in the clinical course and has the potential to decrease long-term morbidity. Iatrogenic underfeeding in critically ill patients is often unrecognized and underestimated. An international prospective study of 201 critical care units showed patients received only 61.2% and 57.6% of prescribed calorie and protein needs. The inability to initiate enteral nutrition within 24-48 hours of ICU admission or frequent interruptions of the enteral regimen lead to insufficient nutrient delivery and a compounding energy deficit. An increase in infectious complications is associated with negative energy balance in patients with Subarachnoid Hemorrhage (SAH). A quality initiative project was developed at Mayo Clinic Florida to measure time to reach enteral nutrition target and common interruptions of enteral nutrition. The target subjects were mechanically ventilated patients in medical and transplant ICU; however the quality measure has recently extended to the Neurocritical Care unit. Data collected included clinical diagnosis, SOFA and APACHE II score, Subjective Global Assessment (SGA) score, Nutric score, enteral tube type and regimen, and reason and duration for interruption of nutrition. A 76 year old subject with SAH and Posterior fossa AVM resection was deemed low nutrition risk with SGA score A and Nutric score 4. Enteral nutrition was initiated via nasoenteric tube within 25 hours of intubation. Target enteral goal rate was reached within 16 hours. The patient received 100% of calorie/protein needs 9 of the 12 days. The most common enteral interruption was for procedure; primarily head CT, for longest duration of 155 minutes. The interruption of enteral nutrition in Neurocritical care patients is likely unavoidable due to procedures. These disruptions, however, need not result in iatrogenic underfeeding. Neurocritical care units may utilize volume-based enteral protocols to allow nurses to compensate for lost nutrition with increased enteral rate. Neurogenic pulmonary edema is challenging to manage in the context of aneurysmal subarachnoid hemorrhage (aSAH) due competing priorities between organ systems. We present a case of refractory neurogenic pulmonary edema due to aSAH necessitating extracorporeal membrane oxygenation (ECMO). Case report. A 30 year-old female with a history of hypertension and diabetes presented neurologically intact with HH2F3 aSAH due to a left posterior communicating artery aneurysm. She underwent coil embolization on SAH day 2 and remained intubated after the procedure due to development of flash pulmonary edema. Transthoracic echo demonstrated normal left ventricular function. On SAH day 3 after unplanned extubation, she was temporized on noninvasive ventilation until reintubation the following day. She progressed to severe ARDS requiring high-dose sedatives and paralytics which obscured her neurologic exam. On SAH day 7, a 9 day course of intrathecal nicardipine was initiated for elevated left MCA transcranial doppler velocities. On SAH day 8, her respiratory status further declined and veno-venous ECMO was initiated as rescue therapy after head CT did not demonstrate new hemorrhage or infarct. Throughout the 7 day ECMO course, a low-dose heparin infusion was utilized along with vasopressors to optimize cerebral perfusion pressure (CPP). She was decannulated on SAH day 15. Surveillance CT head demonstrated left-sided ischemic infarcts in multiple vascular territories. On SAH day 42, she was discharged to an outside facility for ventilator weaning. Upon discharge, she was alert and followed commands with her left arm, however she was aphasic without movement of her other extremities. Veno-venous ECMO was performed in an aSAH patient after coil embolization. Although the patient did not develop intracranial hemorrhage, her course was complicated by severe vasospasm and delayed cerebral ischemia (DCI). While ECMO is a rescue therapy for severe hypoxemia, its effect on CPP remains uncertain and may potentiate DCI.