key: cord-0013140-xukvfszd
authors: nan
title: 18th Annual Meeting, Neurocritical Care Society September 22–25, 2020 Phoenix, Arizona
date: 2020-10-07
journal: Neurocrit Care
DOI: 10.1007/s12028-020-01091-2
sha: 93aca3c8150b26d5be42e5ddb5912cdb3647460b
doc_id: 13140
cord_uid: xukvfszd

nan

Left ventricular assistance devices (LVADs) have emerged as the standard of care for bridgingdestination therapy in heart failure refractory to guideline directed medical therapy. Despite an increasing incidence of cerebrovascular events in this population, the underlying pathophysiologic mechanisms remain not well elucidated as several aspects could play a relevant role. Differentiating between potential contributory mechanisms can provide useful insights on the management and prognosis of these patients. Two of the most commonly implemented stroke classification models (TOAST and ASCOD) do not include LVAD patients. Our classification system is based on a comprehensive understanding of the potential pathophysiological mechanisms operating in each patient, which allows new individualized treatments targeting the specific etiology.

We created an etiological classification based on the subtypes of mechanical circulatory assist device (MCAD) related strokes. We evaluated the interrater agreement between two independent researchers classifying strokes in LVAD patients based on the proposed MCAD classification in a rating scale of 1-5, being 1 highly-unlikely and 5 highly-likely.

We retrospectively reviewed 192 LVAD subjects identifying 31 cerebrovascular events from 2010 to 2020. Among those, 58.06% were ischemic strokes and 41.9% hemorrhagic strokes. The MCAD stroke classification divides LVAD related strokes in 4 categories: 1.Management related(M); 2.Clinically related(C); 3.Acquired related(A); 4.Device related(D). The interrater agreement was extremely high (0.963 ± 0.034). The rates agreed in the primary classification of 31 subjects determining a management related etiology in 67.74% of cases, whereas 16.12%, 16.12%, 12.90% for categories A, D and C respectively. Four subjects fitted in more than one category.

The complex nature of LVAD related strokes brings a challenge to clinicians regarding therapeutic measures. This is the first stroke classification system developed for VAD patients. Classifying subjects in these subgroups shed some light not only in the etiology but in the potential therapeutic pathways as well as prognostic factors.

Transcranial doppler (TCD) is a bedside modality to rapidly assess the presence of vasospasm or stenosis of cerebral arteries. However, manual TCD requires a trained sonographer which may not be available in all healthcare settings. The LUCID TCD system aims to expedite and enhance TCD blood flow measurements through semi-autonomous acquisition. This study aims to evaluate the safety and feasibility of the LUCID TCD system for detection of vasospasm.

Nine patients (average age = 66yrs) with subarachnoid hemorrhages or aneurysms underwent a total of 21 LUCID TCDs to assess the middle cerebral arteries (MCA) for vasospasm. Peak velocities of the MCA were measured and a velocity greater than 120 cm/s was considered indicative of vasospasm. To determine feasibility, maximal MCA velocities from LUCID TCD were compared to computer tomography angiography and the Cohen's Kappa value was calculated to gauge the agreement between the two modalities. Safety was assessed through quantifying the number of complications with central venous lines (CVL) or external ventricular drains (EVD) attributable to the LUCID device.

Previous studies (e.g. progesterone) suggest that TBI edema and lesion volume (LV) may differ by sex. Sur1 (ABCC8) inhibition via glibenclamide has shown promising results-a Phase-II study in contusion is ongoing and a Phase-III is planned. Successful translation requires appropriate patient selection. Hypothesis: Sex impacts effects of Abcc8 knock-out (KO) on LV and hemispheric tissue-loss (TL).

Methods 45 adult mice were divided into 6 groups (n=7-9/group): male Abcc8-KO, heterozygotes, and wild-types (WT), and female Abcc8-KO, heterozygotes, and WT. Mice underwent controlled cortical impact (left parietal cortex, 5m/s, 1.2mm depth, 50-60ms dwell time) before sacrifice (21d). TL and LV were quantified on frozen sections using MCID. LVs were recorded as absolute (LVabsolute) and percent of hemisphere (LV%).

Male-WT had higher LVabsolute vs male-heterozygotes and KOs (18.86±1.5 vs 14.93±1.97ml, p=0.04). There was a trend towards sex moderating effect of genotype on LVabsolute (p=0.11). Male-WT trended towards higher LV% (15.08±1.07%) vs female-WT (12.14±1.0%, p=0.055), and vs male-KO (12.30±1.07%, p=0.077). There was a similar trend towards sex-genotype interaction on LV% (p=0.096). LVabsolute and LV% were similar between all female genotypes. Male heterozygotes had lower TL vs WT (16.6±2.19 vs 21.06±1.55 ml, p=0.045). Controlling for genotype, females had lower TL vs males (17.48±1.3 vs 20.33±1.2, p=0.038), but there was a trend towards higher TL in female Abcc8-KOs vs WTs (16.84±0.89 vs 19.35±1.22, p=0.059) .

Effects of Sur1 (ABCC8) elimination may vary with sex. Female-WTs have lower LVs than males, possibly consistent with protective effects of sex-hormones e.g. progesterone. Abcc8-KO may decrease LV in males, but confer no additional advantage in females. Females have lower TL, however Abcc8-KO may decrease this benefit. Our exploratory findings suggest that Abcc8-KO could produce benefit in males, but may have no or even detrimental effects in females. Studies evaluating pharmacologic inhibition and functional impact are warranted.

Nimodipine (NMP) is the only FDA approved drug for aneurysmal subarachnoid hemorrhage (aSAH) patients proven to improve clinical outcomes. The drug's neuroprotective mechanism per FDA package insert is independent of angiographic vasospasm. Based on our preliminary work investigating pharmacogenomics (PGx) and CSF drug levels, we hypothesize there is a PGx-mediated NMP dose response, which may be influenced by other cytochrome P450 (CYP) mediated drug-drug interactions with other commonly used neuroICU drugs.

A single, retrospective neuroICU center PGx registry of aSAH cases for arterial hypotension and dosereduction (DR) compared to modified Rankin neurological outcomes by death, discharge and/or by 30 days. We risk-stratified all patients with admission WFNS and Fisher scores.

Over four years, we investigated n=111 aSAH patients CYP A4 and A5 PGx genotypes with the phenotypes of dose reduced NMP hypotension and dichotomized modified Rankin outcomes (good outcomes 0-3 vs poor outcomes 4-6). Hypotension and dose-reduction from standard 60mg to either 30mg or 15mg was associated with a statistically significant rate of worse outcomes (P < 0.001) but mirrored worse WFNS scores. Dose-related hypotension with standard 60mg was observed in up to 68% of all patients. No specific A4 and A5 genotype predicted hypotension. We observed more ethnic dosetolerance in African-Americans compared to Caucasians but was underpowered to detect differences amongst other ethnic groups (LatinX, Asians, Pacific Islanders).

Nimodipine dose-related hypotension and dose-reduction is present in more than 50% of aSAH cases, and worse in higher WFNS grades which may paradoxically underdose this neuroprotective drug. We report preliminary ethnic differences in Caucasians and African Americans in terms of dose-tolerance, but the study was underpowered to detect broader ethnic genotypic dose-tolerance. We plan to investigate and validate this data with a larger, more ethnically diverse, multicenter PGx registry in a future biomarker-driven clinical trial.

Perihematomal edema (PHE) growth is associated with worse clinical outcomes in patients with intracerebral hemorrhage (ICH). Exact definitions of perihematomal growth are not known. We aimed to determine the definition for PHE growth that best predicts poor outcome.

We performed an exploratory analysis of the Antihypertensive Treatment of Acute Cerebral Hemorrhage-2 (ATACH-2) trial. Patients underwent initial and 24 hour brain computed tomography (CT) scan. Modified Rankins score (mRS) was calculated at 30 and 90 days. PHE growth was calculated as the difference between PHE volumes (absolute) or absolute PHE volume divided by hematoma volume (relative). PHE growth rate was calculated as the absolute PHE growth divided by the time interval. We generated receiver operating characteristic (ROC) curves for relative PHE, absolute PHE, and PHE growth for the prediction of poor 90 day outcome defined as mRS 3-6 and determined optimal cutoff. As a comparator, we also generated ROC curves for hematoma volume expansion

The area under the ROC curve (AUC) for the prediction of poor 90 day outcome was 0.56 for absolute growth, 0.50 for relative growth and 0.56 for absolute growth rate. The best cutoff for prediction of poor mRS at 90 days was absolute growth was 6.6 mL (sensitivity 49%, specificity 64%). The AUC for the prediction of poor 90 day outcome with hematoma volume expansion was 0.58 for absolute growth, with a cut off for 6 ml at 52% sensitivity and 64% specificity.

PHE absolute growth and absolute growth rate modestly discriminated the risk of poor 30 day outcome. Absolute growth is a better predictor of poor 90 day outcome than relative growth and 6.6 ml is an optimal cutoff. A clear definition of PHE growth will assist in further exploration of this potentially modifiable risk factor for poor clinical outcome.

Identification of Large Vessel Occlusion (LVO) Strokes in the Emergency Department (ED) of Non-primary Stroke Centers is important for triaging patients to Comprehensive Stroke Centers for them to get Tissue plasminogen activator (tPA) and Endovascular Mechanical Thrombectomy which have become standard of care. If LVO is suspected, immediate imaging of the vessels of the head and neck using Computed Tomography Angiogram (CTA) should be done. The aim of our study was to increase the number of CTAs ordered by ED for quicker identification of LVO Strokes leading to transfer to facilities where patients can get appropriate care.

The intervention for this study was a didactic session held on 10/24/2019 for ED residents and staff on identification of LVO Strokes, intervention windows for tPA versus Mechanical Thrombectomy, and the role of CTAs. Pre-intervention data from 4/23/2019 till 10/23/2019 and Post-intervention data from 10/25/2019 until 4/25/2020 was collected from chart review of Code Stroke patients. Data included: time when Stroke CT Head and CTA Head and Neck were ordered. If time between orders was greater than 10 minutes, it was presumed that neurology service was consulted, who requested the imaging. The percentage of CTA studies ordered by ED versus Neurology pre-and post-intervention were calculated. Statistical analysis was done with chi-square test.

In Pre-intervention cohort (n=29): 44.8% (13/29) CTAs were ordered by ED while 55.2% (16/29) were ordered by Neurology. In the Post-intervention cohort (n=22): 77.3% (17/22) CTAs were ordered by ED while 22.7% (5/22) were ordered by Neurology. CTAs ordered by ED were observed to increase from 44.8% to 77.3%, a statistically significant increase, χ2 (1, N=51)=5.44, p= 0.02.

Intervention in the form of ED didactic session led to a statistically significant increase in percentage of CTAs ordered by ED. Results highlight the importance of resident educational interventions to improve patient care.

To describe a case of anti-GAD65 antibody syndrome manifesting as myoclonus status and encephalitis. GAD65 is usually related to autoimmune stiff person syndrome (SPS), a CNS disorder characterized by progressive rigidity of the truncal muscles, superimposed spasm and exquisite sensitivity to external stimuli. Although several subsets such as "stiff-limb syndrome", Jerking SPS or SPS with encephalomyelitis have been described, encephalitis with myoclonus status is rarely reported.

A 49-year-old Caucasian female with a previous medical history of mitochondrial complex III deficiency, hypothyroidism presented with generalized myoclonus for 1 day. A review of her home medication was unrevealing. On exam, she had mild rigidity and hyperreflexia. She got multiple doses of benzodiazepine and levetiracetam with transient improvement of her myoclonus and intubated. Admission labs were remarkable except for lactate 2.6 and CK 349. She was found in myoclonus status on vEEG, which was aborted by multiple AEDs including lacosamide, levetiracetam, phenytoin, midazolam, and propofol. MRI showed questionable right Mesial temporal lobe sclerosis and CSF had mild lymphocytosis with normal protein and glucose. She was on antimicrobial coverage until infectious workup negative. A CT chest abdomen pelvis showed no malignancy. Two rounds of high-dose steroids were given without responses for presumed Hashimoto encephalopathy with high titers of serum TPO and thyroglobulin antibody and suggesting features on ultrasound. Intravenous immunoglobulin was initiated based on positive CSF GAD 65 (titer 3.94) while other CSF antibodies were unrevealing including the TPO antibody. A PET CT and repeated MRI brain were unrevealing. She developed phenytoin related hepatitis and her myoclonus relapsed twice while we were switching her AEDs. She was gradually back to her baseline after five sections of PLEX and thyroidectomy.

Immunotherapy provides a good outcome for anti-GAD65 antibody-related refractory myoclonus status.

We are presenting a case for a patient who presented with altered mental status in the context of the presence of vesicular rash in the V1 cutaneous distribution of Trigeminal nerve without ocular involvement. This case highlights the importance of entertaining the diagnosis of meningitis and encephalitis in patients with altered mental status as this is a potentially reversible conditions in certain instances.

A 61 year old male with past medical history of astrocytoma status post resection on Avastin, seizure disorder on Keppra, CKD stage III, and nonalcoholic liver cirrhosis with ascites who presented to the hospital with a 10 days history of altered mental status. CT and MRI of the brain showed no acute findings. Continuous EEG showed no convincing evidence of seizure activity. Lumbar puncture showed a normal opening pressure, protein of 85 mg/L, glucose of 62 mg/dl, 2000 RBCs and 3 nucleated cells, lymphocytic pleocytosis and PCR positive for VZV. The patient was placed on airborne precautions for disseminated

It has been estimated that 0.01% to 0.25% of patients with varicella develop overt neurological complications such as cerebellar ataxia, encephalitis, transverse myelitis, aseptic meningitis, polyneuritis, cranial neuropathies, and Reye syndrome. In the overall clinical scenario a diagnosis of VZV encephalitis must be entertained, especially in the setting of a positive VZV PCR result. This case demonstrates the importance of lumbar puncture in evaluating altered mental status when there is no obvious cause of encephalopathy. There is a treatment for VZV encephalitis and so entertaining and making the diagnosis is critical

Early diagnostic CSF analysis and PCR should help to confirm the diagnosis; therefore, we are emphasizing the importance of having an early LP in patients with risks factors for disseminated Varicella encephalitis and absence of skin lesions should not withhold the treatment.

Moyamoya Disease (MMD) & Charcot Marie Tooth Disease (CMTD) are both rare neurological conditions. MMD is a unique cerebrovascular disease with progressive stenosis of large intracranial arterties and a hazy network of basal collaterals called moya vessels. CMT hereditary neuropathy refers to a group of disorders characterized by a chronic motor and sensory polyneuropathy, also known as Hereditary motor and sensory neuropathy. We report a case of an adult with a known diagnosis of CMTD-Type 2N, who had an acute bilateral MCA distribution ischemic strokes and was confirmed to have Moya disease on a cerebral Angiogram.

A thorough chart review was performed for the patient, including review of images of cta and cerebral angiogram with the treating Endovascular Neuroradiologist in regards to diagnosis of MMD.

Our patient was diagnosed with CMTD-2N variant, confirmed with genetic testing and EMG/NCS. She also had WPW (Wolf-White-Parkinson) disease. Cardiac abnormalities are rare but have been found to be associated with CMTD. Patient had presented with acute bilateral arm weakness with slurred speech and was found to have bilateral MCA watershed infarcts. CTA studies concerning for severe proximal ICA/MCA stenosis without any evidence of extracranial Carotid disease or Cardiac abnormalities other than those from WPW. Her Diagnostic Angiogram confirmed MMD and she safely underwent EDAS (Encephalo-duro-arterio-synangiosis) procedure. Based on ODC (Orphan Disease Center) report, CMTD-2N is associated with AARS gene mutation. MMD on the other hand is associated with ACTA2 and RNF213 genes. These are indirectly related to each other via the mutations in genes/proteins, however no direct link has been established yet.

We report first case of two rare conditions MMD and CMTD-2N in the clinical setting of acute bilateral strokes that was treated successfully with a surgical EDAS procedure. This association warrants further genetic, lab research to establish link between these conditions Introduction Fiberoptic bronchoscopy (FB) plays an important role in the diagnosis of nosocomial pneumonia and in guiding percutaneous tracheostomy (PT) in critically ill patients. In patients with brain injuries, FB can lead to intracranial hypertension. Sustained increases in intracranial pressure (ICP) can lead to poor outcome in these patients.

We report a case in whom transcutaneous end tidal CO2 monitor (TCOM) was used during FB for PT and correlated with ICP levels obtained from external ventricular drain (EVD).

Patient is a 58 year-old man admitted with aneurysmal subarachnoid hemorrhage who underwent coiling of anterior communicating artery aneurysm. Patient developed hydrocephalus for which he underwent EVD placement. Patient underwent percutaneous tracheostomy due to difficulty weaning from mechanical ventilation. During FB for PT, TCOM was used to correlate CO2 levels with ICP measurements. The TC-CO2 (mmHg) and ICP (mmHg) were recorded every 60 seconds and were correlated as follows: First FB insertion : ICP: 26:38.1; 30:38.3; 37:38.9; 30:39.6; 25:39.6; 20:38.6) , and second FB insertion: ICP: 30:38; 35:39.6; 44:41; 32:43.3; 18:40; 14:39) , with Pearson correlation coefficient of 0.93 (p = 0.007). TC-CO2 levels correlated strongly with ICP levels during FB with about 1-minute lag (CO2 lagged behind ICP). Even though ICP rises in response to hypercapnia (PaCO2 in the blood), this lag of TC-CO2 behind ICP is expected and explained by the lag of TC-CO2 behind PaCO2 which is a well-documented and an unavoidable consequence of time for carbon dioxide diffusion through skin.

This report strongly suggests that hypercapnea could be one of the mechanisms behind FB causing elevations in ICP. TCOM may be used to control the time bronchoscope is inserted in the airway during FB of patients with brain injury to minimize dangerous elevation in ICP, especially in patients who may not have an ICP monitor.

Semaglutide is a glucagon-like peptide-1 receptor agonist recently used off-label for weight loss in patients without diabetes. This is a case report of a patient who received off-label semaglutide for weight loss who developed a life-threatening subdural hematoma secondary to coagulopathy as a consequence of achieving appetite suppression.

Data collection included medical chart review and direct observation of the patient, followed by a literature review encompassing relevant topics.

A 45 year old female on chronic anticoagulation with warfarin who was prescribed semaglutide for weight loss and presented with a right convexity subdural hematoma in the setting of an INR greater than 12. Initial laboratory evaluation significant for severe metabolic acidosis. Coagulopathy was corrected with 4 factor prothrombin complex concentrate and vitamin K and the consulting neurosurgeon evacuated the subdural hematoma. Metabolic acidosis was attributed to a starvation ketosis in the setting of appetite suppression from semaglutide, compounded by chronically reduced caloric intake and appetite suppression. The resultant nutritional deficiency of vitamin K in combination with the chronic use of a vitamin K antagonist led to supratherapeutic INR levels, coagulopathy, and the subdural hematoma.

This case illustrates an interesting cause of subdural hematoma. With the significant public health crisis of obesity, many practitioners are prescribing medications with off-label uses. Providers should exercise caution when applying trial-data clinically in an off-label manner. Patients often have more complex comorbidities than those who are enrolled in industry-funded trials and further research may be necessary to fully evaluate medication safety in these populations.

Nonketotic hyperglycinemia (NKH) is an autosomal recessive disorder of glycine metabolism due to deficient activity of the glycine cleavage system resulting in increased levels of glycine in multiple tissues, including the brain. The vast majority of reported cases present in the first year of life, and are typically associated with severe epilepsy and grim neurodevelopmental outcomes.

Serial plasma amino acid, urine amino acid, and urine organic acid analyses, as well as exome analysis.

A 38 year old female with learning disability and depression initially presented at age 28 with involuntary movements and progressive encephalopathy. She was treated for suspected Serotonin Syndrome with symptom resolution. At age 38, she presented with facial twitching, polyminimyoclonus, chorea and progressive severe encephalopathy after developing pneumonia and respiratory failure; the encephalopathy and involuntary movements were refractory to intensive medical therapies. Blood chemistries, routine CSF tests, infectious serologies and cultures, EEG, brain MRI and whole body PET were unremarkable.. Family history was significant for two sisters with similar intermittent, undiagnosed neurologic condition. The patient's initial plasma amino acid analyses showed moderately high levels of glycine (~800 uM.) Treatment for NKH was initiated, including a low protein diet, sodium benzoate, dextromethorphan and pharmacologic doses of multiple vitamins. Exome analysis showed biallelic variants of the GLDC gene in the patient and her similarly affected sisters, confirming the diagnosis of NKH. Subsequent to NKH-related treatment she was no longer ventilator dependent, her mental status normalized, and the involuntary movements improved.

This case demonstrates the phenotypic spectrum of NKH presentations. It also illustrates the importance of family history taking and judicious metabolic and genetic testing, even in an intensive care context, for the recognition of medically treatable inborn errors of metabolism that can present in the adult years and the clinical ramifications of the genetic diagnosis for other family members.

Hematoma expansion post tissue plasminogen activator (tPA) induced symptomatic intracranial hemorrhage (sICH) can have deleterious outcomes. Limited data exists on management of sICH but guideline recommendations include cryoprecipitate in patients with low fibrinogen or antifibrinolytic therapy such as tranexamic acid (TXA) or aminocaproic acid. A fibrinogen level of <200 mg/dL has been established as a risk factor for hematoma expansion and can help guide cryoprecipitate administration but there is a paucity of data on selecting patients for agents like TXA. Viscoelastography testing, such as thromboelastography (TEG), may be beneficial at identifying patients that could benefit from TXA by evaluating lysis 30 percentage (Ly30%). Ly30% is a measure of fibrinolysis over time and since tPA has shown prolonged coagulopathy TEG may be able guide resuscitation.

Here we present two cases in which patients received tPA for acute ischemic stroke and unfortunately developed sICH.

TEG Ly30% were evaluated as a way to identify any ongoing hyperfibrinolysis and evaluated for hematoma expansion. Utilizing TEG we were able to create a stewardship approach to the neuroresuscitation of this cohort. The two patients did not have increases in hematoma size as evaluated on repeat computed tomography scans of the brain.

TEG and Ly30% may be useful tools at guiding resuscitation of patients who develop sICH and prevent erroneous administration of blood products or agents such as TXA.

Intracranial epidermoid cyst is rare. Microscopically, it consists of a capsule layer of striated squamous epithelium and is filled with keratin and cholesterol. Delayed intracranial vasospasm and ischemia after epidermoid cyst removal is rarer and reportedly self-limiting. Here we present a case with severe delayed intracerebral vasospasm after epidermoid cyst removal.

A retrospective chart review was performed on the patient's medical records. We present a case of severe delayed intracranial vasospasm after epidermoid cyst removal.

A 59-year-old woman with worsening right-sided trigeminal neuralgia was diagnosed with right cerebellopontine angle epidermoid cyst. Her surgery was uneventful, and she was discharged home soon after. On postoperative day (POD) 12, she presented with progressive global aphasia and confusion. CT head showed a small new left frontal convexity subarachnoid hemorrhage. MRI showed diffusion restriction within left fronto-temporo-parietal and cerebellar regions. MRA showed diffuse intracranial vasospasm most severe in left middle cerebral artery (MCA) territory. Patient was admitted to neuroscience ICU and treated with blood pressure augmentation and dexamethasone for presumed chemical meningitis. She continued to worsen clinically, so a catheter angiogram was performed. The angiogram revealed severe vasospasm within both MCAs and right PICA distributions. Intra-arterial verapamil was injected into left internal carotid artery with moderate radiographical improvement. On POD 14, she became mute and MRI showed evolution of old infarcts and a new right PICA infarct with brainstem compression. Patient was taken for decompression and CSF diversion. Her vasospasm persisted radiographically on POD 19, although her exam remained stable clinically. Ultimately, she was discharged to skilled nursing facility.

Vasospasm after epidermoid cyst removal is rare but could be life-threatening. Multiple hypotheses were raised to explain this phenomenon: 1. release of content from the resected tumor; 2. accumulation of blood products in the basal cisterns; 3. manipulation of the blood vessel; 4. autoimmune/paraneoplastic response.

The presence of cortical susceptibility-weighted imaging (SWI) lesions on MRI brain can be a sign of cerebral amyloid angiopathy (CAA), a disease typically affecting the elderly and conferring a higher risk of intracerebral hemorrhage (ICH). SWI lesions may also be present in patients with poorly controlled hypertension or patients with mechanical valves (MV). In patients with MV, the SWI lesions are typically located in frontoparietal lobes and associated with amount of time spent on bypass. We report a case of a young man with history of MV and cortical SWI lesions who developed recurrent ICH.

A 44-year-old Asian man, on Eliquis for MV replacement, presented with severe headache and left sided weakness. He was found to have a right frontal ICH. Past medical history was significant for left frontal ICH two months earlier while on Coumadin with INR 2.49 and a right frontoparietal ICH three months earlier with INR 5.7. Blood pressure was normal during each admission. MRI showed scattered cortical punctate SWI lesions. Five years earlier the patient was diagnosed with rheumatic heart disease and had metallic mitral and aortic valve replacements. An MRI performed post-operatively for an episode of PEA arrest had shown similar SWI lesions. Genetic testing revealed Apo-E genotype of E3/E3, not suggestive of increased risk of CAA. Family history was negative for history of strokes, dementia, or hemorrhage in his 10 siblings and parents. Social history was pertinent for 10 pack-years smoking. He worked as a manicurist before he had his first stroke.

Punctate SWI lesions associated with valve replacement surgery may be associated with an increased risk for recurrent ICH while on anticoagulation therapy like those in CAA.

Malignant reversible cerebral vasoconstriction syndrome (RCVS) is a rare and difficult condition to manage.

Case series. Patients were treated at an academic medical center in 2019.

We report on two young females (ages 33 and 43) presenting with acute severe thunderclap headaches and ultimately diagnosed with pharmacologically-refractory RCVS. Symptoms of both patients were refractory to standard-of-care intravenous nimodipine. In one patient, intravenous nimodipine therapy was initiated day 3, followed by repeated intra-arterial verapamil, and later multiple angioplasties with stenting. The other patient presented from an outside hospital after receiving intravenous nimodipine and underwent twelve angiograms with intra-arterial verapamil. Further neurological decline with radiographic confirmation of ongoing vasospasm in setting of increasing ischemic burden, resulted in death in one patient and a vegetative state in the other.

There is a lack of literature on both pharmacologically-refractory and stenting-refractory RCVS. The role of multiple angiograms, as well as other factors that may have contributed to a malignant course will be discussed. While a combined therapeutic approach involving intra-arterial vasodilators and angioplasty plus stenting proved ineffective in these two patients, there may be a role for developing treatment protocols involving prompt angioplasty with stenting, which may help mitigate unnecessary procedural risk and complications associated with repeated courses of intra-arterial therapy alone.

Left Ventricular thrombi with hypermobile segments seen on echo have been demonstrated to associate with high risk of arterial embolization with incidence rates of 22-100%. Large strokes with petechial hemorrhages in the setting of a hypermobile LV thrombus present a challenge in terms of timing of anticoagulation. We present a similar case in which a large hypermobile LV thrombus disappeared within 48 hours of initiation of anticoagulation without any trace of systemic or cerebral injury.

68-year-old white female with history of HTN, DM2, CKD3, PVD, TIAs presented to ED with AMS. Patient did not receive iv tpa for unknown last known well and she was intubated for gcs 7. Head CT showed a large left PCA stroke with scattered petechial hemorrhage. CTA showed occlusion of left ICA with involvement of pca. No basilar thrombus. A stat echo showed a large hypermobile LV thrombus. Despite the evolution of hemorrhage on follow-up CT, patient was urgently anticoagulated with heparin. Repeat echo at 48 hours demonstrated absence of thrombus. Patient did not develop renal failure or any evidence of Pulmonary embolic, splenic infarcts, hepatitis or mesenteric or limb ischemia during her hospital course.

The role of heparin in the setting of petechial hemorrhage and stroke has not been studied well especially in terms of its timing of initation. A case series of 23 patients with LV thrombi had a decrease in its size within 2 weeks of intiation of low dose heparin or warfarin. Our case is the first to suggest complete resolution or disappearance of LV thrombus within 2 days of initiation of heparin. The possibility of heparin breaking the LV thrombus into microthrombi with systemic arterial embolization remains a possibility.

It is safe to anticoagulate patient even in presence of petechial hemorrhages for acute ischemic stroke resulting from an LV thrombus.

Myoclonic Jerks are sudden, brief, shock-like "jerky" movements due to muscular contractions (active) or sudden lapse of muscle contraction (passive). Etiologies include various medications and physiologic conditions.

Arterial blood gas was used to monitor and adjust AVAPS setting.

A 56-year-old male with history of COPD, bipolar disorder, and drug abuse presented with multiple mechanical falls. Patient was recently started on olanzapine for bipolar disorder. Since then he began experiencing occasional jerking movements. Patient stated the jerking movements caused him to fall, hit his head and bite his tongue prior to admission. Patient was in hypercapnic respiratory failure and was started on AVAPS and admitted to ICU. The patient continued to have recurrent myoclonic jerks with myoclonic status. Patient became impulsive, climbed out of bed, and stumbled his way to urinate when he had severe jerking movements, and fell hitting his head. Patient was returned to bed while continuing to have myoclonic jerks following which he lost all strength in his extremities. Patient was awake and belligerent throughout. He was placed back on AVAPS. Olanzapine was tapered with concerns for drug-induced myoclonus. Patient was positive for cocaine and opiates. Within two days, the patient's hypercapnia improved with AVAPS and his jerking movements gradually resolved.

Myoclonic jerks can be caused by commonly prescribed medications. In this case the patient was abusing opiates and cocaine, on a common anti-psychotic, and was hypercapnic all of which may have exacerbated the patient's condition and lowered his threshold for myoclonic jerks. Along with the aforementioned, antibiotics including beta lactams, quinolones, sulfonamides, aminoglycosides as well as antidepressants, including Selective Serotonin Reuptake Inhibitors, Tricyclic Antidepressants and lithium have shown to cause focal or generalized myoclonus. Myoclonic jerks should be a consideration in any patient with multiple comorbidities and polypharmacy, since this is eminently treatable.

Coronavirus disease 2019 (COVID -19) is a global pandemic that causes flu-like symptoms. There is a growing body of evidence suggesting that both the central and peripheral nervous systems can be affected by SARS-CoV-2, including stroke. We present three cases of arterial ischemic strokes and one venous infarction from a cerebral venous sinus thrombosis in the setting of COVID-19 infection who otherwise had low risk factors for stroke. The mean age was 55 (48-70) years. All arterial strokes presented with large vessel occlusions and had mechanical thrombectomy performed. Two cases presented with stroke despite being on full anticoagulation.

We retrospectively reviewed patients presenting to a large tertiary care academic US hospital with stroke and who tested positive for COVID-19. Medical records were reviewed for demographics, imaging results and lab findings.

Case 1: A 51-year-old male with history of HTN, CAD, HLD was admitted to an outside hospital (OSH) with progressive shortness of breath and cough for four days. The patient was on 6 liters nasal canula oxygen saturating 92%. In accordance to the OSH COVID-19 treatment policy, the patient was started on therapeutic dose Lovenox (1 mg/kg) upon admission. On hospital day 2, he was found to hemiplegic on the left side with an NIHSS of 20. The patient did not receive IV tPA given he was on therapeutic lovenox. CTA head and neck demonstrated a tandem occlusion: acute thrombus in the right internal carotid artery (ICA) from its origin and an M1 occlusion. He was transferred to our hospital for endovascular intervention. Shortly after transfer, the patient developed worsening hypoxia and required mechanical intubation while in the angiography suite. He underwent mechanical thrombectomy (TICI 0 to 2B) with five stent placements to the right ICA. He was loaded with aspirin and clopidogrel and therapeutic lovenox was discontinued. Post stroke day 1, a repeat CT head in the neurocritical care unit (NICU) showed a large R MCA territory stroke. Laboratory testing was significant for the presence of anticardiolipin IgA antibodies, anti-B2-glycoprotein IgA and IgG antibodies. Unfortunately, the patient had progressive hypotension requiring multiple vasopressors and worsening hypoxia. The patient's family ultimately decided to withdraw life sustaining treatment and the patient died on hospital day four.

Case 2: A 70-year-old female with no significant past medial history was admitted to an OSH for shortness of breath, fever, hypoxia and new onset atrial fibrillation (A-Fib). She was confirmed to be COVID-19 positive and started on hydroxychloroquine and azithromycin. Her hypoxia was treated with non-rebreather mask. Per the OSH policy she was initiated on therapeutic Lovenox (1 mg/kg) for the new onset A-Fib. On hospital day 3, she was found to be altered with acute left hemiparesis and left facial droop with an NIHSS of 28. The patient did not receive IV tPA given she was on therapeutic lovenox. CTA head and neck demonstrated a acute right M2 occlusion. She was transferred to our hospital and underwent a successful thrombectomy (TICI 0 to 2A). While in the interventional suite, she became increasingly hypoxic and required intubation and mechanical ventilation. Patient was treated with prone positioning for her ARDS. CT head was obtained post stroke day 1 demonstrating a hemispheric right MCA stroke with mass effect and right to left herniation. Despite a decompressive hemicraniectomy and hypertonic therapy, the patient's neurological exam continued to decline and the family ultimately decided to withdraw life sustaining treatment.

Case 3: A 54-year-old male with history of HTN was admitted to an OSH with shortness of breath and cough 10 days after testing positive for COVID-19 infection. Over the next 24 hours, he complained of worst headache of his life and had progressively worsening mental status into coma. MRI brain showed diffuse T2 hyperintensities in the bilateral thalami and basal ganglia, 4 mm right to left shift, cerebral edema, and hydrocephalus with transependymal flow. MRI brain and CT venogram demonstrated filling defects in the vein of Galen, straight sinus, bilateral internal cerebral veins and right basal vein of Rosenthal consistent with thrombosis. Patient was transferred to our facility and immediately had an external ventricular drain (EVD) placed along with hypertonic therapy. After which, the patient was started immediately on a heparin drip for cerebral venous sinus thrombosis. Patient's pertinent labs show elevated D-dimer and other inflammatory markers. CSF analysis showed RBC 401, WBC 10, glucose 38, and protein 1,104 and his CSF SARS-CoV-2 RNA PCR was negative twice. Despite aggressive EVD drainage and medical management of intracranial pressures (ICP), the patient continued to have refractory ICPs. There was no evidence for meningitis or encephalitis. The patient's neurological exam continued to decline and the family ultimately decided to withdraw life sustaining treatment.

Case 4: A 48-year-old male with a history of HLD admitted to an OSH with an acute right sided hemiplegia after being found on the floor by family. On exam, the patient had global aphasia and right sided weakness with an NIHSS of 31. HCT was unremarkable but his CTA demonstrated an acute left middle cerebral artery (MCA) occlusion. He received IV tPA and was transferred for endovascular intervention. He underwent a successful left M1 thrombectomy (TICI 0 to 3). In the NICU, the patient was positive for COVID-19 despite not having any recent symptoms. MRI brain showed a large acute infarct in the left MCA territory, with mass effect and mild petechial hemorrhage. Aspirin and statin therapy have been initiated for secondary stroke prophylaxis along with subcutaneous heparin for DVT prophylaxis. His labs are pertinent for significantly elevated D-dimer. His NIHSS improved to 10 and currently he is in the stroke unit awaiting discharge to a rehab facility.

Neurologic manifestations are more common than expected in COVID-19 patients. In a recent review in China, 36.4% of COVID-19 patients had neurologic manifestations, ranging from skeletal muscles injury to acute cerebrovascular disease. The prevalence of cerebrovascular disease was low in this review with only 1-5% of patients developing symptoms of cerebrovascular disease. A multi-center prospective study in intensive care units (ICUs) in France found that close to 3% of COVID-19 patients develop ischemic strokes. Patients with severe symptoms are more likely than patients with mild to moderate symptoms to have neurological manifestations. In our case series, we present four cases of cerebrovascular disease who had concurrent COVID-19 infection. Our patients displayed evidence of hypercoagulability and significantly increased inflammatory markers. All four cases were positive for anticardiolipin antibodies. The first patient was positive for anti-B2-glycoprotein I IgA and IgG antibodies. A recent publication showed an association with the development of ischemic infarcts and the presence of these antibodies.All patients had elevated D-Dimers and other inflammatory markers. The mechanism on how the SARS-CoV-2 virus causes ischemic strokes is unclear. Several theories have been proposed. The elevated D-dimer in COVID-19 patients can be associated with the development of ischemic stroke. Elevated D-dimer increases blood coagulation, thrombin formation, and intravascular fibrin. In addition, elevated D-dimer concentrations have been reported in cerebral venous sinus thrombosis, acute pulmonary embolism, spontaneous intracerebral hemorrhage.The increased hypercoagulability can directly lead to clot formation in the large intracranial vessels and increased risk for ischemic stroke. In addition, the SARS-CoV-2 virus has been associated with a cytokine surge that is linked with the development of acute cerebrobasilar disease. In our patients, the LDL and HBA1C were at normal levels. Their bedside echocardiograms were unrevealing. It is possible that the hypercoagulability and inflammatory surge that accompanies COVID-19 infection can lead to ischemic infarcts through clot formation in large intracranial vessels. Moreover, the accompanying hypoxemia can lead to ischemic stroke by increasing intracellular acidosis and free radicals. This process can eventually lead to cell damage and apoptosis. Damage to the central nervous system can also come from direct invasion of the virus to brain cells. Brain autopsy in deceased COVID-19 patients detected the nucleic acid of the virus in their cerebrospinal fluid and brain tissue. Perhaps the invasion is a result of ACE2 receptors in cerebral blood vessels similar to the invasion that occurs in lung tissue, but further research is required to validate this.Another animal model showed that the olfactory nerve could be the point of entry of the virus to the brain.

According to the WHO, nearly 228 million cases of malaria occur worldwide including 405,000 deaths with adults accounting for 133,000 (2018). The CDC estimates 2,000 cases occur annually in the US. Due to this low incidence, close monitoring is imperative to avoid delayed recognition of progression to cerebral infection leading to additional complications including death.

Data collected via chart review and direct patient care.

A 56-year-old female with history of renal cell carcinoma post partial nephrectomy presented with fever, nausea, emesis, abdominal pain, and malaise for 2 days with subsequent development of confusion and difficulty ambulating. The patient endorsed travel from an endemic malarial region. Initial vitals were unremarkable with neurologic exam notable for mild expressive aphasia and disorientation. Labs demonstrated hyponatremia and thrombocytopenia. Malarial smear demonstrated plasmodium falciparum ring forms with 0.23% parasitemia. Presenting symptoms resolved post volume expansion and acetaminophen administration which preceded initiation of anti-malarial therapy with artemetherlumefantrine. Within 24 hours of admission, the patient developed global aphasia, dysarthria, inattention, and the inability to follow commands concerning for progression to cerebral malaria which prompted immediate transfer to ICU. Further workup revealed an unremarkable brain CT, continuous EEG with mild diffuse slowing, worsened thrombocytopenia, new transaminitis, and elevated LDH. Subsequent initiation of IV artesunate followed rapid resolution of encephalopathy within hours. The patient continued to have an expected mild cognitive dysfunction with no additional neurological deficits noted for the duration of hospitalization. Over several days, abnormal EEG tracings and thrombocytopenia resolved, however, the transaminitis persisted and mild anemia developed. Overall, the patient remained clinically stable and was discharged after six days with planned follow up.

Due to the rarity of cerebral malaria in the US, only with rapid recognition and close monitoring for deterioration can significantly improved outcomes and avoidance of additional complications be achieved.

Although patients with COVID-19 initially present with pulmonary pathology, they are at risk for neurologic complications. We present three cases of patients with COVID-19 who developed catastrophic ICH and cerebral edema.

Case series and review of the literature.

Three men ages 62-74 were admitted to our hospital in March 2020 for respiratory distress secondary to COVID-19. All patients required intubation for acute hypoxic respiratory failure. Prior to intubation, the patients were neurologically intact. Within the first 2 weeks of hospitalization, all three patients were empirically started on therapeutic heparin for elevated D-dimer. Sedation was discontinued after a mean of 10 days, at which time neurocritical care was consulted due to persistent coma. Exam of all three patients demonstrated coma with few or absent brainstem reflexes. All three NCHCT revealed multifocal ICH, global anoxic injury with cerebral edema, and downward herniation. There was no significant blood pressure fluctuation, coagulopathy, or thrombocytopenia to explain the imaging findings. Two patients died after WLST and the third had a cardiopulmonary arrest.

Catastrophic ICH may develop in critically ill patients with COVID-19 who are on therapeutic anticoagulation. It is well-known that anticoagulation can increase risk of bleeding and COVID-19 can cause ischemic strokes, but the imaging findings suggest that these injuries were not simply the result of anticoagulation-induced hemorrhage or hemorrhagic transformation of multifocal ischemic strokes. This hypothesis is supported by speculation in the literature that ischemic brain injury can be worsened by the cytokine cascade triggered by COVID-19. The use of empiric anticoagulation in patients with severe COVID-19 is controversial as there is no data to clarify whether the potential benefits of doing so outweigh the risks. As such, while these are only three cases, our findings should be considered when deciding whether or not to empirically start anticoagulation in a patient with COVID-19.

Post-stroke epilepsy (PSE) is divided into two categories. Early seizures (ES) typically occur within one week after stroke onset and are also termed 'acute symptomatic seizures', whereas late seizures (LS) have a peak within 6-12 months with a higher recurrence rate. PSE is about 10% as a stroke complication, but its rate is as high as 30 to 40% in the cause of elderly-onset epilepsy. We investigated the clinical characteristics of patients with status epilepticus after stroke who were transferred to our critical care center.

Forty-seven patients with status epilepticus who had been admitted to our critical care center were included. We retrospectively investigated the patient background (age, stroke type and risk factors), epileptic seizure type, and electroencephalogram (EEG) findings based on the information in the medical records.

Twenty patients (42.6%) had ES and 27 patients (57.4%) had LS. Median age was 71 years (range, 64.5-75 years), and stroke type was classified as follows: cerebral thrombosis, 21.3% (n=10); cerebral embolism, 27.7% (n=13), cerebral venous thrombosis, 4.3%, (n=2); intracerebral hemorrhage, 27.7% (n=13) and; subarachnoid hemorrhage, 19.1% (n=9). Hypertension was the most common risk factor for stroke in 33 patients (70.2%). Brain imaging revealed cortical lesions in 39 patients (78.0%). The seizure types consisted of 25 patients (53.2%) with a focal origin, 15 patients (31.9%) with a generalized origin, and 7 patients (14.9%) with unknown origin. In LS, patients with a focal origin were common (63.0%, n=17). EEG findings included rhythmic delta activity in 15 patients (31.9%), periodic discharges in 6 patients (12.8%), and spike-and-wave or sharp-and-wave in 7 patients (14.9%).

Based on our observation, PSE were characterized by elderly, comorbid hypertension, and cortical lesions.

Chimeric antigen receptor T (CAR-T) cell therapy is highly effective anti-neoplastic therapy; however, cytokine release syndrome (CRS) and neurotoxicity are observed in up to 77% of patients. Stroke and intracranial hemorrhages are only observed in 1-2% and seizures in 1%-8% of cases.

We present a 30 year old male with primary mediastinal large B-cell lymphoma who received anti-CD19 CAR-T cell therapy. On day six post CAR-T infusion, he developed grade 2 CRS and was treated with Tocilizumab. His mental status rapidly deteriorated and he developed myoclonic jerks in upper extremities and flaccid paralysis in lower extremities. Grade 3 neurotoxicity failed to respond to Dexamethasone. EEG demonstrated 2Hz generalized periodic discharges (GPDs), however concurrent brain PET showed global hypometabolism, suggestive of non-epileptic etiology of these findings. CSF was not inflammatory but did demonstrate elevated IL-6 levels. The patient received Siltuximab which led to a significant decrease in CSF IL-6 levels and coincided with rapid clinical improvement. Brain MRI demonstrated non-enhancing T2 changes in bilateral external capsule and thalami without diffusion restriction. MRI of the spine demonstrated longitudinal T2 changes involving predominantly grey matter across most of the spinal cord. At 6 month follow up the patient had a normal cognitive exam but remained with flaccid paraplegia in lower extremities and reported a T10 sensory level.

Neurotoxicity follows CRS and is commonly observed in patients after CART therapy. To our knowledge, this is the first reported case of neurotoxicity directly involving the spinal cord. Additionally, it has been postulated that exposure to tocilizumab may exacerbate neurotoxicity by IL6 receptor mediated elevation of CNS IL-6 level which was demonstrated in this case. Siltuximab (direct IL6 inhibitor) was temporally associated with a decrease in CSF IL-6 and rapid clinical improvement.

To increase awareness of contrast induced neurotoxicity as an uncommon but important neurologic complication of procedures involving the use of intracarotid injection of iodinated contrast media.

72-year-old, right-handed woman with diabetes, hypertension, right cavernous internal carotid artery aneurysm with prior pipeline stenting developed decreased alertness, dysarthria, right gaze preference, left hemiparesis, left hemisensory loss; 2 hours after elective right carotid catheter arteriogram. Commuted Tomography (CT) head showed fullness and diffuse effacement of sulci in the right hemisphere. CT angiogram head and neck showed patent vasculature. Brain magnetic resonance imaging (MRI) showed no changes associated with early ischemia hence intravenous tissue plasminogen activator (tPA) was deferred. Patient developed two episodes of asystole due to symptomatic cerebral edema and was treated with intravenous mannitol, scheduled 23.4% sodium chloride and dexamethasone taper for the next 72 hours. Electroencephalography showed generalized rhythmic delta activity. Mental status improved on day 4 and patient was subsequently discharged to skilled nursing facility

Contrast induced neurotoxicity is uncommon and should be suspected in patients presenting with encephalopathy and focal neurological deficits, after recent contrast exposure. Neuroimaging should be interpreted in conjunction with clinical history as failure of recognizing this stroke mimic might lead to unnecessary endovascular therapy.

Reversible cerebral vasoconstriction syndrome (RCVS) is a clinical-angiographic syndrome that manifests with thunderclap headache and transient multifocal segmental cerebral artery vasoconstriction. Common risk factors include the use of antidepressants, vasoactive drugs as well as the postpartum period. Red blood cell transfusions have not been well identified as a risk factor for RCVS. We report a rare case of acute brain injury resulting from RCVS after a red blood cell transfusion.

This is a retrospective chart review of a single case.

A 49-year-old female with a history of menorrhagia initially presented with generalized weakness. She was found to have a hemoglobin (Hgb) of 1.7 g/dL in the setting of a fundal fibroid for which she received 5 units of packed red blood cells. Post transfusion she complained of several days of thunderclap headache and later returned with new-onset seizures. She was admitted to the neurocritical care unit for treatment of status epilepticus. Metabolic, infectious and toxic work-up were unremarkable except for an elevated lactate. Magnetic resonance imaging (MRI) of the brain with contrast showed extensive bilateral hemispheric and cerebellar white matter T2/FLAIR hyperintensities with areas of enhancement. A diagnostic cerebral angiogram was performed to evaluate for a vascular etiology and revealed focal segmental stenoses in bilateral A1 segments of the anterior cerebral arteries (ACA) and in branches of the bilateral middle cerebral arteries (MCA). These findings were suggestive of RCVS.

Clinicians should have a high degree of suspicion for RCVS in patients presenting with neurological manifestations, such as thunderclap headache or seizures after recent transfusion. The window for injury may be longer than that seen in other organs, such as in transfusion-related acute lung injury (TRALI).

Opsoclonus Myoclonus syndrome (OMS) is a rare disorder, may present in children or adults, characterized by ataxia, dysarthria, vertigo, myoclonus, opsoclonus and encephalopathy. Etiology is paraneoplastic or parainfectious of which paraneoplastic OMS has worse clinical course.

Case report.

66-year-old man with history of laryngeal cancer s/p resection, admitted with vertigo, nystagmus, opsoclonus, ataxia and myoclonus. Initial workup was unremarkable including labs, MRI brain. CSF showed increased protein and lymphocytosis. He developed agitation, hallucinations, dysarthria and was started on leviteracetam for possible seizures. High dose steroids were started for possible OMS but he quickly progressed to respiratory distress and required intubation. His exam was GCS 3T, preserved brainstem reflexes, opsoclonus and increased tone with myoclonus. Serum GAD antibodies were elevated, 125 IU/ml. He was also treated with IVIG and subsequently plasmapheresis without improvement. Symptomatic management was maximized with leviteracetam, baclofen, diazepam and addition of valproate and phenytoin which were both later discontinued due to hyperammonemia. Whole body CT revealed enlarged supraclavicular lymph nodes and biopsy showed metastatic squamous cell cancer. Patient was diagnosed with paraneoplastic OMS. He was started on lacosamide which led to significant improvement in opsoclonus and myoclonus. Due to diaphragmatic myoclonus and inability to wean off ventilator, he received tracheostomy. He was treated with Rituximab infusion which caused marked improvement in encephalopathy and baclofen, diazepam, levetiracetam, lacosamide were tapered off. Serum GAD antibodies 10 days after rituximab were undetectable. He started following simple commands and was weaned off ventilator. He was discharged to a long-term acute care facility with close follow up.

This is the first reported case of paraneoplastic OMS in setting of laryngeal squamous cell carcinoma which responded to treatment with lacosamide and rituximab and was unresponsive to usual treatment with anti-epileptics, GABAergic medications and other immunomodulatory treatments (steroids, IVIG, plasmapheresis).

Cerebral venous thrombosis is a well-recognized and serious complication in adult treated for acute leukemia, clinically described with several symtomps and signs like: pulsatile holocranial headache, seizures and neurological deficits in the extremities depending on the affected vascular territory. It is much common that it appears related to the intravenous and intrathecal application of L-asparaginase, a common drug in the treatment for certain types of leukemia in humans, as a catalyst of endogenous amino acids vital for neoplastic cells.

We report the case of a 41-year-old man, diagnosed with acute lymphoblastic leukemia, who after receiving this antineoplastic presented headache, decreased muscle strength in the left arm, dizziness, self-limited diplopia and generalized tonic-clonic seizure.

Brain Magnetic Resonance Imaging (MRI) was performed where cerebral venous thrombosis was observed, which rapidly evolved into hemorrhagic cerebral infarction and finally ended in intracranial hypertension and brain death. The rapid evolution of the mentioned vascular event is striking.

The report discusses the relation of the thrombosis, leukemia and treatment with L-asparaginase. Also emphasizes the importance of early recognition and prompt management, while incorporating a collaborative multidisciplinary approach to prevent longterm consequences.

Aneurysmal subarachnoid hemorrhage (aSAH), a prothrombotic condition, is associated with delayed cerebral ischemia and deep venous thrombosis (DVT). In addition to prophylactic heparin for venous thromboembolism (VTE), cerebral angiogram (CA) utilizes intra-arterial heparin. However, sparse reports exist for heparin-induced-thrombocytopenia (HIT) resulting in life-threatening DVT/VTE that is refractory to conventional treatment of HIT.

Case report of a patient who developed type II HIT resulting in critical limb ischemia due to severe upper extremity (UE) DVTs.

Sixty-three year old woman admitted with aSAH due to ruptured anterior communicating artery aneurysm that was secured with coil embolization on hospital day (HD) 1. She received intra-arterial heparin during the CA and was started on prophylactic heparin on HD2. HD8, patient underwent screening CA that showed distal cerebral vasospasm and received procedural intra-arterial heparin. The platelet count dropped to 88,000/μL on HD9 from 248,000/μL on HD7, and she was diagnosed with right UE DVT. HIT antibody was confirmed and she was placed on intravenous argatroban. Due to persistent thrombocytopenia (9,000/μL), she underwent a 4-day course of intravenous immunoglobulin (IVIG) starting HD11. Platelet transfusions were necessary prior to CA to treat symptomatic vasospasm on HD 12 and external ventricular device removal on HD15. On HD18, patient had progression of UE DVT with edema and compromised limb perfusion. Anticoagulant changed to bivalirudin for argatroban failure. With platelet nadir of 4,000/μL on HD20, a 3-day course of methylprednisolone and a 5-day course of IVIG for immunosuppression and immunomodulation led to resolution of thrombocytopenia by HD24. Patient's UE swelling improved with re-establishment of venous circulation.

The etiology of persistent HIT after aSAH is multifactorial: prothrombotic aSAH, heparin exposure, and repeated platelet transfusions. Refractory HIT may need alternate anticoagulation (bivalirudin) and immune modulation to suppress HIT antibody production as well as avoiding platelet transfusions unless absolutely necessary.

Objective: To describe three cases of Cardio-Cerebral Infarction (CCI) with acute proximal L middle cerebral artery (MCA) occlusion, and highlight the variable approaches to management in the absence of evidence-based guidelines.Introduction: CCI is a rare clinical presentation of simultaneous acute ischemic stroke (AIS) and acute myocardial infarction (AMI). Management of CCI is challenging as acute interventions for one organ system may lead to adverse effects in the other.

We describe three cases of CCI in a 43-year-old male, a 72-year-old male, and an 80-year-old female. All three patients presented with symptomatic proximal left MCA (M1) occlusion, ST elevation myocardial infarction (STEMI), and left ventricular apical thrombus. Examination revealed a full LMCA syndrome with aphasia and R sided weakness . Interventional therapies such as alteplase and mechanical thrombectomy (MT) were discussed in all cases, but only alteplase was utilized in one. In one case, left heart catheterization was done without intervention/ stenting. Two patients received therapeutic heparin (no bolus) for LV thrombus; all patients received antiplatelet therapy with aspirin and/or clopidogrel. Ultimately, all three patients were transitioned to comfort care.

CCI poses a therapeutic challenge for practitioners. Clinicians must be mindful of 1) risk of symptomatic hemorrhagic conversion associated with heparin and antiplatelet therapy; 2) relative contraindication for alteplase if an MI has occurred within the last 3 months due to risk of myocardial rupture; and 3) difference in dosage of alteplase for AIS vs AMI. Approach to CCI must be highly individualized based on prior co-morbidities, bleeding risk, and degree of individual organ infarct. Outcome following CCI with LMCA stroke may be poor. We hope more institutions will publish their approach and outcomes.

Clevidipine is commonly used in the critical care setting as a drug of choice in acute hypertension management. It can uncommonly cause hypertriglyceridemia as it is in a lipid emulsion. We report a teenager with no metabolic issues at baseline who developed a rapid increase in serum triglycerides within 24 hours of initiation of a clevidipine infusion in our neurocritical care unit.

Single center retrospective chart review.

A 15-year-old girl was admitted to the neurocritical care unit for further management of left thalamic intracranial hemorrhage due to a posterior choroidal AVM with venous ectasia s/p partial embolization. To optimize her blood pressure control, we initially started a nicardipine infusion which did not achieve blood pressure goals. Then we began a clevidipine drip at 1mg/hour titrated up to 32mg/hour to maintain SBP <110. The following morning her blood tests had abnormalities attributed to the samples being hemolyzed on repeated draws and then suspected secondary to the lipid load in the clevidipine. Her triglyceride level was found to be 3668 mg/dL. The clevidipine infusion was discontinued and she was started on an alternative anti-hypertensive regimen. She had no symptoms of pancreatitis and her lipase level remained normal. Her triglyceride levels trended down to normal range (112 mg/dL) in the next 48 hours. The related hemolysis was problematic for other lab measures. She continued to improve neurologically. Intralipid infusions are known to have impact on triglycerides however this sudden significant increase was unexpected within 24 hours of initiation of the medication. Fortunately we did not find any long term consequences of this metabolic derangement and adverse drug reaction in our patient.

Periodic monitoring of the serum triglyceride levels for patients on clevidipine is warranted to detect this drug induced side effect.

Outbreak of a novel coronavirus began in December of 2019 in China and quickly disseminated across the globe. This virus primarily affects the respiratory system.

In this case report, we discuss a case of acute motor and sensory axonal neuropathy in a patient infected with COVID-19.

Ten days prior to hospitalization, the patient experienced cough and shortness of breath. One day prior to hospitalization, the patient began to experience progressive bilateral lower extremity weakness and paresthesia. Cerebral spinal fluid analysis showed cytoalbuminoloigc dissociation. The patient was started on 400mg/Kg/day of intravenous immunoglobulin for five days. She required intubation on day four of admission due to respiratory compromise secondary to ascending paralysis.

This case report shows that COVID-19 infections are associated with rapid onset of acute motor and sensory axonal neuropathy and should be considered in the differential diagnosis of all COVID-19 associated weakness. We believe that the incidence of such cases may be undiagnosed and underreported in patients requiring early intubation for respiratory compromise. Further investigation is warranted to determine the incidence of acute motor and sensory axonal neuropathy in patients with COVID-19. Thus far, electromyography studies in patients with progressive paresthesia and weakness remains lacking. Further research is also warranted in futility of coronavirus vaccination in COVID-19 associated acute motor and sensory axonal neuropathy.

Segmental arterial mediolysis is a non-atherosclerotic, non-inflammatory arteriopathy causing visceral artery aneurysms as a result of vacuolization and lysis of the outer arterial media. While aneurysms are most commonly found in the abdominal aortic branches, cerebrovascular involvement has been documented. We report a patient presenting with aneurysmal subarachnoid hemorrhage and found to have multiple visceral artery aneurysms.

Literature review and direct patient care and observation were undertaken to evaluate this topic.

A 65 year old female presented with subarachnoid hemorrhage secondary to a ruptured 12mm right MCA bifurcation aneurysm. Right sided craniotomy was performed for aneurysm clipping. The patient had a prolonged ICU stay secondary to vasospasm with delayed cerebral ischemia, hydrocephalus, and prolonged respiratory failure. During hospitalization a CT abdomen incidentally showed multiple visceral artery aneurysms involving the hepatic, superior mesenteric, splenic, and gastric arteries raising concern for segmental arterial mediolysis a potential etiology of subarachnoid hemorrhage.

Segmental arterial mediolysis is a rare but potential causative etiology of aneurysmal subarachnoid hemorrhage. It should be considered in patients presenting with simultaneous subarachnoid hemorrhage and intra-abdominal hemorrhage. Conversely, intra-abdominal hemorrhage from visceral artery aneurysm rupture should be considered in patients with subarachnoid hemorrhage who experience sudden abdominal pain, an acute drop in hemoglobin, or unexplained hypotension or shock. Further case studies and literature review is needed to determine the incidence between the two.

Osteogenesis imperfecta (OI) is a heterogenous group of heritable connective tissue disorders characterized by fragile and brittle bones. Neurovascular manifestations of OI are scarce and most commonly present as cerebral aneurysms. To our knowledge, this is the first case report of an adult with OI presenting with moyamoya syndrome.

Data was collected via chart review.

We present a rare case of a 37 year old man who experienced a syncopal event followed by blindness, inability to move his bilateral legs and proximal bilateral arms. The initial NIH stroke scale was 15. His head CT showed acute strokes in the left ACA, MCA/PCA and right MCA/ACA territory as well as an old left PCA infarct. CT angiogram was significant for marked bilateral ICA and bilateral vertebral artery hypoplasia. Additionally there was multifocal narrowing and irregularity of intracranial vasculature suggestive of moyamoya phenomenon. The rest of the stroke work up remained unremarkable. He was never diagnosed with a stroke before, but patient stated that he had symptoms of dizziness, complicated by fall 6 months prior for which he did not seek medical attention. The patient was diagnosed with OI in childhood. He had suffered numerous bone fractures and complete edentulism due to dentinogenesis imperfecta.

Connective tissue disorders are an uncommon etiology of stroke. However, in patients presenting with stroke, who have a history or clinical features of connective tissue disorders; arteriopathies due to defects in the extracellular matrix must be suspected.

Stroke related to bacterial meningitis in children contributes to high mortality and morbidity among survivors. The etiology of stroke in bacterial meningitis has not been well elucidated. Cerebral vasospasm has been proposed as one of the mechanisms that leads to stroke in bacterial meningitis. Monitoring for vasospasm in meningitis, and treatment options are not well reported in the literature. We present a case of bacterial meningoencephalitis complicated by stroke and refractory cerebral vasospasm treated successfully and safely with intrathecal (IT) nitroprusside. To our knowledge, this is the first report of IT nitroprusside used in pediatrics and for the treatment of vasospasm in meningitis.

This is a case report of a 7-year-old male with autism spectrum disorder admitted to the intensive care unit with Streptococcus Pneumoniae meningitis complicated by multifocal, acute arterial-pattern lacunar ischemic infarctions, status epilepticus, depressed consciousness, and obstructive hydrocephalus with subsequent external ventricular drain placement. The patient had serial transcranial dopplers (TCDs), MR angiogram, and cerebral angiogram indicative of diffuse vasospasm, which was refractory to intravenous (IV) nimodipine, IV milrinone infusion, and cerebral-intra-arterial injection of verapamil. The patient was treated with serial IT (via external ventricular drain) nitroprusside injections of 4mg every 12 hours, for a total dose of 40mg.

After IT nitroprusside, there was normalization of vasospasm by TCD measures, and clinical improvement. His level of consciousness improved, he was extubated, and transferred to rehabilitation, where he has begun ambulating with assistance.

Consider serial TCDs to monitor for vasospasm to prevent stroke in patients with bacterial meningoencephalitis admitted to the intensive care unit and consider IT nitroprusside for cases of refractory vasospasm in bacterial meningoencephalitis.

Reversible cerebral vasoconstriction syndrome (RCVS) is characterized by thunderclap headache, focal neurological deficits and/or seizures due to increase in cerebral vascular tone. Vascular imaging typically shows beaded or sausage-shaped appearance of cerebral arteries. The major complications include cortical subarachnoid hemorrhages and parenchymal strokes. The condition is self-limiting with a favorable prognosis. Herein, we present a case of resistant to treat RCVS that showed clinical and radiographic improvement after stellate ganglion block.

Single-center retrospective chart review.

A 25-year-old, woman was admitted with bi-occipital, bi-frontal, bi-temporal infarcts of varying size. She had a two-week history of recurrent thunderclap headache, photophobia, nausea, vomiting, vertigo, and altered mental status. Following administration of sumatriptan, at an outpatient facility, patient progressed to left face and left upper extremity weakness along with cortical blindness. Head MRA showed diffuse segmental narrowing of the major intracranial vasculature consistent with RCVS. Diagnostic cerebral angiogram showed diffuse moderate to severe vasospasm in both anterior and posterior circulation. We utilized transcranial doppler to monitor vasospasm besides conventional angiogram. Oral verapamil and hypertensive therapy was used as treatment. Multiple doses of intraarterial verapamil improved but did not resolve her focal deficits. Severe headaches persisted. Repeat imaging showed persistent diffuse vasospasm. Given resistance to standard treatment, patient underwent ultrasound guided bilateral stellate ganglion block which resulted in significant improvement in clinical symptoms and the Lindegaard ratio. She showed some clinical improvement and was discharged to a rehabilitation facility.

RCVS management remains largely supportive and includes bed rest, analgesics and removal of precipitating factors. Calcium channel antagonists have been administered with a reported reduction in headache intensity but with minimum effect on the time course of cerebral vasoconstriction. Invasive neuro-interventional techniques and ganglion blocks are therapies that should be explored further as a treatment option in cases that are resistant to conventional treatment.

Approximately 20 percent of ischemic stroke is caused by spontaneous dissection in young adults. Due to subclinical presentations, incidence of dissection may be greater than reported. A reliable method of detection is important for preventing morbidity and mortality from stroke and transient ischemic attacks in young adults. Blacked blood study MRA sequence may have utility in the diagnosis of intracranial vessel dissection. The appropriate early diagnosis of dissection may change the management of presenting patients. Blacked blood sequences MRI technique has been utilized successfully for the diagnosis of other vascular conditions such as vasculitis, subacute hemorrhages, and focal abnormalities of vessel walls.

We report a unique case with left middle cerebral artery (LMCA) dissection that was diagnosed using black-blood MR sequence imaging of the vessel wall as well as NeuroICU management and complications.

The reported case presented with intermittent weakness and numbness and was found to be perfusion dependent.

The potential of this blacked blood MRA technique may substitute invasive procedures such as conventional angiography. In addition it plays a role in a faster diagnosis of dissection and the quicker initiation of dissection management.

While it is well established that hypertonic saline is beneficial in the setting of malignant cerebral edema, maintaining normal to high normal Na levels can theoretically improve neuronal function and reduce cerebral edema in non-emergent cases. This project was performed to determine which modality is most optimal for correction of Na level in non-acute settings

Patients selected from AUMC Neuro-ICU from March 15th to May 15th, 2020. Inclusion criteria: (1) Large hemorrhagic Infarction, (2) Large ischemic infarction. Exclusion criteria: (1) Heart Failure, (2) Severe Hypertension, (3) DI/SIADH/CSW, (4) Na Level < 132 mEq/L. Patients were started on Salt tabs (5 gm x 1, then 3 gm q8hrs) or 3% hypertonic saline (20ml/hr). Primary outcome was Na level 140-145 mEq/L. Secondary outcomes were duration to goal and complications. Correction rate was 4.5-21 mEq/L per 24hr for HS and 2-24 mEq/L per 24hrs for NA. Average duration to primary outcome was NA (25 hrs) and HS (31 hrs).

Both modalities had an equivalent complication rate and success rate . In general, Na tabs had a greater range in correction rate and emesis, but had a shorter duration to goal Na level. Hypertonic saline had a mild but not statistically significant increase in rate of correction > 6/24hrs. From this study, it is reasonable to consider starting Na tabs in settings where hypertonic saline is relatively contraindicated. Further studies investigating benefits on non-emergent correction of relative hyponatremia is warranted.

The 2019 novel coronavirus (SARS-CoV-2) pandemic has been challenging different healthcare systems around the world. We present a case of COVID19 that was complicated with Cryptococcal meningitis. One of the worse characteristics of SARS-CoV-2 is severe cytokine storm, therefore Immunosuppression has been broadly used around the world. Immune reconstitution inflammatory syndrome (IRIS) frequently associated with HIV, but also seen with withdrawal of immunosuppressive medications.

We present a kidney transplant patient on immunosuppression (Mycophenolate, Prednisone, and Tacrolimus) with encephalopathy, cough, fever, and vomiting. COVID-19 positive on admission. Underwent a 5-day course of Hydroxychloroquine, 1 dose of Tocilizumab, and cessation of chronic immunosuppressors. Due to recurrent fever and encephalopathy LP performed on hospital day-25 with WBC 405 (79 segs, 16 lymphs, 5 monos), RBC 84, glucose 20, and protein 1639. Cryptococcal Ag positive and titers of 1:2560 in CSF, 1:640 in blood. Serum CMV PCR positive. Amphotericin, Flucytosine, and Ganciclovir initiated. EVD placed on day-31 due to worsening hydrocephalus and ventriculitis. Lumbar drain subsequently required due to inability to wean EVD on day-53.

Case of COVID-19 associated pneumonia on immunosuppressive medication, overcame infection by discontinuation of these agents and treated with Hydroxychloroquine and Tocilizumab, resulting in negative repeat COVD-19 test and respiratory improvement. Recurrence of fever and decreased level consciousness, Brain MRI showing hydrocephalus and ventriculitis, LP with elevated opening pressure, CSF leukocytosis with evident positive CSF cryptococcal Ag and serum CMV PCR. Development of Cryptococcal meningitis and CMV viremia likely due to augmenting immunosuppression with Tocilizumab and decline in patient condition after discontinuation of immunosuppressive medication suggests IRIS.

Given uncertain treatment guidelines, management of COVID-19 in patients with underlying immunosuppression is challenging. Providers may be forced to discontinue immunosuppressive agents. Provoking CNS-IRIS is anticipated. Recognition of infection will provide an opportunity to save lives and prevent complications.

Reversible Cerebral Vasoconstriction Syndrome (RCVS) is usually associated with benign outcomes. Patients with vascular risk factors, like underlying Sickle Cell Disease (SCD), and sudden onset neurological deficits after a period of flu-like symptoms and/or hemolysis can raise concern for vasoocclusive crisis (VOC). RCVS should not be overlooked when considering VOC or cerebral vasculitis in patients with underlying vascular disease.

We present a teenager SCD patient with flu-like symptoms for 10 days, jaundice and abrupt onset of non-traumatic headache, followed by right side weakness and confusion. MRI brain was remarkable for multifocal strokes and non-traumatic subarachnoid hemorrhage (NT-SAH). Cerebral angiogram showed petrous left-ICA stenosis with post-stenotic bilateral ACA and left-MCA branches dilation. Autoimmune vasculitis workup was unremarkable except for elevated ANA titer. Two months later, he was readmitted for recurrent thunderclap headache. Head CT showed new frontal NT-SAH without new infarct.

Repeat cerebral angiogram showed stable right-ICA stenosis with post-stenotic dilatation, occluded right-ACA with flow reconstitution, and improvement of left-ACA and MCA stenosis. Interval improvement of intracranial stenosis supports the diagnosis of RCVS. Treatment with erythropheresis and Verapamil were associated with a good outcome. Outpatient follow up neurologic exam was normal, and TCD revealed normal anterior/ophthalmic/posterior circulation and asymmetric MCA velocities.

SCD vasculopathy can be a risk factor for RCVS. RCVS exacerbation can be precipitated by similar risk factors as VOC. Due to a lack of diagnostic criteria for RCVS, extensive workup is often required. Early recognition and management with erythropheresis and calcium channel blocker can lead to better outcomes and decreased hospital stay in RCVS. Further studies regarding diagnosis, treatment methodology and outcomes for RCVS in patients with underlying vasculopathy are needed.

Extracorporeal membrane oxygenation (ECMO) is being used more frequency for cardiopulmonary support. Neurologic complications, including stroke are common in patients on ECMO. Recognition of potentially treatable neurologic complications is important. ECMO is not an absolute contraindication to many neurointerventional treatments. Catheter-directed neurointerventions are being performed for stroke in cardiac patient with good results. Although literature is sparse, catheter-directed neurointerventions should be considered for patients on ECMO with acute stroke.

A 48-year-old male presented in cardiogenic shock related to arrhythmias. He was placed on mechanical ventilation and cannulated peripherally for VA ECMO. To unload the left ventricle an Impella® was inserted in the left femoral artery. A heart catherization was unremarkable and he was extubated on day three of his ECMO run. Echocardiograms showed improving heart function and removing the Impella® was planned while continuing ECMO support. On day four of his ECMO run he went into atrial fibrillation with rapid ventricular response. A transesophageal echocardiogram showed no intracardiac thrombi and he was cardioverted to sinus rhythm. Soon after cardioversion he developed right side facial drop, arm and leg weakness. A CT angiogram and perfusion study showed an acute M1 middle cerebral artery occlusion. After Impella® removal the femoral artery was used for neurointerventional access. A thrombectomy was performed and revascularization was achieved with full perfusion. Over several days, his neurologic deficits and heart function improved. De-cannulation occurred on ECMO day eight and the rest of his course was unremarkable

This patiented in cardiogenic shock, was supported with ECMO and developed an acute MCA stroke. The patient received a thrombectomy and an excellent result was achieved.

Patients on ECMO are at risk of neurologic complications. Prompt recognition neurologic complications may allow patients to be candidates for neurointerventional treatments. ECMO is not an absolute contraindication for neurointervential treatments.

Diaphragmatic pacing (DP) has been shown to decrease morbidity, mortality and length of hospital stay in patients with respiratory failure from diaphragmatic dysfunction. However, patients with neuromuscular diseases are commonly excluded from DP studies. We present three novel cases of DP in the treatment of failure to wean from the ventilator in the setting of acute neuromuscular respiratory failure.

Cases of acute-onset neuromuscular respiratory failure were identified in the diaphragmatic pacing database with additional information obtained through chart review.

Case #1: Guillain-Barre Syndrome. 68-year-old woman with ascending numbness/weakness 3 weeks after influenza vaccination, progressing to respiratory failure 3 weeks after symptom onset despite IVIG and plasmapheresis. She underwent DP 3 months after intubation and weaned 2 months later. Case #2: Botulism. 70-year-old woman with rapidly progressive bulbar and generalized weakness after ingestion of contaminated shellfish, progressing to respiratory failure 3 days after symptom onset. She underwent DP 3 months after intubation and weaned 2 months later. Case #3: Myasthenia Gravis (MG). 54-year-old woman with one year history of diplopia, ptosis and bulbar weakness presented with acutely worsening symptoms and respiratory failure requiring intubation. She was diagnosed with anti-MUSK MG and felt to be in crisis. She underwent DP 1.5 months after intubation and weaned 17 days later. All cases had little if any detectable diaphragm EMG activity at time of placement and were ultimately weaned from DP.

This is the first report of DP used to successfully treat failure to wean in the setting of neuromuscular respiratory failure. All cases had significant symptoms of disease at the time of DP placement suggestive of ongoing neuromuscular disease. Given the likelihood of superimposed ventilator-induced diaphragmatic dysfunction, earlier initiation of DP should be considered. Larger trials are needed to determine the benefit of DP in this population.

Vaping has become a widespread recreational drug activity, with products spanning from nicotine to tetrahydrocannabinol (THC) combined with other unregulated substances and flavors. Although recently vaping-associated lung injury has become more known, there exists a gap in the literature regarding an association with status epilepticus.

A case series of four patients presenting with status epilepticus associated with vaping.

Four patients, aged nineteen to thirty, presented in convulsive status epilepticus with three also having documented non-convulsive status epilepticus. All patients regularly and frequently vaped THC. Two patients also vaped nicotine. None of the patients had seizures prior to vaping. Past medical history was notable for a remote subdural hematoma and a cysto-peritoneal shunt in one patient who had no prior seizures until he increased frequency of vaping THC. The other three patients had no significant past medical history. Imaging was notable for hippocampal atrophy in one patient, who along with the other two patients also had sequelae of seizures noted on imaging. One patient had a prodrome of malaise and vomiting, however the rest were all in their usual state of health, without sleep deprivation, recent head trauma, or infection. All seizures were eventually controlled with medications and patients were able to be discharged. One patient had super-refractory status epilepticus with residual cognitive deficits. Beyond vaping, no clear precipitating factor was found in any of the cases despite extensive work up.

Though vaping has been associated with seizures, we believe this may be the first case series of THC vaping linked to status epilepticus. All four patients habitually vaped THC though two also vaped nicotine. Further research is required to delineate the components in vaping that may precipitate seizures. Public education on the potential link between vaping of THC and status epilepticus may be needed.

Anti-NMDA receptor encephalitis is a well described immune mediated neurological disease with neurological and psychiatric manifestations. Standard treatment involves escalating degrees of immunosuppression, however relatively few cases have been reported regarding best management strategies in pregnant patients presenting with NMDAR encephalitis.

Direct clinical observation and management as well as data was reviewed from electronic medical records for this case report.

A 33 year old previously healthy woman presented 14 weeks pregnant with refractory seizures preceded by two weeks of headache, progressive behavioral changes and impaired memory. Her initial evaluation was notable for EEG that showed focal status epilepticus with frequent bitemporal electrographic seizures and MRI with abnormal signal in the bilateral insula, right anterior temporal and inferior frontal lobes extending to the bilateral hippocampi. MRI of pelvis showed a 3.4 cm teratoma in the right ovary, while CSF studies were notable for lymphocytic pleocytosis (WBC 49/μL) and the presence of oligoclonal bands. She was taken to the OR for salpingo-oophorectomy and teratoma removal. This was accompanied by 5 days of high dose steroids and plasma exchange, and later by two courses of IVIg due to minimal behavioral improvement. After surgery and initiation of immunosuppression, result of CSF NMDAR antibody titer returned positive.

NMDAR encephalitis is an uncommon neuroinflammatory condition with accepted clinical and diagnostic criteria, as well as initial management strategies. However, cases reported during pregnancy are uncommon and even fewer with associated teratoma. Our patient presented with neuropsychiatric symptoms and seizures consistent with typical NMDAR encephalitis and showed significant improvement in symptoms after teratoma removal and initiation of immunosuppression. She continues to clinically improve, although as of this writing, she continues to have neuropsychiatric and behavioral problems, now off all immunosuppression. Her fetus remains viable although does have evidence of mild polyhydramnios on ultrasound at 36 weeks.

Stiff Person Syndrome(SPS) is a rare autoimmune disorder and is often confused with seizures. It is characterized by progressive rigidity and muscle spasms affecting the axial and limb muscles. Clinical diagnosis of SPS may be delayed in patients with underlying seizure disorder. Anti-glutamic acid decarboxylase (anti-GAD65) antibodies are found in majority cases of SPS.

A 35 years old female with medically refractory temporal lobe epilepsy underwent right amygdalahippocampectomy in 2009. Patient was seizure free until 2019 when she developed spells of severe muscle spasms, palpitations, dyspnea, and panicked restlessness, and occasional fever of 39 C. On admission, she had sinus tachycardia and forceful bilateral leg thrashing that resolved with benzodiazepines. creatinine kinase(CK) and lactate were found. Examination revealed increased spasticity in bilateral lower extremities, a wide based gait, and hyperreflexia with crossed adductor sign and bilateral Babinski sign. Serum and cerebrospinal fluid(CSF) labs were obtained, and two typical spells occurred during cvEEG monitoring. Subclinical seizures, neuroleptic malignant syndrome and psychogenic nonepileptic spells were considered high in differential diagnosis.

Typical spells did not correlate with seizure activity on cvEEG. CSF analysis showed normal opening pressure, gram stain, culture, cell count, protein, glucose. Serum and CSF GAD65 antibody levels were 5153nMol/L and 53 nMol/L respectively along with 14 oligoclonal bands in CSF. Diagnosis of SPS was made. Treatment with diazepam caused improvement in spells and myalgia. CK trended down from 5096 to 1080 U/L over the next three days and lactic acidosis resolved.

Few cases of SPS and epilepsy have been reported. Ours is first case of SPS in patient with prior hippocampal resection. Development of autoimmunity to neuronal GAD65 protein, exposed during surgery, is suspected in pathogenesis. Early recognition of SPS with serum GAD65 antibody testing can prevent misdiagnosis. Appropriate therapy can stop spells and prevent rhabdomyolysis, lactic acidosis, and autonomic dysfunction.

Cerebral amyloid angiopathy is the most common cause of nontraumatic lobar hemorrhage in older adults, resulting in significant morbidity and mortality. The diagnosis is often suspected based on bleed location, patient age, and presence of microhemorrhages. However, in this population, the incidence of malignancy also increases and is often part of the differential. While advanced imaging can be helpful in differentiating the cause of hemorrhage, it is not a definitive diagnostic tool and at times can be misleading.

We present a case of a 78-year-old male with atrial fibrillation on Coumadin and history of metastatic prostate cancer who presented with one-week history of lethargy and confusion in the setting of a supratherapeutic INR. Information was obtained through chart review.

CT head showed a large right temporal intracerebral hemorrhage (ICH) with surrounding vasogenic edema and 0.2 cm midline shift. Subsequently, MRI brain with and without gadolinium was obtained demonstrating a 4.4 x 4.0 cm lesion with peripheral contrast enhancement and associated vasogenic edema. GRE sequence did not show evidence of significant microhemorrhages. Based on the radiographic appearance and clinical history, suspicion for underlying neoplasm was high and the patient underwent a right craniotomy for tumor resection. Pathological evaluation of intraoperative tissue samples showed cerebral amyloid angiopathy with hemorrhagic necrosis. No viable tumor cells were identified and a diagnosis of cerebral amyloid angiopathy induced ICH was made.

MRI alone cannot conclusively determine the etiology of ICH. Multiple microhemorrhages are not necessary for cerebral amyloid induced ICH and rarely, patients may present with tumefactive cerebral amyloid angiopathy or amyloidoma that mimics an underlying malignancy. A tissue sample is required to definitively diagnose the etiology of ICH, however this is impractical in most cases. Further markers (clinical and radiographic) are needed to more accurately identify the etiology of ICH through noninvasive means.

UTHSC: Neuroscience, Memphis, TN, United States

We present a case of ruptured intracranial dermoid cyst in a 57 yo African American female.

This case report is noteworthy because of her rare imaging finding and unusual diagnosis for a common presenting symptom: thunderclap headache.

The patient presented with "the worst headache of my life" lasting approximately 20 minutes at its maximum intensity. Despite relief with acetaminophen, the patient called 911 to be evaluated. She underwent emergent noncontrasted CT scan of her head which revealed a large (4 x 4.5 centimeter) ruptured dermoid cyst in the right frontal lobe region. Characteristically, there was prominent, scattered subarachnoid and intraventricular fat. Additionally, there was a prominent calcification at the ruptured cyst's inferior margin, and a significant right to left midline shift with associated sub-falcine herniation. She was admitted to the neurological intensive care unit for supportive care and close observation. Over the course of the initial 48 hours after admission, she developed some mild left arm weakness; this combined with the midline shift and mass effect led to neurosurgical intervention on hospital day 3. The patient underwent right pterional craniotomy and subsequent microsurgical resection in the right frontal lobe and along the right Sylvian fissure.

Intracranial dermoid cysts are a rare and benign etiology and arise in the first trimester of embryonic life from tissue of ectopic ectodermal origin incorporated into the neural tube. These insidious lesions are often composed of a stratified squamous epithelium capsule with dermal components comprising of sweat and sebaceous glands, cholesterol, and hair follicles with chronic expansion due to intra-cystic accumulation of desquamated debris and sebaceous secretions. Space occupying symptoms manifest predominantly in the second and third decade of life; patients present with headache, visual symptoms, dizziness, seizures and sensory / motor deficits (secondary to cerebral ischemia); however, most patients are asymptomatic and identified incidentally.

Post-stroke delirium (PSD) affects 13-48% of stroke patients, increasing their risk of in-hospital and 1year mortality, length-of-stay, and poor functional outcomes. Yet guidance on pharmacotherapy is lacking. Due to its unique pharmacological properties and intramuscular formulation, phenobarbital is gaining popularity in adult critical care. Literature supports its use for prophylaxis and treatment of alcohol withdrawal syndrome, agitation during end-of-life care, and sedation for those refractory to typical regimens. Here we present the first known case of successful management of severe, refractory post-stroke hyperactive delirium with phenobarbital.

A case report from a critical care unit that was exempt from review by and deidentified per our institutional review board process.

A middle-aged man with left middle cerebral artery (MCA) stroke, urinary tract infection (UTI), and bacteremia presented with severe hyperactive delirium. His symptoms lasted 13 days despite treatment and rapid resolution of infection. He proved refractory to high-dose trials of typical and atypical antipsychotics, oral and continuous infusions of alpha-2 agonists, gabapentin, therapeutic valproic acid, benzodiazepines, and ketamine infusion. He continued to require physical and intermittently violent restraints on day 12, prompting family to consider limiting goals of care. Despite low serum drug levels of phenobarbital, a dramatic improvement in the patient's hyperactive delirium was observed within 36 hours of initiation of drug. His mixed aphasia improved, he avoided gastrostomy tube and was cooperative with therapies facilitating discharged to a skilled nursing facility for rehabilitation and home soon after.

Post-stroke delirium is a common complication with significant consequences on patient outcomes and little evidence to guide treatment. Our report supports the use of phenobarbital for management of severe PSD and suggests potential for broader clinical use as a safe and effective treatment option for refractory hyperactive ICU delirium of any etiology.

NYU Langone Medical Center, Department of Neurology, New York, NY, United States

Wernicke's encephalopathy (WE), an acute neurological presentation of thiamine deficiency, is typically associated with alcoholism, but it can be a rare complication of Crohn's disease. Confusion, ophthalmoplegia, and gait instability are the classic triad of WE, but seizures may also occur.

Case report.

A 59-year-old woman with Crohn's disease requiring multiple surgeries presented with nausea, vomiting, and weight loss of 50lbs for several months. She was treated for cellulitis and electrolyte abnormalities. Several days after initiation of antibiotics and nasogastric feeds, she had 3 generalized tonic-clonic seizures. She was given keppra and vimpat, then intubated, started on propofol, and transferred to the Neuro-ICU. She was noted to have profound encephalopathy and lateral gaze palsy. Video EEG demonstrated frequent bioccipital sharp and slow wave complexes, predominantly in sleep, present both in isolation and in very brief runs of 3 Hz. MRI demonstrated restricted diffusion and FLAIR signal abnormalities in the thalami, periaqueductal midbrain, tectal plate, and mammillary bodies consistent with WE. Intravenous thiamine 500mg every 8 hours was initiated. Further workup revealed deficiencies in thiamine, vitamin A, vitamin D, folate, and zinc; she was supplemented accordingly. Mental status and ophthalmoplegia initially improved with thiamine, but she ultimately developed other medical complications, was palliatively extubated based on her prior wishes, and died.

There are few case reports of WE resulting from Crohn's disease. Thiamine deficiency occurs in Crohn's disease due to malabsorption, decreased dietary intake, and a catabolic state. Thiamine plays an important role in glucose metabolism and neurotransmitter synthesis. Deficiency impacts GABA and glutamate levels which can result in seizures, particularly in the setting of concurrent metabolic derangements. Treatment consists of intravenous thiamine. The prognosis for nonalcoholic WE is favorable; however, persistent neurologic dysfunction is common. 

Amaurosis fugax is a sudden transient monocular visual loss with different etiologies that requires an immediate evaluation to rule out possible devastating consequences. Usually the most frequent cause are thromboembolic events. Herein we present a rare case of amaurosis fugax in a subarachnoid hemorrhage after coiling of a paraclinoid aneurysms.

B-mode and Color-Doppler ultrasound and automated pupillometry were used as point-of-care diagnostic devices.

A 20 years-old male with a silent past medical history were admitted overnight for a sudden excruciating headache (GCS 15). A brain CT-scan showed a subarachnoid hemorrhage (Fisher modified 1, Hunt-Hess grade 2). The subsequent angiography revealed a 9mm left supraclinoid rupted aneurysm of the internal carotid artery which was coiled. No complications were evident at the end of the procedure and the patient was admitted to ICU. Thirty minutes later, during the neurological assessment, the patient reported a left amaurosis. To rule out secondary etiologies an immediate evaluation with point-of-care devices were performed. An occlusion of the ophthalmic artery and an ICP increase were excluded by a color-doppler and B-mode ultrasound, as the blood-flow and the optic nerve sheath diameters were in the normal range in both eyes. As a further test the Neurologic Pupil index was evaluated bilaterally. All values were within physiological ranges, providing evidence of the responsiveness of the third cranial nerve. Those findings were confirmed by a brain AngioCT-scan. The symptomatology resolved progressively over 120 minutes with low-dose steroid therapy, 30° head of bed and mean blood pressure management.

Amaurosis fugax can occur after endovascular procedure and should be investigated. Suspected visual loss is a neurological emergency that deserve a prompt evaluation. Ultrasound and pupillometry were presented as two useful point-of-care diagnostic tools for neurointensivist as an immediate assessment of the second and third cranial nerve.

Spinal cord infarction accounts for 1% of all central nervous system infarction. It is mostly due to traumatic or surgical etiologies. Sickle cell disease (SCD) is a rare cause of spinal cord infarction. Here we report a patient with SCD who experienced an anterior spinal cord infarction and subsequent paraplegia.

This is a retrospective chart review of a single case.

A 36-year-old African American female with a history of sickle cell beta thalassemia and insulindependent diabetes who presented with an acute onset of abdominal pain, nausea and vomiting. While in the emergency department, she rapidly developed urinary retention, bilateral lower extremity weakness and paresthesia from the lower sternum to her toes. Neurological examination revealed paraplegia, absence of patellar and ankle reflexes, and a sensory level at T6 dermatome characterized by losing pain/temperature sensation below that level with relative sparing of vibration/joint position sense. Magnetic resonance imaging (MRI) of the thoracic spine revealed intramedullary central T2 hyperintensity on FLAIR extending from T6 to T10 with corresponding restricted diffusion on diffusionweighted imaging/apparent diffusion coefficient (DWI/ADC) map. The clinical and radiological findings were suggestive of anterior spinal cord infarction. Hemoglobin electrophoresis revealed HbS of 71.4% with elevated HbA2 and Hb S/β Th. Patient underwent red blood cell (RBC) exchange transfusion which lowered HgbS level to 18% but with no significant improvement in her symptoms. She was started on Aspirin and no other interventions were attempted. She was subsequently discharged to rehab.

We report a rare case of spinal cord infarction in the setting of sickle cell disease. Some cases might be more subtle and physicians should be concerned about spinal cord infarction when encountering SCD patients in the appropriate clinical scenario.

Infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been associated with numerous neurological sequela, including Guillain-Barre syndrome (GBS).

We present two patients who presented with severe quadriparesis, areflexia, eye-movement abnormalities and albuminocytological dissociation after prolonged ICU admissions for COVID-19. Both were diagnosed and treated for GBS before ultimately being determined to have severe critical illness neuromyopathy based on electromyography and nerve conduction study (EMG/NCS) findings.

A 20-year-old female and a 58-year-old female presented to outside hospitals with COVID-19 and were intubated and sedated for multiple weeks. After extubation, they were both noted to have diffuse weakness, areflexia, and eye-movement abnormalities, for which they ultimately required reintubation, and were then transferred to the neurological ICU at our institution. In both patients, CSF studies were notable for elevated protein and normal white blood cell count. Both were treated with 0.4g/kg IVIG for 5 days. EMG/NCS studies were delayed while the patients remained positive for SARS-CoV-2, but were ultimately performed 6 and 8 weeks after the patient's initial COVID admission. EMG/NCS in both patients did not show any evidence of demyelination, but were instead consistent with a combined critical illness neuropathy and myopathy in one, and critical illness myopathy alone in the other.

Despite presenting to our institution with severe weakness, areflexia, eye-movement abnormalities and albuminocytological dissociation, these two patients were ultimately diagnosed with critical illness myopathy. Clinicians should be mindful of the potential for GBS in patients with COVID-19, but at the same time recognize that critical illness myopathy is extremely common in patients requiring prolonged intubation, and therefore is a likely etiology even in patients with classic exam and CSF features of GBS. EMG/NCS can distinguish the two diagnoses and potentially avoid costly and inappropriate treatments.

Young patients with malignant cerebral edema (MCE) have been shown to benefit from early decompressive hemicraniectomy (DHC). The impact of concomitant infection with COVID-19 and how this should weigh in on the decision for surgery is unclear. Here we report a case series of COVID-19 positive ELVO patients admitted for MCE monitoring.

We retrospectively reviewed all COVID-19 positive patients admitted to the Neuroscience ICU for malignant edema monitoring from 3/24/2020 to 4/30/2020. Patients with >50% of middle cerebral artery involvement on CT-imaging were considered at risk for malignant edema. All patients were followed until death, discharge from the hospital, or the data cut-off date, whichever came first. Good outcome on discharge was defined as awake, alert and able to participate in rehabilitation.

Seven patients were admitted for monitoring of which four died. Cause of death was related to COVID-19 complications and these were either seen very early or several days into the ICU course after the typical window of malignant cerebral swelling. Three cases underwent DHC and one of these patients died post-operatively from cardiac failure. A good outcome was attained in the other two cases.

COVID-19 positive patients with large hemispheric stroke can have a good outcome with decompressive hemicraniectomy. Mortality in our patients appeared related to the downstream complications of severe COVID-19 infection including refractory shock and renal failure, rather than the stroke itself. Thus, clinicians should be mindful of these complications, but a positive test for COVID-19 should not be used in isolation to exclude patients from a potentially lifesaving procedure.

Edoxaban is factor Xa inhibitor approved for treatment venous thromboembolism (VTE) and atrial fibrillation (AF). Currently, there are no approved reversal agents for edoxaban.

This case report describes the use of andexanet alfa for reversal of an edoxaban-associated subdural hematoma (SDH).

An 82-year-old male presented to the emergency department (ED) after a fall resulting in head trauma two days prior with ongoing unsteadiness and new onset left-sided vision loss, headache, and rightsided weakness. GCS upon arrival was 15, and he was noted to have neurological decline. Baseline laboratory results revealed: Hgb 12.5 g/dL, platelets 121 x 109/L, and INR 2.0. He was on edoxaban 60mg daily for AF, last taken > 8 hours prior to arrival. A non-contrast CT of the head showed a large acute left SDH measuring 2 cm in dimension with approximately 1 cm of left-to-right midline shift with significant mass effect. The patient met criteria for acute surgical intervention. He was reversed with low-dose regimen of andexanet alfa. Subsequently, he was taken to the operating room for hematoma evacuation with no evidence of further hemorrhage and an estimated blood loss of 500 mL, within the expected range for this type of emergent procedure. Initial post-operative CT demonstrated complete evacuation with resolution of midline shift with no re-accumulation. At this time a thromboelastogram showed normal parameters and INR was 1.4. The patient was hospitalized for 19 days with no thrombotic events noted and no re-accumulation of intracranial blood products. Unfortunately his recovery was complicated by seizures, encephalopathy and decompensated heart failure, and he was transitioned to hospice, expiring 19 days after emergent surgical procedure.

This case report describes the use of andexanet alfa for effective reversal of edoxaban prior to an emergent neurosurgical procedure without any intraoperative or post-operative bleeding or thrombotic complications.

To describe an interesting case of SMART syndrome (stroke-like migraine attacks after radiation therapy) in a young female who presented with new onset headache, aphasia and status epilepticus. SMART syndrome is a rare disorder presenting with new headache or migraine and a focal neurologic deficit months to years after radiation therapy. Status epilepticus is rarely reported with this condition.

A 38-year-old female with a history of a left frontal meningioma status-post excision and brain radiation fourteen years prior presented with 3 days of headache, nausea/vomiting, and aphasia. Her initial MRI without gadolinium showed several small recurrent meningiomas of the bilateral upper cerebral convexities and left hemispheric cortical diffusion restriction. She developed seizure activity with right arm and leg clonic activity. EEG changes were consistent with a left temporal seizure lasting approximately 25 minutes. She was given lorazepam 2mg followed by levetiracetam 2g and lacosamide 200mg. Following the seizure, EEG showed abundant lateralized periodic discharges (LPD) in the left parieto-occipital region. Both the seizure focus and LPD+ were from different areas than her new meningiomas. Clinically she developed profound new right sided weakness with persistent aphasia. Repeat MRI with gadolinium showed gyriform enhancement involving the cortex of the entire left hemisphere, most prominent in the left frontal and temporal lobes, concerning for SMART Syndrome. She was started on dexamethasone 4mg every 6 hours for three days. After completion of steroids, her MRI and clinical exam improved with only mild residual diffusion restriction in the left parietal region and significant improvement in her right sided weakness.

SMART syndrome may present with status epilepticus. While the dose and duration of steroids is unclear, three days of dexamethasone lead to significant clinical and radiographic improvement in this case.

Super-refractory status epilepticus (SRSE) occurs in 10% to 15% of patients with status epilepticus and is associated with poor outcomes. There are many potential but unproven treatments including immunosuppression, ketogenic diet, and invasive surgery. Invasive surgery, including lesionectomy, has been studied most prominently in children with simple focal lesions. In adults, the most common cause of new onset of epilepsy is chronic stroke, which causes regional and diffuse disruption in cerebral connectomes. We present a case of SRSE from chronic stroke in which lobectomy was attempted.

A 69-year-old woman with history of acute ischemic stroke and subsequent seizures, presented with nonconvulsive status epilepticus with seizures emanating from the area of stroke (right frontal lobe). She was sedated and intubated but had breakthrough seizures despite ketamine, midazolam, and propofol infusions. Over the course of the subsequent 4 weeks continuous sedation was weaned, but she relapsed into NCSE several times. The decision was made to place a subdural EEG grid for planning and resection of the epileptogenic region.

The subdural EEG grid revealed multifocal seizure foci which were removed via open craniotomy. Continuous EEG was free from any seizure activity until post-op day 3 when focal seizures emerged from a previously electrographically quiet area of her right posterior head region. She never returned to SE, but continued to have at least 1 seizure daily. She was discharged to a long-term care facility on hospital day 41 on 3 ASDs (levetiracetam, clobazam, lacosamide) and on a ketogenic diet.

SRSE is a devastating disease with no proven treatment. Though surgical resection may be an appealing option, it may only be reserved for simple and focal lesions. This case reveals the widespread nature of cerebral connectomes and that seizures from chronic stroke have regional and diffuse involvement.

The novel coronavirus SARS-CoV-2 is known to cause hypoxemia and acute respiratory distress syndrome (ARDS) in a significant portion of those affected. Survivors of critical illness and ARDS often experience neurocognitive impairment but, to date, there is a paucity of literature linking MRI findings of hypoxic brain injury to ARDS-related hypoxemia. In this case series, we describe three cases of hypoxic brain injury seen on MRI in patients with hypoxemia secondary to COVID-19 related ARDS.

In this retrospective study, the clinical courses and radiographic findings of three patients with COVID-19 related ARDS were reviewed. All patients were seen by the General Neurology service due to encephalopathy following prolonged mechanical ventilation. The patients selected were neurologically intact at baseline and did not suffer from cardiac arrest during their clinical course. Clinical attributes including SpO2, PaO2, P/F ratio, and hypotension were reviewed.

All reviewed patients required prolonged mechanical ventilation as well as the use of vasopressor therapy. There were no documented episodes of hypoglycemia in any patients included in this study. Interestingly, despite the lack of severe observed hypoxemia in two of the three patients, these patients developed hypoxic brain injury. The same two patients also developed cytokine release syndrome (CRS) requiring tocilizumab, a monoclonal antibody against the IL-6 receptor. Each patient remained below their neurologic baseline at the time of discharge.

The lack of severe observed hypoxemia in two of the cases suggests that unrecognized or asymptomatic hypoxemia may play a role in hypoxic brain injury related to COVID-19. We also propose that the presence of cytokine release syndrome should be further studied as a contributor to the neurologic sequelae seen in these patients.

Neurology Department, Spectrum Health Grand Rapids, Grand Rapids, MI, United States

Amyloid-beta related angiitis (ABRA) is a form of inflammatory, cerebral, amyloid-related, angiopathy, which presents as CNS vasculitis. Patients with ABRA typically present with altered mental status, seizures, and focal neurological deficits. These symptoms are also seen in other neurological disease processes like primary CNS vasculitis and posterior reversible encephalopathy syndrome; therefore it delays early treatment of ABRA. We present a case of ABRA that was initially diagnosed as posterior reversible encephalopathy syndrome.

A 77-year-old male with a past medical history of hypertension presented to the ED with a 1-week history of progressive confusion and lethargy. The patient was noted to have altered mental status, with an initial blood pressure of 240/100. A CT scan was performed which showed significant hypoattenuation bilaterally, concerning for cerebral edema, right greater than left. The patient was admitted to the medical ICU with a working diagnosis of posterior reversible encephalopathy syndrome. The patient was started on IV nicardipine. MRI was done which showed vasogenic edema with cortical ischemia in the left parietal and temporal lobe. MRA showed luminal irregularity involving distal arterial branches. The patient was transferred to neurocritical care, placed on an EEG monitor. A spinal tap was performed which showed WBC 0, RBC 38, and protein of 394 with normal glucose. Infective workup was negative. The patient was noted to have non-convulsive status epilepticus requiring multiple AEDs, was later intubated and sedated. A diagnostic angiogram was done which was concerning for vasculitis for which systemic vasculitis workup was done which negative.

A repeat MRI was done with GRE sequencing which showed similar cerebral vasogenic edema and GRE concerning multiple microhemorrhages. A brain biopsy was also done which showed amyloid-beta related angiitis. The patient was started on steroids followed by cyclophosphamide.

Brain biopsies remain the gold standard for the diagnosis of ABRA. Earlier diagnosis can prevent significant disability and neurological impairment as immunosuppression has shown to improve outcomes.

Ventriculoperitoneal shunt (VPS) infections due to methicillin-resistant staphylococcus aureus (MRSA) can result in both meningitis and ventriculitis and can rarely present as diffuse vasospasm and result in ischemic stroke. It is suggested that inflammatory cytokines are the predominant mechanism of elevated cerebral blood flow (CBF), leading to vasospasm and ischemic stroke. Additionally, it has a high rate of mortality (30-50%).

A 41-year-old-woman with history of epilepsy, migraines with aura, and congenital hydrocephalus status-post VPS since age of 15, presented with right-sided hemiplegia, somnolence, and generalized tonic-clonic seizures. Two weeks prior, patient had discontinuation of the VPS due to concern of CNS infection manifesting with headaches and worsened mental status, and discharged home on oral antibiotics. In our hospital, the non-contrast CT-head showed severe communicating hydrocephalus. Brain-MRI demonstrated T1 contrast enhancement of the lateral ventricles and areas of restricted diffusion with ADC correlation in the left anterior cerebral artery territory and right occipital lobe consistent with infarction. CT head angiogram with no arterial occlusion, but diffuse vasospasm. Therefore, an external ventricular drain was placed urgently. CSF studies consistent with bacterial meningitis and cultures grew MRSA. She started IV vancomycin and cefepime. Serial transcranialdoppler (TCD) revealed increased mean CBF-velocities in middle cerebral arteries (right:153cm/s, left:140cm/s). Patient received intraventricular vancomycin with remarkable laboratory, imaging, and examination improvement. She remained with mild right-side hemiparesis.

Ventriculitis and meningitis are a life-threatening neurological emergency and rarely provoke cerebral vasospasm which can lead to infarction. If left untreated, as in our case for one week, it can lead to severe disability or death. Surgical removal of infected VPS and placement of CSF flow diverting device is essential in the initial management. Medical management would include systemic IV antibiotics as well as prompt consideration of intraventricular antibiotics for improvement of vasospasm.

45-year-old man with HIV admitted for confusion and new onset seizures. MRI showed FLAIR hyperintensities involving the bilateral hippocampi, temporal lobes without enhancement. He was diagnosed with HSV2 encephalitis with CSF evaluation and was treated with acyclovir showing significant improvement. He was readmitted after 1.5 months with worsening confusion, episodes of unresponsiveness and gait abnormality. CSF showed increased protein and mononuclear cells. MRI showed worsening FLAIR hyperintensities in bilateral hippocampus and parietal lobes without enhancement. Differentials included herpes encephalitis, HIV encephalopathy, progressive multifocal leukoencephalopathy, limbic encephalitis. He was treated with acyclovir for concern of incompletely treated HSV encephalitis and high dose steroids for presumed NMDA encephalitis with some improvement. CSF autoimmune encephalitis panel was negative. He was readmitted in 1 month with worsening confusion, shortness of breath, fever, hematemesis, and respiratory failure and treated for klebsiella pneumonia. He had seizures and VEEG showed nonconvulsive status epilepticus. Antiepileptics were escalated and he was intubated. LP showed elevated protein, lymphocytosis and oligoclonal bands. Encephalitis panel, autoimmune/paraneoplastic panel were negative. MRI showed worsening FLAIR hyperintensities involving bilateral hippocampal and frontal subcortical white matter. Brain biopsy showed reactive/inflammatory infiltrate, later confirmed to be CD8+ on immunohistochemical stains. He was diagnosed with CD8+E. Treatment with steroids was deferred by family due to concern of worsening behavior and infections.

CD8+E is a rare HIV associated neurological disease which can be fatal if untreated. CD8+E is steroid responsive. There is no consensus on long-term management. Some studies suggest use of mycophenolate mofetil. Early recognition and prompt treatment can reduce morbidity and mortality

Neuromonitoring can detect physiological disturbances and help minimize secondary injury in patients with brain injury. Recommendations support intracranial pressure (ICP) monitoring in a variety of patients with traumatic brain injury (TBI). Extracorporeal membrane oxygenation (ECMO) is being used with increasing frequency for cardiopulmonary support. ECMO has risks of neurologic complications and is being used with increasing frequency in patients with neurologic injury.

This is a report of a patient with a TBI with an ICP monitor on ECMO for hypoxemic respiratory failure from pulmonary contusions. Additional neuromonitoring included cerebral oximetry and continuous electroencephalogram (cEEG). Also included is review of literature on neurologic complications, invasive and noninvasive neuromonitoring in patients on ECMO.

A 24-year-old male suffered a motor vehicle accident resulting in severe TBI, orthopedic injuries, pneumothorax and pulmonary contusions. Brain imaging showed contusions, Intraparenchymal and subdural blood and edema. His Glasgow coma scale (GCS) was 6T. Due to hypoxia, he was cannulated for VV ECMO. Hypertonic saline was given three times for elevated ICP, cerebral saturations remained acceptable and no seizures were noted on cEEG. He was decannulated on Hospital day 5, continued to improve neurologically and went to neurorehabilitation with a favorable prognosis. Neurologic complications such as bleeds, stroke, seizures, anoxic injury and brain death occur in thirteen percent of ECMO patients. Literature on neuromonitoring in ECMO patients is limited to mostly observational studies of differing modalities. Reports supporting invasive neuromonitoring of ECMO patients exist. Several noninvasive modalities can identify variables that correlate with poor outcomes. Investigators advocate the study of multimodality neuromonitoring in ECMO patients.

ICP monitoring guided the management of this patient. ECMO patients are at risk of neurologic complications and those with neurology injury do get placed on ECMO. The future may include standardize multimodality neuromonitoring for ECMO patients, however, more quality studies are needed.

COVID-19 is suspected to increase risk of stroke, possibly via coagulopathy. We present a patient infected with SARS-CoV-2 who presented with acute ischemic stroke due to dissection of the internal carotid artery (ICA) with severe subsequent complications, including coagulopathy, hemorrhage, and acute respiratory distress syndrome (ARDS).

A 58-year-old man with a history of hypertension and hyperlipidemia presented with acute onset of right hemiparesis and altered level of consciousness following fever and severe cough. The initial NIHSS score was 10. CT Head revealed early ischemic changes in the left frontal lobe and CTA showed complete left ICA occlusion. Patient was treated with intravenous tPA followed by mechanical thrombectomy under endotracheal general anesthesia. Nasopharyngeal swab for SARS-CoV-2, done on arrival at the emergency room, returned positive. Diagnostic angiography revealed tandem left ICA and MCA occlusion with "candle flame" sign, suggesting dissection, as well as irregularity of the right intracranial ICA. Balloon angioplasty and thrombectomy was performed with TICI-2b reperfusion. Subsequently, residual dissection of the ICA and subarachnoid hemorrhage were noted. Post-procedure, he developed disseminated intravascular coagulation. EVD was placed for early hydrocephalus after transfusions of FFP and cryoprecipitate. Subsequently, he developed ARDS and was treated with a course of steroids for presumed cytokine storm before developing aspiration pneumonia requiring antibiotics. The coagulopathy and ARDS gradually improved, and the ventilator and EVD were successfully weaned. Neurologically, the right hemiparesis significantly improved with residual aphasia.

The relationship between SARS-CoV-2 and ischemic stroke is not fully understood. This case suggests that vasculopathy and dissection may be associated with SARS-CoV-2. Also, coagulopathy may significantly complicate presentation of COVID-19 and possibly affect risks associated with tPA. Moreover, during endovascular procedures, special attention should be directed not only towards intravascular clotting from hyper-coagulable state but bleeding tendency in cerebral arteries.

Cerebral vasospasm is a known complication of subarachnoid hemorrhage. Risk factors include hypertension, cigarette smoking, family history, alcohol consumption and sympathomimetic drugs. We report the case of a 32-year-old male with thyrotoxicosis contributing to SAH and diffuse vasospasm causing cerebral infarction.

A 32-year-old male with no history was admitted to the hospital with left-sided weakness. CTA demonstrated right M1 occlusion with perfusion mismatch. Repeat CT demonstrated Hunt-Hess 4, modified Fisher Grade 1 SAH from ruptured right supraclinoid aneurysm. He was taken to IR and found to have diffuse vasospasm and IA verapamil was administered. He transferred to our institution for management of SAH. Lab work revealed biochemical hyperthyroidism (undetectable TSH, positive TPO antibody) consistent with Graves' disease. The patient was given SSKI and taken to IR for aneurysm securement with coils. He remained with diffuse spasm involving the right ICA, MCA and ACA, as well as moderate spasm of the basilar artery and left vertebral artery. IA verapamil, milrinone and nicardipine were administered with subjective improvement. The patient's thyroid function improved with methimazole, however, his vessels continue to spasm despite conventional medical management and he remains with poor neurological examination.

There are no documented cases of cerebral vasospasm in the presence of thyrotoxicosis. Multiple reports describe coronary vasospasm caused by thyrotoxicosis. The presumed mechanism of spasm involves decreased vasodilation, increased hypersensitivity to vasoconstrictive agents and increases in both cerebral metabolism and oxygen consumption. Hypercoagulability has also been reported. The degree of vasospasm on initial presentation without apparent SAH suggests that aneurysmal rupture may have been precipitated by cerebral vasospasm caused by thyrotoxicosis as were ongoing challenges in management of persistent spasm.

Thyroid evaluation can be useful in patients admitted with cerebrovascular disease. Thyrotoxicosis should be considered in patients with diffuse vasospasm, particularly when present on admission.

Chronic subdural hematoma (cSDH) is a common neurosurgical pathology in elderly patients due to decreasing brain volume, increasing use of anti-coagulants and anti-platelets, and increased fall risk. Traditionally, surgical treatment options include burr hole drainage or craniotomy for evacuation with or without use of a surgical drain. Due to the prevalence of cSDH, new techniques are constantly under review and include arterial embolization, use of anti-inflammatories and steroids, and bedside evacuation. We previously reported the first use of an irrigating surgical drain placed after craniotomy in the United States. We now present a novel drainage protocol which utilizes a combination of early high irrigation and a net fluid output to improve postoperative outcomes and decrease length of stay.

A 75 year-old male presented with two weeks of progressive dizziness, headache, confusion and left sided weakness. He was found to have a 25 mm thick right sided cSDH with 7 mm of right to left midline shift. The patient was taken to the OR for right sided craniotomy with placement of an IRRAflow (IRRAS, Stolkholm, Sweden) irrigating drain in the subdural space on hospital day 2.

The drain irrigated at 100 cc/hr for 23 hours with net output decreasing over time with an average of 43.6 cc/hr. Head CT showed decrease in subdural collection to 12 mm thick with trace midline shift. Irrigation was stopped for 3 hours, where net output decreased to 3 cc/hr, before being pulled on postoperative day 1 without complication. The patient was discharged to acute rehab facility on postoperative day 2. On follow-up at POD 11 he was asymptomatic and full strength.

Initial elevated irrigation and monitoring net output are safe parameters and provide a novel approach to decreasing hospital stay and recurrence of cSDH.

Central venous sinus thrombosis (CVST) is a life-threatening condition that has numerous etiologies. CVST often goes underdiagnosed when caused by trauma. Conventional guidelines recommend anticoagulation as first-line management, and endovascular intervention when the patient fails anticoagulation or has a severe clinical presentation, but literature on endovascular treatments to traumatic CVST is lacking. Traumatic CVST can also involve concurrent diagnoses such as subarachnoid hemorrhage (SAH) and subdural hemorrhage (SDH), which put the patient at high risk of worsening hemorrhage. We present a unique case of traumatic CVST with concurrent SAH and SDH that underwent a successful mechanical thrombectomy.

Patient chart and imaging were collected through the electronic medical record.

A 36-year-old man was brought in by ambulance with a 9mm SDH, small paramedian SAH, skull fracture, and CVST after hitting his head from a fall from a one-story building then subsequently jumping out of the ambulance-in-motion. Anticoagulation was not initially started due to high risk of worsening hemorrhage. Patient's Glasgow Coma Scale was initially 14, but patient had worsening mental status the next day and repeat imaging with venography revealed evolving flow defects in the confluence of sinuses, right transverse and sigmoid sinus. Urgent mechanical aspiration thrombectomy with balloon angioplasty was used to successfully restore flow to the bilateral transverse and right sigmoid sinuses, and the patient returned to baseline condition without any residual neurological deficits.

This case illustrates that endovascular intervention can be a safe and reasonable alternative to treating traumatic CVST especially in the setting of conditions contraindicating anticoagulation therapy. We call for providers and trainees to recognize CVST in head trauma and to consider endovascular intervention for cases with high bleeding risk, and for further studies to investigate the safety and efficacy of different endovascular treatments to various types of CVST by location and etiology.

Eisenhower Medical Center Department of Internal Medicine, Rancho Mirage, CA, United States

Neurogenic shock is a vasodilatory shock characterized by autonomic instability typically seen following high spinal cord injury. Non-traumatic causes are rare, making clinical recognition challenging. We report a case of acute transverse myelitis (TM) presenting as concurrent neurogenic and spinal shock in a patient ultimately diagnosed with neuromyelitis optica.

Case review of a single patient experience.

A 76-year-old male presented with a 1-week history of lower extremity weakness and numbness. There was no reported trauma or history of neurological disease. Medical history was significant for sick sinus syndrome requiring biventricular pacemaker. BP was 80/50mmHg on arrival with labile temperature (92-99.3°F). There was no bradyarrhythmia due to ventricular pacing. Neurological exam revealed a sensory level, quadriparesis, brisk deep tendon reflexes diffusely with positive Babinski and Hofmann sign; there were no ocular abnormalities. BP was unresponsive to fluids necessitating norepinephrine infusion. Spinal MRI showed edematous central cord signal abnormality extending from the level of C2-C3 to T11. MRI brain was unremarkable. CSF revealed elevated protein (134mg/dL) and IgG (13.7mg/dL) levels. Serum aquaporin-4-IgG autoantibody titers measured at 1:10,000. High-dose IV methylprednisolone was initiated, although his cardiopulmonary status continued to decline and he ultimately elected for comfort-focused care.

Acute TM is a rare cause of neurogenic shock, which should be suspected in any SCI accompanied by hemodynamic instability, regardless of trauma history. Sympathetic vasomotor tone disruption and unopposed vagal drive results in characteristic hypotension, bradycardia, and temperature dysregulation. Prompt recognition is essential in optimizing neurological outcomes and preventing cardiovascular collapse. Management is centered around hemodynamic stabilization and maintaining spinal cord perfusion. Traumatic cases often necessitate surgical decompression, although treatment of non-traumatic cases remains unclear. Ideally, rapid identification of underlying causes and targeted therapy should be initiated. However, prognosis often remains poor despite appropriate medical therapy.

External ventricular devices (EVD) are placed to relieve intracranial hypertension and remove intraventricular blood in patients with aneurysmal subarachnoid hemorrhage (aSAH). Over-drainage can cause intracranial hypotension leading to common complications, subdural hygromas, low-pressure headaches, or produce slit-like ventricles leading to brainstem compression or "brain sag." We present the case of "brain sag" caused by over drainage of cerebral spinal fluid.

A 35-year-old female with history of uncontrolled hypertension presented with dizziness, worst headache of her life, progressive lethargy, coma (Hunt Hess 4) and systolic blood pressure 250. Head CT demonstrated diffuse SAH and IVH (Modified Fisher 4) and CTA demonstrated an anterior communicating artery aneurysm. EVD was initially placed at 15 cm H2O for acute hydrocephalus and intraventricular hemorrhage. She underwent surgical clipping, which EVD was lowered to 10 cm H2O. Her neurological exam consisted of opening eyes spontaneously and following simple commands, GCS 11T. HD 7 her exam worsened, no eye opening and only a flicker of movement, concerning for vasospasm. Diagnostic angiogram showed vasospasm in bilateral ACA/MCA territories, intra-arterial verapamil administered. Her neuro exam improved to GCS 7T. Repeated angiogram showed improvement in vasospasm, but no improvement in exam.

HD 13 repeat Head CT showed slit ventricles and downward herniation. Head of bed lowered to Trendenlenburg and EVD raised to 20 cm H2O. An immediate improvement in GCS, localization to noxious stimulation. 24 to 48 hours after raising the EVD, CT head showed increase in the ventricular system. She started following commands and opening her eyes spontaneously, GCS 11T. HD 17 EVD removed and she was downgraded from the ICU.

Brain sag is a rare condition that can be fatal if not diagnosed quickly. We recommend inclusion of brain sag in the differential diagnosis for a worsening neurological exam in the setting of CSF over-drainage and slit-like ventricles.

Dural arteriovenous fistula (dAVF) is a vascular anomaly caused by an abnormal arteriovenous shunt located within the wall of dural venous sinuses. While the clinical presentations can vary depending on the location of dAVF, the most common symptoms of spinal dAVFs include progressive pain, lower extremity weakness and sensory disturbances. Area postrema syndrome (APS) is a clinical syndrome of intractable nausea, vomiting and hiccups caused by neuroinflammatory involvement of the area postrema, an emetic reflex center. It has been considered to be one of the most specific presentations of neuromyelitis optica spectrum disorders (NMOSD). Here we present a unique case of a spinal dAVF presenting with area postrema syndrome that can falsely suggest an alternative diagnosis of NMOSD.

A 45-year-old male with history of migraine and pituitary cyst presented with 6 weeks history of intractable nausea, emesis and generalized weakness. On examination, he was noted to have mild bilateral lower extremity weakness with clonus and a C3 sensory level.

MRI cervical and thoracic spine showed longitudinally extensive enhancing hyperintensities extending from the medulla to the upper cervical cord up to C5-C6 level and multiple tortuous flow voids within the thecal sac at the cervicomedullary junction, cervical and thoracic cord to T9-10 level. Diagnostic angiogram showed spinal dAVF originating from the left vertebral artery at the level of foramen magnum. He underwent successful posterior fossa craniectomy with C1 laminectomy for ligation of spinal dAVF with improvement of his symptoms.

Area postrema syndrome has not been reported as a symptom of spinal dAVF. This case demonstrates an interesting presentation of spinal dAVF with longitudinally extensive enhancing lesions from medulla to the cervical spine resulting in area postrema syndrome and myelopathy. It is important to acknowledge that in addition to NMOSD, spinal dAVFs may present with area postrema syndrome.

The Mount Sinai Hospital Department of Neurology, New York, NY, United States

Stenotrophomonas maltophilia, a Gram-negative bacillus, is a rare cause of adult bacterial meningitis and is usually associated with neurosurgical procedures and immunocompromised states. Treatment is usually complicated by its multi-drug resistance properties although most reported cases respond to trimethoprim/sulfamethoxazole. To our knowledge, there are only 2 reported cases of community acquired S. maltophilia meningitis in patients without recent neurosurgical procedures.

Data was obtained by retrospective chart review of the case.

A 30-year-old female with a history of spina bifida and congenital hydrocephalus status post ventriculoperitoneal shunt at 6 months presented with a month-long history of progressive headache and two episodes of loss of consciousness. The patient had not had revision of her shunt for several years. One month prior to presentation she started having worsening headaches and neck pain. On the day of presentation, she had an episode of loss consciousness lasting one minute. She returned to her baseline and brought herself to the emergency department. On arrival, her vital signs were normal and she was able to provide the history. However, upon laying on the stretcher she lost consciousness, became tachycardic, pupils became fixed and dilated, and had loss of brainstem reflexes. She was immediately intubated and given hyperosmolar therapy without improvement in her exam. Vancomycin and cefepime were started. CT scan showed global cerebral edema, downward herniation, and hydrocephalus. Neurological exam was consistent with brain death but she was too unstable for apnea testing. She went into cardiac arrest and could not be successfully resuscitated. CSF culture from autopsy was positive for S. maltophilia.

S. maltophilia is an extremely rare cause of adult bacterial meningitis and it should be considered in patients with remote history of instrumentation who are experiencing a new persistent headache or new neurological symptoms. If suspected, trimethoprim/sulfamethoxazole should be initiated promptly.

We present the case of a 38-year-old woman with Influenza A infection and spontaneous aortic and carotid thrombosis with resultant ischemic stroke. There are numerous reports of increased stroke and thrombotic risk after respiratory viral infections including influenza and COVID-19. However, there are few reports of spontaneous arterial thrombosis in adults with Influenza A.

The data presented in this case report was gathered by family interview, chart and imaging review, and discussion with relevant team members. An ethical standard of care was followed throughout the patient's hospitalization.

The patient had a past medical history significant only for obesity and asthma when she was admitted with hypoxemia, cough and fever and diagnosed with Influenza A. The morning after admission she developed an acute left middle cerebral artery (MCA) distribution stroke with M1 occlusion and was treated with IV tPA and mechanical thrombectomy. Within 12 hours of intervention, she developed worsening of the left MCA stroke and clinical evidence of a right hemispheric stroke as well. Vascular imaging showed spontaneous aortic and carotid thrombus. She was taken for an emergent left hemicraniectomy but declined further after developing bilateral cerebral hemispheric strokes with cerebral edema and progressed to brain death. An autopsy was performed and showed thrombus in the left common carotid and aorta without evidence of a focal or systemic vasculopathy or vasculitis. A hypercoagulable workup was unrevealing. COVID-19 PCR or antibody testing was unavailable at that time of her presentation.

We report a rare case of spontaneous aortic thrombosis and stroke in an adult with Influenza A infection. There are similar, rare reports in children but none in adults. The mechanism of thrombosis is unclear but could be related to activation of the extrinsic coagulation cascade. This case adds to the list of respiratory virus-related neurologic complications.

Cerebral venous thrombosis (CVT) has a variable presentation in terms of severity ranging from headaches to coma. The standard of treatment is heparin with relatively few reports on the efficacy of new oral anticoagulants (NOACs). We present a case of a young woman with coma from CVT with good response to NOACs after failing heparin, argatroban and enoxaparin therapy.

Data collected while patient remained in hospital neurological exam monitoring, physical/ occupational therapy reports and laboratory data

39-year-old woman on oral contraceptive pills presented with worsening headaches, left hemiparesis, altered mental status progressing over 2 days to coma requiring intubation. MRV showed extensive venous thrombosis involving superficial and deep venous systems extending into the right internal jugular vein. MRI brain revealed multifocal venous infarctions in right frontoparietal lobes in watershed territory, bilateral subarachnoid hemorrhages, and left parietal venous hemorrhage. Hypercoagulable panel revealed elevated phosphatidylserine IgM levels, antithrombin III deficiency, MTHFR heterozygous mutation, high Factor 8 level. Heparin drip was started which was switched to argatroban and then therapeutic enoxaparin with inability to reach therapeutic factor Xa levels with all three agents and no improvement in exam. On day 3, apixaban was started and by day 4 she was awake and following commands with fluctuating attention. She was extubated on day 6 however, remained abulic and had minimal verbal output. By Day 16, she was discharged to rehabilitation on apixaban.

NOACs are safe and an efficacious treatment for severe cases of CVT such as our patient. There are reports highlighting the use of NOACs in multiple cases of CVT with improvement but standard of therapy remains low-molecular/ultra-filterate heparin, especially in acute settings. NOACs can be used in severe presentations, warranting improved mortality and morbidity. Further studies and detailed trials are required for standardizing the role of NOACs for severe CVT in ICU setting.

COVID-19 has overwhelmed many healthcare systems, re-purposed neurocritical care units have been managing patients with SARS-CoV-2 infection. We describe the compassionate use of plasma exchange (PLEX) in two critically ill patients with COVID-19, managed in a national quaternary neurocritical care unit, with significant experience of PLEX.

Authors directly cared for patients during their admission. Data was collected retrospectively by interrogating the hospital electronic health record. Clinical, biochemical and imaging data are reported.

Both patients were admitted for management of acute respiratory distress syndrome and multiple organ failure. They were each supported for 34 days prior to initiation of PLEX. Patient A was a 50 year old man admitted to ICU eleven days after onset of COVID-19 symptoms. He required ventilatory and extracorporeal renal support prior to initiation of PLEX. After 10 days of PLEX, respiratory and renal function improved. He was discharged to the ward day 54, and home day 61, with no ongoing organ support. Patient B, a 64 year old man admitted to ICU seven days after onset of COVID-19 symptoms. He required ongoing ventilatory support, with failure to improve by day 34. He completed eight days of PLEX, due to superimposed infection. He is tracheostomised on a ventilatory wean and receiving treatment for critical illness-associated polymyoneuropathy.

Therapeutic plasma exchange represents a rescue therapy for either the most acutely unwell patients with COVID-19, or those whose recovery has plateaued and are experiencing prolonged ICU admission with increased risks of associated complications. PLEX resulted in a prompt reduction in inflammatory and thrombotic parameters underpinning severe COVID19 illness, with improved organ function. A randomised controlled trial of PLEX in COVID-19 has been developed.

Religious views on death by neurologic criteria (DNC) vary. We sought to evaluate how Muslim allied healthcare professionals view DNC.

We recruited participants from two listservs of Muslim American health professionals to complete an online survey questionnaire. Survey items probed views on DNC and captured professional and religious characteristics. Religiosity was determined based on the responses to five questions about religious importance, practices and motivation, each of which was assigned a point value and summed to determine the religiosity index. Respondents were dichotomized based on religiosity index and perspectives of the more and less religious respondents were compared.

There were 49 respondents (54%) in the less religious cohort and 42 (46%) in the more religious cohort. The majority of respondents (84%) believed that if the American Academy of Neurology (AAN) guidelines are followed and a person is declared brain dead, they are truly dead; there was no difference on this view based on religiosity. Less than a quarter of respondents believed that outside of organ donation, mechanical ventilation, hydration, nutrition or medications should be continued after DNC; again, there was no difference based on religiosity of the sample. Importantly, half of all respondents believed families should be able to choose whether an evaluation for DNC is performed (40% of the less religious cohort and 60% of the more religious cohort, p=0.09) and whether organ support is discontinued after DNC (49% of both cohorts, p=1).

Although the majority of allied Muslim healthcare professionals we surveyed believe DNC is death, half believe that families should be able to choose whether an evaluation for DNC is performed and whether organ support should be discontinued after DNC. This provides insight that can helpful when making medical practice policy and addressing legal controversies surrounding DNC.

Chronic liver failure patients are increasingly recognized to be at risk for developing cerebral edema and intracranial hypertension. In the largest case series to date, we document the clinical characteristics of cerebral edema in cirrhotic patients while exploring noninvasive techniques to monitor intracranial pressure (ICP).

Patients were identified using Morbidity & Mortality data (2014-2020) from the Liver Medicine Division.

All patients had acute-on-chronic liver failure (ACLF) according to the CLIF-Organ Failure system. Neurological decompensation was defined as development of focal neurological deficits or development of abnormal movements. Elevated ICP was diagnosed clinically by unresponsive pupils which improved with hyperosmolar therapy or radiologically on CT, pulsatility indices > 1.0 on transcranial dopplers (TCDs), and/or optic nerve sheath diameter (ONSD) > 0.6mm on ocular ultrasound. Arterial ammonia laboratory values were assessed when available. MELD and MELD-Na were calculated based on criteria present on the date of neurological decompensation.

Seventeen patients with ACLF were identified. Median MELD was 35 (IQR 31, 43) and MELD-Na was 36 (IQR 31.5, 43). Neurological decompensation was associated with abnormal pupil reactivity in 76% and abnormal movements in 65% of patients. Cerebral edema was diagnosed by CT imaging (n = 14), which was confirmed by TCDs (n = 3) and ONSD (n = 5). For those too ill to receive a CT (n = 3), elevated ICP was confirmed following TCDs for all three patients and ONSD for two. All documented ammonia levels (n = 6) were significantly elevated at the time of decompensation. Unfortunately, every patient died.

Cirrhotic patients with neurological decompensation often exhibit distinct clinical changes indicative of fulminant cerebral edema. Noninvasive beside techniques proved to be valuable methods to monitor ICP in these critically ill patients. In the future, patients with ACLF should be monitored closely for development of cerebral edema.

University of California, San Francisco, San Francisco, CA, United States

20% of all deaths in the United States now occur in or after admission to the Intensive Care Unit (ICU).

Checklist are now a common practice in the ICU. The process of withdrawing someone from life support is often challenging and emotionally draining, anxiety provoking for both staff and family. Our academic institution developed a checklist to ensure family center end of life care (EOLC), prevent mistakes and omissions, and ensure consistency. We hypothesized that the critical care nurse practitioner group would successfully introduce and maintain the checklist at 80% compliance for 18 months in IAP ICUs, and demonstrate compliance in non-IAP ICUs

We used Epic electronic medical record (EMR) dot phrase to ensure that a note confirming the ICU Huddle Withdrawal Checklist was used in all patients that had a withdrawal of care order. Our objective was to compare the NP group to our physician colleagues in implementing this new checklist. We understand that in our physician colleagues groups, residents were included, due to the monthly rotation of the residents, they may not have written a comfort care huddle note. We went into the EMR for every patient that had a withdrawal comfort care order set to see if we could find a supplementary notes that a comfort care huddle did indeed take place. In order for the ICU Huddle checklist to be successful it must become part of the culture.

Data collected between September 2018 and February 2020 found that the NP group had a 99% compliance using the Comfort Care Huddle Checklist. Our physician colleagues on the EPIC reported 62% compliance. After reviewing all the charts there was supplemental notes that supported a 76% compliance.

Working in the ICU is very complex, multifaceted, and requires a multidisciplinary approach. Implementing the ICU Comfort Withdrawal Huddle Checklist has improved communication within the ICU team, promote standardization of care, reduce preventable errors, relieved excessive stress and anxiety, and minimized uncertainty to achieve a smooth transition for end of life care. Our ICU NP group has successfully implemented and demonstrated compliance to the end of life checklist.

The term "brainstem death" can be used to mean loss of function of the whole brain or loss of function of the brainstem; it is also sometimes used ambiguously. We aimed to determine what is meant by this term in protocols on death by neurologic criteria from around the world.

As part of a separate project, we collected 78 unique international protocols. Of these, 8 referred to "brainstem death." We reviewed these protocols to estimate the intended meaning in each protocol that used this term.

We found that of the 8 protocols, 2 used the term "brainstem death" synonymously with loss of function of the whole brain; 3 used the term "brainstem death" to mean loss of function of the brainstem; and 3 were ambiguous and open to interpretation.

There is international variability in the meaning of the term "brainstem death," resulting in ambiguity and potentially inaccurate or inconsistent diagnosis. We advocate for uniformity in terminology when referring to death by neurologic criteria, to reduce differences in interpretation by providers, the lay public and legal powers in countries that use the term "brainstem death."

Science continues to search for a neuroprotective drug therapy to improve outcomes after cardiac arrest (CA). Glibenclamide (GBC) has shown promise in preclinical studies, but its effects on neuroprognostication tools is not well understood. We aim to investigate the effect of GBC on somatosensory evoked potential (SSEP) waveform recovery post-CA and how this relates to early prediction of functional outcome, with close attention to thalamic and brainstem functional recovery, in a rodent model of cardiac arrest.

16 male Winstar rats were subjected to 8-minute asphyxia CA and assigned to GBC (n=8) or control (n=8) groups. GBC was administered as loading dose of 10ug/kg intraperitoneally 10 minutes after return of spontaneous circulation, then maintenance dose of 1.6ug/kg every 8 hours for 24 hours. SSEPs were recorded at baseline and 150 minutes following CA. Coma recovery, arousal, and brainstem function, measured by subset of neurological deficit score (NDS), was compared between both groups. Univariate analysis compared SSEP N10 amplitude after cardiac arrest between the two groups, and ANOVA test was used to compare SSEP recovery at 30, 60, 90, and 120 minutes post-CA.

Rats treated with GBC had higher sub-NDS scores post-CA, with improved coma recovery (p=0.007). Both groups showed gradual improvement of SSEP N10 amplitude over time, from 30 minutes to 120 minutes post-CA. Rats treated with GBC showed significantly better SSEP recovery at every time point (p<0.001 for 30, 60, 90 min; p=0.003 for 120 min). In the GBC group, N10 amplitude recovered to baseline by 120 minutes post-CA.

Glibenclamide improves coma recovery, arousal and brainstem function after cardiac arrest in a rat model. GBC improves SSEP recovery post-CA, with N10 reaching baseline by 120 minutes, suggesting early electrophysiologic recovery with this treatment. This medication warrants further exploration as a potential drug therapy to improve functional outcomes in post-CA patients.

The effect of apnea testing on left ventricular function (LVF) is controversial in the literature. This study aims for evaluating the effect of hypercarbia and acidosis during apnea testing on LVF using transesophageal echocardiography (TEE).

Five consecutive patients suspected brain death and considered as potential organ donors were prospectively studied. After optimizing hemodynamic status, preoxygenation with 100% O2 was done for 10 minutes. Apnea testing was initiated and serial Arterial blood gases (ABGs) were obtained till achieving an adequate rise of PaCO2. A TEE with a 6 MHz probe was used for assessment of LVF. Fractional area changes (FAC), Left ventricular end-diastolic area (LVEDA), Left ventricular end-systolic area (LVESA), heart rate (HR), Systolic wall motion (SWM) and Mean arterial pressure (MAP) were all mesured. Serial ABGs were recorded at baseline, after 8, 10 and 12 minutes.

No spontaneous respiratory moment occurred. While PaCo2 gradual increase, a decrease in PH was noticed. Dropping in O2 concentration was kept above hypoxic level. There was a significant drop in MAP associated with increase in HR in all patients. By the time an adequate PaCo2 was achieved, FAC was increasing significantly in 3 patients.

Results of this small sample size is heterogenous. Multicenter prospective study should be conducted to better assess the effect of apnea testing-induced respiratory acidosis on FAC.

To describe practices around neurologic prognostication after resuscitation from out-of-hospital cardiac arrest (OHCA) in a metropolitan region.

This is a sub-analysis of a population-based randomized control trial assessing the effect of pre-hospital initiation of hypothermia. Subjects were adults in King County, Washington with OHCA from 2007-2012. Primary objective was to describe the incidence and characteristics associated with withdrawal of life sustaining therapy based on perceived poor neurologic prognosis (WLST-N). Hospital records were reviewed for demographics, neurologic examination findings, prognostic tests, medications, and decisions about goals of care.

This study included 1040 OHCA patients who were successfully resuscitated. Overall, mean age was 65 years (SD 16), 37% female, 41% Caucasian, and 44% had an initial rhythm of ventricular fibrillation. Patients were categorized into three groups based on awake status at 3 days: 1) awake (33%), 2) not awake and alive (27%), 3) never awoke and died (40%). Factors associated with awakening by 3 days were an initial rhythm of ventricular fibrillation 71% (p<0.05) and shorter arrest (mean 31.5 vs 35.2 minutes, p<0.05). Patients who were not awake and alive on day 3 more often received sedation and paralytics compared to the other two groups (p<0.001). Of the 413 patients (40%) who never awoke and died in <3 days, 70% of deaths were due to WLST-N. The following were associated with WLST-N <3 days: cerebral edema on head CT (45 vs 25% p<0.05), myoclonus (32 vs 12% p<0.05), and receipt of sedating medications (87 vs 74% p<0.05).

A majority of deaths occurring <3 days after OHCA were due to WLST-N, which is discordant with current guidelines recommending waiting at least 3 days prior to prognostication. Utilization and timing of prognostic tests varied widely, and sedating medications were often co-administered. Standardizing prognostication and restricting early WLST-N might improve overall outcome after OHCA.

Many cardiac arrest (CA) survivors report some form of near-death experience (NDE), but the mechanism and evolutionary purpose of this powerful phenomenon remain unclear. In our preclinical asphyxial CA model, we observed a surge in functional connectivity (coherence) in rats during nearelectrocerebral silence (denoted asphyxial segment 3; "AS3") within the first minute after asphyxia onset. Here, we characterize neurovascular coupling during AS3 via optical imaging of cerebral hemodynamics to investigate the link between AS3 and neurological recovery post-resuscitation.

Electrocorticography (ECoG) was continuously recorded in our preclinical CA model. Coherence between ECoG channels was quantified by calculating phase angle differences. Laser Speckle Imaging (LSI) and Spatial Frequency Domain Imaging (SFDI) were used to monitor cerebral blood flow, brain oxygenation (StO2), and cerebral metabolic rate of oxygen (CMRO2). Short-term neurological recovery was quantified via ECoG Information Quantity (IQ) 90 min post-resuscitation. Longer-term neurological/behavioral outcome was measured using Neurological Deficit Score (NDS) 24 hr postresuscitation.

During AS3, an ECoG coherence surge and an inflection point in CMRO2 were observed. Time to maximum surge in ECoG coherence was correlated with time to CMRO2 inflection (R = 0.76, p = 0.028). The CMRO2 inflection consistently occurred a few seconds after the coherence surge, suggesting a neurovascular coupling mechanism whereby the surge in electrical activity caused increased brain oxygen consumption. Shorter AS3 duration correlated with higher 90 min ECoG IQ (R = 0.58, p = 0.02) and 24 hr NDS (R = 0.62, p = 0.03).

The consistent occurrence of an ECoG coherence surge immediately followed by a change in CMRO2 suggests momentary enhanced neurovascular coupling during near-death. Our data presents the first demonstration that a surge in functional connectivity during impending death is linked to increased cerebral demand for oxygen, and that a more transient surge correlates with better neurological recovery.

Neurological deficits can severely impact the lives of cardiac arrest (CA) survivors. Therefore, improving neuroprotective methods for CA patients is critical for optimizing neurological recovery. Specifically, identifying physiological parameters that are predictive of neurological outcome after CA may lead to development of effective neuroprotective therapies that target those parameters.

Diffuse optical imaging technology (Laser Speckle Imaging; Spatial Frequency Domain Imaging) and electrocorticography (ECoG) were used to non-invasively and continuously measure cerebral blood flow (CBF), brain tissue oxygenation (StO2), and cerebral metabolic rate of oxygen (CMRO2) in rats during asphyxial CA and cardiopulmonary resuscitation (CPR). ECoG information quantity (IQ), an entropybased measure of neural activity, was used to quantify cerebral electrical recovery post-CPR. The relationship between baseline (pre-CA) CBF, StO2, and CMRO2 and ECoG IQ 30 min and 90 min post-CPR was assessed. Linear regressions with leave-one-out cross-validation were performed to assess accuracy of predicting ECoG IQ using the aforementioned variables.

Higher baseline CBF (r = 0.67; p<0.05) and higher baseline CMRO2 (r = 0.71; p<0.05) were significantly correlated with higher ECoG IQ within the first 90 min post-CPR. Interestingly, a greater, more rapid depletion of cerebral oxygen (i.e. fast metabolic shutdown, measured by the magnitude and slope of the StO2 drop) during the first minute of asphyxia strongly correlated with higher ECoG IQ within 90 min post-CPR (r = 0.89; p<0.05, r = 0.68; p<0.05, respectively). CMRO2 at baseline predicted ECoG IQ values 30 and 90 min post-ROSC with 11% and 25% accuracy, respectively.

In our preclinical model, we observe that baseline perfusion, oxygenation, and metabolism are strong determinants of how the brain responds to global ischemia and reperfusion. Future studies to manipulate these parameters, either at baseline (pre-CA) or during ongoing CPR, may provide insight to improve targeted neuroprotective therapies for CA.

Classic moyamoya disease is well-known in East Asian populations, however there are few longitudinal epidemiology studies in the United States. Our objective is to assess the incidence, baseline characteristics, and complications of moyamoya disease in the United States between 1993 and 2014.

We performed a retrospective population-based study on data from the Nationwide Inpatient Sample using ICD-9 code for moyamoya disease to analyze patient characteristics such as age, gender, race, mortality, comorbidities, and procedures.

From 1993 to 2014, 5868 patients were identified with moyamoya disease and the incidence increased from 0.31 to 3.62 per million person at risk (p<0.001). The age and sex-adjusted rate of moyamoya in white patients decreased from 66.4% (95% CI: 51.9% -78.3%) to 40.9% (95% CI: 35.3% -46.8%, p=0.0112). The rate increased in black patients from 5.4% (95% CI: 1.7% -15.7%) to 32.5% (95% CI: 28.1% -37.2% with p=0.0057), with no significant change in other race groups. Rate of hemorrhagic stroke in moyamoya patients decreased from 34.3% (95% CI: 24.0% -46.4%) to 9.0% (95% CI: 6.4% -12.6%, p<0.001). The rate of ischemic stroke increased from 6.8% to 16.2% though this did not reach statistical significance (p=0.135). Rate of procedures such as extracranial-intracranial (ECIC) bypass increased from 0.8% of patients (95% CI 0.1%-5.7%) to 13.8% (95% CI: 10.0%-18.8%, p=0.0095. Gender and age distribution did not change significantly. Mortality rate remained 0.6-6.3% without significant change over the study period.

From 1993 to 2014, incidence of moyamoya disease increased overall and the rate of diagnosis increased in black patients while decreasing in white patients. The rate of ischemic stroke also increased in comparison to hemorrhagic stroke among all patients, and despite increased rate of ECIC bypass, there has been no significant change in mortality.

Massage therapy is a commonly used integrative technique that has been shown to improve pain, reduce anxiety, and enhance the immune system across diverse healthcare settings and patient populations. However, the utility of massage therapy on neurocritical care (NeuroICU) patients has not been adequately studied. In a prior single-center pilot case-control study, massage therapy appeared to be safe, feasible, and subjectively effective. The present study aimed to build on these findings by using objective outcome measures to further evaluate the impact of massage therapy, with the ultimate goal of better informing a holistic treatment care plan for patients with severe neurological injury.

Twenty-one patients who received massage therapy during their stay in the NeuroICU at an academic medical center in southern California were identified and demographic and outcome data was extracted from patient charts. For each of these massage patients, retrospective matched controls were identified based on age, gender, primary neurologic diagnosis, and disease severity; outcomes data for these control patients was extracted from the electronic medical record. We utilized student's t-tests and Pearson's chi-squared tests to compare outcomes data between massage and control groups.

There were no statistical differences between the massage and control groups with respect to inhospital mortality, discharge destination, overall hospital length of stay, or incidence or duration of delirium. Despite no difference in overall LOS, NeuroICU LOS was slightly longer for massage patients than for control patients.

The findings of the present study in combination with our prior research is encouraging. They suggest that massage therapy in the NeuroICU is largely safe, feasible, and improves patient subjective experience without negative impact on objective outcomes. Massage therapy may have a valuable role as an adjunct therapy for patients with severe neurological injury.

Rotor wing air transportation (AT) is a valuable but costly resource when time-sensitive neurological interventions (TSI) are required. We sought to determine the appropriateness of AT to a neurosciences intensive care unit at a university hospital.

The air ambulance company registry at our institution was reviewed to identify all patients transferred to the neurosciences ICU during a one-year period. Demographics, diagnosis, and the performance of TSIs or complex diagnostic studies (EEG, multimodal CT/MRI) within 6 hours of admission were obtained. TSIs were defined as any of the following interventions performed within 6 hours: craniectomy/craniotomy, aneurysm coiling, aneurysm clipping, ventriculostomy placement, spinal decompressive surgery, mechanical thrombectomy, or stent placement. Fishers exact test was used for categorical variables, proportion of the means test for continuous.

A total of 106 patients underwent AT; complete data was available for 101. The median age of the patients was 63.5 (range 19-92), males comprised 42/101 (42%) and 59/101 (59%) were Caucasian. Transportation indications included intracranial hemorrhage (43%), subarachnoid hemorrhage (17%), subdural hematoma (13%), status epilepticus (11%), brain metastasis (8%), ischemic stroke (4%), and traumatic brain injury (4%). A total of 25 TSIs were performed in 23 patients; ventriculostomy placement in 16 (16%), subdural drain placement in 5/101(5%), and craniotomy in 4/101 (4%) patients. Subarachnoid hemorrhage was associated with TSI (p< 0.008). Advanced imaging was performed in 77/101 (77%) patients. Transports of less than 75 miles were performed in 27% of patients; TSIs were done in 22% of such patients. Cost of AT was $6,721,146; there was no statistical difference in the cost of transporting TSI vs. non-TSI patients.

The majority of patients using AT do not undergo TSIs but undergo complex diagnostic studies. Subarachnoid hemorrhage is associated with the need for TSI. Short transfers, equivalent to ground transport, were not associated with TSIs.

Neurocritical care nurses are an integral part to the Neurocritical Care team. They face many challenges in the daily delivery of high quality intensive care to their patients. Specialty-specific nursing societies, like the NeuroCritical Care Society (NCS), can offer support and education to help nurses confront these difficulties. We conducted a survey to better understand these challenges and to identify barriers to nursing involvement in the NCS.

A multi-answer survey was created and analyzed through Survey Monkey™. We collected responses to gather nursing specific demographic data, current work environment, training, organizational involvement and future educational plans. The survey was dispersed via multiple associates involved in Neurocritical care.

Out of 112 responses from multiple regions in the United States, (86.6%) of the respondents were female, mostly 20-30 years old (39.2%). 92.8% work in a Comprehensive stroke center with a designated Neuro Critical Care unit with 8-24 Neuro ICU beds (68.7%) staffed by Neuro Intensivitsts (89.2%). While 62.5% of nurses responded they get compensation for additional certifications and training courses, the vast majority report not being a Certified Neurosciences Registered Nurse (CNRN) or a member of the American Association of Neuroscience Nurses (AANN) (68.7%). Surprisingly, 61.9% of responders were unaware they could become members of the NCS; still, 77.1% said they were unwilling to pay the annual membership fee and would prefer a lower fee. Nonetheless, many of the respondents expressed interest in the services provided by the NCS (76.1%).

Based on this survey, the primary barriers to involvement in the NCS seem to be a lack of awareness of the organization in general along with an unacceptably high membership fee.

Some critically-ill patients with the novel 2019 coronavirus infection (COVID-19) remain comatose for prolonged duration following sedation discontinuation. It is unclear if this represents prolonged encephalopathy or brain injury despite the absence of documented sustained hypoxia/ hypotension.

Review of medical records. We describe the clinical course, imaging and outcome of two critically-ill COVID-19 patients who remained comatose >7 days following sedation discontinuation.

A 59-year-old man and a 53-year-old man, both with history of hypertension and neurologically intact on admission, developed worsening COVID-19 associated acute respiratory distress syndrome (ARDS). Both required benzodiazepine, opioid, neuromuscular blockade, therapeutic anticoagulation and vasopressor infusions in addition to renal replacement therapy. Echocardiography demonstrated normal chamber size and systolic function in both cases. Both patients demonstrated only trace flexion to pain 10 days following discontinuation of all sedation. Magnetic Resonance Imaging (MRI) on both demonstrated multifocal Diffusion Weighted Imaging (DWI) lesions with apparent diffusion coefficient (ADC) correlate in bilateral middle/ anterior cerebral artery (MCA-ACA) borderzones. In both patients, continuous electroencephalography demonstrated no seizures and carotid ultrasound was unremarkable. The 59-year-old man remained comatose, with only flexion to pain, 60 days following initial presentation. The 53-year-old man demonstrated neurological improvement starting 40 days from initial presentation, and by 50 days was alert and conversing with family, without significant motor deficit. Neither patient had any documented period of sustained hypotension (mean arterial pressure <60mmHg) or hypoxia (SpO2<90%).

Critically-ill COVID-19 patients with prolonged coma following discontinuation of sedation may demonstrate imaging features of ischemic injury in borderzone regions despite the absence of documented sustained hypotension or hypoxia. However, excellent neurological recovery is possible despite these findings.

Decision aids (DAs) for goals-of-care (GOC) decisions in severe acute brain injury (SABI) patients provide a critical opportunity to improve surrogate decision-making. We sought to determine feasibility of a novel paper-based GOC-DA for surrogates of critically-ill SABI patients.

We conducted a two-center, parallel-arm randomized-controlled feasibility trial enrolling surrogates of critically-ill SABI patients (ICH, TBI, or hemispheric acute-ischemic stroke requiring tracheostomy and/or long-term feeding-tube for survival) between 2/2018-12/2019. Surrogates were randomized 1:1 at the patient-level >1day(s) before the GOC family-meeting to intervention (DA, including values-clarificationexercise worksheet) or control (no DA). Primary feasibility outcomes were participant screening, recruitment, participation, and retention. Secondary outcomes included surrogates' decisional-conflict, decision-confidence, decision-regret, psychological burden, assessed with validated scales.

We screened 1233 SABI patients; 74(6%) were eligible; we approached the surrogates of 58(78%) eligible patients. We consented 66/95 surrogates (70%); n=33 were randomized to each study arm. Baseline characteristics were balanced except for more ICU-complications in the control group (p=0.03).

In the intervention arm, 27 surrogates remained in the study at the family-meeting-time-point; 19 had read the DA; 15 had read and completed the DA (56% participation). Study retention rates at 3-months were 23/33(70%) for intervention and 22/33(67%) for control. There were no between-group differences in surrogates' decisional-conflict and decision-confidence. Mean Hospital-Anxiety-and-Depression-Scale (HADS) scores were high in both groups at baseline (I:C.18±7 vs.17±7.5;p=0.82), remained high after the family-meeting (I:C.18±8.8 vs.18±7.9;p=0.81), but declined in the control group 

GOC-DA use is feasible in neuroICUs. Our study's secondary outcomes may potentially be explained by the small sample-size with imbalance in patients' ICU complications, parallel-arm design limitations (control-group contamination), and low protocol fidelity. Our study offers important "lessons-learned" for future neuroICU DA studies.

The safety of deep nasopharyngeal swab (NPS) testing for COVID-19 in ruptured intracranial aneurysms and other brain-injured patients at risk for raised intracranial pressure (ICP) and herniation is unknown.

Literature review and descriptive small case series. We searched National Library of Medicine for various search terms "intracranial pressure," "nasopharyngeal", "swab," "COVID-19" as well as Google and Google Scholar. We additionally reviewed the last two months of patient data from 3 academic Neuro Critical Care Units (NCCU).

There were zero articles in the literature clarifying safety of NPS testing in this population. Separately we found one article on COVID-19 in the New England Journal of Medicine that cautioned this testing in those with recent nasal surgery or deviated septum [1]. Over 3 academic NCCUs, we observed 138 cases of aSAH, AIS, ICH and seizure cases at risk for neurological complications that underwent COVID-19 testing via NPS over the period of 4/1/2020 -5/26/2020. One aSAH case in the emergency department had increasing expansion of hemorrhage from initial CT to CTA and GCS decrease from 13 to 8 requiring intubation. COVID-19 testing was negative in that case and felt unrelated to timing of testing. Of 138 patients tested, with many receiving multiple tests, there were no incidences of epistaxis, ICP crisis or neurological worsening believed to be related to NPS testing. Further study is needed to support these findings. Testing with the use of oral secretions or serum IgG testing may overall be safer within this specialty population [2] .

There is no nasopharyngeal swab safety data in at risk neurocritical care populations. Alternate means of testing may decrease the risk for raised intracranial pressure, re-rupture of aneurysms, and herniation within this group, however as of this time, there were no observed complications to NPS testing within 3 NCCUs.

The coronavirus disease 2019 (COVID-19) pandemic has strained health care resources and personal protective equipment (PPE) supplies globally. We hypothesized that a collaborative robot system could reduce both PPE use and exposure to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) if feasibility could be proven in a simulated setting of selected common health care tasks that normally require entry into an intensive care unit (ICU) room.

A prospective proof-of-concept feasibility and design pilot study tested 5 discrete medical tasks in a simulated ICU room of a COVID-19 patient through use of a collaborative robot: push button on intravenous pole machine when alert occurs for downstream occlusion, adjust ventilator knob, push button on ICU monitor to silence false alerts, increase oxygen flow on wall-mounted flow meter to allow the patient to walk to the bathroom and back (dial-up and dial-down oxygen flow), and push wallmounted nurse call button. Feasibility was defined as task completion robotically.

A training period of 45 minutes to 1 hour was needed to program the system de novo for each task. In less than 30 days, the team completed 5 simple effector task experiments robotically, which reduced PPE use and worker exposure to SARS-CoV-2. Selected collaborative robotic effector tasks appear feasible in a simulated ICU room of the COVID-19 patient.

Theoretically, a collaborative robotic approach can be used reduce PPE use and staff SARS-CoV-2exposure and requires worker learning similar to other ICU device training.

Refractory intracranial hypertension is strongly associated with mortality. Current guidelines recommend medical management involving sedation, hyperosmotic agents, barbiturates, hypothermia, paralysis, and surgical intervention. When these interventions are maximized, refractory intracranial hypertension poses risk for brainstem herniation and death. We describe a novel intervention of verticalization for lowering refractory intracranial hypertension.

Single center retrospective review of five cases of refractory intracranial hypertension in a tertiary care center. All patients were treated with maximal standard-of-care for lowering intracranial pressure (ICP) and, despite these interventions, maintained persistent ICP > 20cm H2O. They were then treated with verticalization. Using paired t-tests, we compared hourly ICP, number of ICP spikes >20cm H2O, and percent of ICP >20cm H20 in the 24 hours prior to verticalization vs. 24 hours post-verticalization. We also assessed effects on the use of hyperosmotic therapies, MAP, and CPP.

Four patients were admitted with SAH, one with ICH. Admission Glasgow Coma Scale ranged 3T-4T, and imaging from all patients revealed cisternal effacement and intraventricular hemorrhage. All patients had ICP monitoring by external ventricular drain. Mean opening pressure was 25cmH20. All patients received maximal medical therapies for intracranial hypertension, including barbiturate coma. All patients showed reduction in ICP after verticalization, with a significant decrease in overall mean ICP (x cm H20 vs. y cm H2o, p<0.001), number of ICP spikes (x spikes vs. y spikes, p<0.01), and percent of ICPs >20cmH2O (x% vs. y%, p<0.01) after intervention. There was no difference in number of doses of hyperosmolar therapies, propofol infusion rate, MAP, or CPP after intervention. The most common adverse events were bradycardia (n=3; 60%) and pressure wounds (n=3, 60%). Survival to hospital discharge was 40%, with discharge modified Rankin Scale 5 in both surviving patients

Verticalization is an effective, non-invasive intervention for lowering intracranial pressure in refractory intracranial hypertension warranting further study.

Many patients admitted to the neurocritical unit (NCCU) are in a non-responsive state. Numerous reports have been published on the use of neurostimulant medications (NSMs) to help improve consciousness in brain injured patients. Most of these reports have been on chronic vegetative or minimally conscious state patients. There is limited prospective data on the use of NSMs in the acute setting.

We prospectively collected data on patients admitted to Thomas Jefferson University Hospital's NCCU from 2/2019 to 3/2020. We collected demographic data, GCS on day one of starting NSMs and three days after, medications, infections and creatinine. Modafinil or Adderall were started on stabilized patients with persistent GCS ≤ 9. A positive response was defined as an improvement in GCS by 2 points within 3 days.

73 patients received NSMs (17 Adderall and 58 Modafinil). Patients admitted with stroke were unlikely to benefit from NSMs (p < 0.001). There was no significant relationship between NSMs and improvement in GCS in patients admitted with seizures, anoxic brain injury, brain trauma, tumors or CNS infections. There was no significant relationship between starting NSMs and improvement in GCS in patients with encephalopathy, interictal brain activity (GPDs or LPDs), systemic infections or on antibiotics or AEDs. Adderall was discontinued in 5 patients (4 tachycardia, 1 hypertension). Modafinil was stopped in 3 patients (2 seizures and 1 tachycardia).

Using NSMs for three days was insufficient to improve GCS in our patients. Longer durations of NSM use may be provide benefit but further research is needed.

The purpose of this study is to explore gender differences in attitudes toward end-of-life (EOL) care and life support among medical students. Studies have indicated that gender influences EOL decision making and care. It has been suggested that individuals in the medical profession are more likely to decline prolonged life support for themselves in the setting of incurable or debilitating disease. However, little has been published qualifying this notion and even rarer are studies analyzing medical students' preferences regarding life-sustaining treatment based on gender.

A 66-item questionnaire was distributed among medical students during their neurology clerkship at a large U.S. medical school. The differences between genders were examined by Chi Square test. SAS 9.4 was used.

There were 303 responses received over a 22-month period, of which 165 were male students and 138 were female students. Females were more likely to have had discussions about advanced directives with their family members regarding their own health preferences (males, 33%; females 49%, p<0.01) and their family members' health preferences (males, 50%; females 64%, p=0.01) compared to males. There were slight gender differences in wanting to continue with life support in the following scenarios: Coma with a small chance of full recovery (males, 41.8%; females, 29.9%, p=0.18), Persistent vegetative state (males, 7.2%; females, 2.2%, p=0.17), Painful and debilitating but not life-threatening disease (males, 78.8%; females, 72.3%, p=0.14), and Not-painful but incurable disease with life expectancy 3 months or less (males, 62.4%; females, 56.2%, p=0.4).

Female students were more likely to have discussed advanced directives with their families. While males had a consistent trend towards preferring continuation of life-sustaining measures compared to their female counterparts, this difference was not statistically significant. Recognizing gender differences in future physicians' attitudes toward EOL care may help current healthcare providers in education of advanced directives and EOL curriculum.

Medical-grade ultrasound devices are now pocket-sized and can be easily transported to underserved parts of the world, allowing health care providers to have the tools to optimize diagnoses, inform management plans, and improve patient outcomes in remote locations. However, broader awareness of the impact of these mobile health technologies is needed among health care providers, along with training in how to use them in valid and reproducible environments for accurate diagnosis and treatment. The word "resource" appears frequently in the definition of health, which implies that health is composed of resources to deal with physical, mental, and social issues.

We sought to use the resources of the Mayo International Health Program (MIHP) to travel to Bwindi, Uganda, in November 2019. This article focuses on the use of a portable "mobile health unit" (MHU) as an added, relatively low-cost resource for the Bwindi Community Hospital and its potential impact on the global health value equation in terms of diagnostic and management decision-making. The MHU was comprised of 2 portable ultrasound units, donated in-kind by FUJIFILM Sonosite, Inc, a digital stethoscope with 1 lead electrocardiogram donated in-kind by Eko Devices Inc (compatible with iPhone [Apple Inc]), and an otoscope with video.

Among the patients seen in Bwindi, we performed approximately 46 point-of-care ultrasound (POCUS) examinations in 33 patients during a period of 23 days (about 2 examinations per day). Lung, cardiac, optic nerve, vascular, and abdominal ultrasounds were performed according to clinical relevance. POCUS took, on average, 10 to 20 minutes, depending on amount of ultrasound clarity and anatomical difficulties. In 15 of 18 (83% (15/18) of the documented cases, use of these bedside diagnostic tools enhanced the formation of diagnoses, and thereby positively impacted the management and outcomes of these patients. In 4 of the 18 (22% (4/18) the initial diagnosis was confirmed and management approach remained unchanged.

While each health care community should evaluate the value of these technologies in their own context, the technologies we describe are safe and non-invasive and can be used in conjunction with physical examination skills, are relatively simple to operate.

Anecdotal evidence seems to demonstrate increased agitation in critically ill patients with COVID-19 and acute respiratory distress syndrome (ARDS) requiring mechanical ventilation. This may be associated with higher doses and increased number of analgosedative medications.

We conducted a retrospective cohort study comparing analgosedation in critically ill patients with ARDS caused by COVID-19 (admitted January-May 2020) with a historical cohort of ARDS patients (admitted May-December 2019). Patients were included if they were mechanically ventilated, had ARDS defined as bilateral infiltrates on chest imaging, PaO2/FiO2 ≤200 mmHg, PEEP ≥5 cm H2O, and received cisatracurium for at least 6 hours. A comprehensive list of analgosedatives was compared between cohorts during the time each was receiving cisatracurium. Wilcoxon rank sum and Poisson tests were used to compare daily analgosedative doses and sedative counts, respectively. Dose comparisons were conducted only in patients receiving each specific agent.

Patients had a median age of 51 years, 78% were white, and 60% male. Those with COVID-19 and ARDS (n = 10), when compared with a historical cohort with ARDS (n = 18) received higher median daily doses of fentanyl (2725 mcg [8 of 10 patients] vs 1501 mcg [15 of 18 patients], P = 0.03). There was no difference between cohorts in daily dose of propofol, dexmedetomidine, ketamine, and midazolam. The median number of continuous infusion (3.5 vs 2, P = 0.09), enteral (2 vs. 0.5, P = 0.002), and total (5.5 vs. 2.5, P = 0.002) analgosedatives received during neuromuscular blockade was higher in the COVID cohort, though only the latter two met statistical significance.

When compared with a historical cohort, patients with COVID-19 ARDS received significantly higher doses of fentanyl, and an increased number of enteral and total analgosedative medications. Future studies should compare varying analgosedative regimens and dosing strategies in patients with COVID-19 and ARDS.

Traditional pathways for advanced neuroimaging such as magnetic resonance (MRI) require patient transportation to centralized hospital imaging suites. During the COVID-19 pandemic, critically ill patients have had limited neuroimaging because of infection control and safety concerns. Many patients are sedated or paralyzed, precluding the routine assessment of neurological status. In this context, we report the acquisition of advanced neuroimaging in patients with severe COVID-19 using a low-field MRI device at the bedside.

A 64mT point-of-care (POC) MRI was used to acquire neuroimaging in COVID-19 intensive care units at Yale New Haven Hospital from April 2020 through May 2020. COVID-19 patients with altered levels of consciousness were scanned if patients were clinically stable and had no MRI contraindications. Exams were acquired at the bedside using a standard 110V, 15A power outlet. Hospital rooms included equipment such as vital signs monitors, ventilators, dialysis machines, and intravenous infusion pumps. Image acquisition was obtained by trained research staff, without the need for an MRI technician. POC MRI exams were interpreted by two board-certified physicians (one neuroradiologist and one neurologist). The device was cleaned with a hospital-approved disinfectant following each scan per radiology guidelines for infection control.

POC MRI exams were obtained on 19 patients (16% female, ages 42-74 years). GCS and RASS at time of scan ranged from 3 to 14 (median 7) and 0 to -5 (median -3), respectively. T1-weighted (T1W), T2weighted (T2W), fluid-attenuated inversion recovery (FLAIR), and diffusion-weighted imaging (DWI) sequences were obtained for all patients. POC MRI exams demonstrated intracranial pathology in 37% of patients: intracranial hemorrhage (n=1), cerebral infarction (n=2), diffuse cerebral edema (n=1), and leukoencephalopathy (n=3).

We report the first experience of advanced neuroimaging using low-field, POC MRI at the bedside during a pandemic. This approach may hold promise for bedside assessment of neurological injury in other resource-limited settings.

Critically ill patients with Coronavirus Disease 2019 (COVID-19) often exhibit markers of hypercoagulability and develop thrombotic complications. As a result, empiric anticoagulation has been advocated for these patients, but there are few data on its efficacy and a great deal of variation among providers. We therefore examined the association between empiric anticoagulation and outcomes among COVID-19 patients at two hospitals in New York City.

We performed a retrospective cohort study of adult patients admitted with PCR-confirmed COVID-19 to intensive care units (ICU) at two New York City Hospitals from March 1, 2020 to April 19, 2020. Our exposure variable was empiric therapeutic-dose anticoagulation (i.e., no traditional indication such as confirmed venous thromboembolism) with intravenous heparin or subcutaneous enoxaparin. The primary outcome was a composite of death or arterial or venous thrombosis. We used marginal structural models adjusted for age, sex, SOFA score, and Charlson-comorbidities. Patients were considered at risk until death, hospital discharge, or 28 days after admission, whichever occurred first.

Among 1,530 patients (626 female, mean age 64 years), 102 (6.7%) received empiric therapeutic-dose anticoagulation at some point during their ICU stay. Patients who received empiric anticoagulation had a higher baseline (9.5 vs 3.7 p<0.001) and peak sofa score, (11.8 vs 4.7 p<0.001) and were more likely to be male. We found no significant evidence of an association between empiric anticoagulation and death or thrombotic events (HR, 0.7; 95% CI, 0.3-1.8), death alone (HR, 0.4; 95% CI, 0.1-1.2), or thrombotic events alone (HR, 1.5; 95% CI, 0.5-4.5).

Empiric anticoagulation in COVID-19 ICU patients was not significantly associated with improved survival or a decreased risk of thrombotic events. Based on our confidence intervals, our study was relatively underpowered, but given the scarcity of data on the prevention of thrombotic complications in COVID-19, our findings may be of interest to intensivists.

With advances in neuro science there is also increasing utilization of intrathecal medication administration such as nicardipine and tPA via external ventricular drains (EVDs). The practice of medication administration via EVDs are being more wildly used within neurocritical care settings. The most common technique for these injections tends to remain a multi puncturing process into the EVD tubing and thereby increase the opportunity for cerebral spine fluid (csf) infection. We describe our success with a single injection process compared with the traditional multi-puncture approach.

While maintaining a sterile field and using a 2 stopcock closed system with 3 sterile syringes and 1 blunt tip cannula we demonstrated one can easily aspirate, inject, and flush while only puncturing the closed system one time. With a single injection using the pre-constructed stopcock system we aspirate CSF to equal the total volume of injectable medication plus 2 ml flush thus maintaining isovolemic fluid balance.

Due to the simplicity and reproducibility the 2 stopcock approach has become the standard of care for EVD medication injection in our intensive care unit. With over 400 injections we have had no associated adverse reactions. Furthermore our EVD infection rates continue to remain below the national average at less than 1% per year.

While multifactorial, we believe this method attributes in part to our lower than the national average EVD infection rates. This method decreases the chance for contamination and the risk for cerebral spinal fluid infection by simply reducing the frequency of access. It a simple reproducible technique that can safely be performed by either physician or advance practice provide.

The SARS-CoV-2 pandemic has affected health care systems worldwide in a multitude of ways, many of which are just now becoming evident. Early studies during the COVID-19 pandemic have shown a decrease in the number of hospitalized acute ischemic stroke (AIS) patients compared to historical averages, with patient/caregiver reluctance to seek care hypothesized as a leading driver of this decrease. However, no studies have been done looking at the effect of the pandemic on overall stroke and neurocritical care unit admissions.

We retrospectively reviewed the number of patients transferred via an active Neuroemergency transfer program to a 28-bed NSICU at an academic tertiary care center. Data was compared between a prepandemic March-April 2019 and pandemic March-April 2020 epoch. The number of transfers, discharge diagnoses and mortality was compared between epochs utilizing Fisher's exact test. Patients were divided into stroke [AIS, intracranial hemorrhage (ICH) and aneurysmal subarachnoid hemorrhage (SAH)] and non-stroke sub-groups.

There was a decrease in patient volume of 50.7% between the pre-pandemic and pandemic epochs, with an absolute decline of 43.5% in stroke admissions. The decrease was proportional across stroke sub-groups ( 2= 2.9334, p-value= 0.23) and between stroke and non-stroke diagnoses (47.2% of admissions in 2019 vs 54.2% in 2020, P= 0.0967). There was no statistical difference in mortality.

A proportional decrease was seen in the number of patients presenting during the COVID-19 pandemic across both stroke and non-stroke subgroups. The decrease was more pronounced in the April subgroup, correlating with increased prevalence of COVID-19 in the community. This is suggestive that the pandemic is a contributing factor to the decline in ICU transfers. The etiology of this decrease remains unclear, as mortality and clinical severity did not differ significantly between epochs. Further studies on the effect of COVID-19 on presentations to care for neurologic emergencies are warranted.

Clinical practice guidelines (CPGs) and consensus statements are published and distributed to guide healthcare decisions. Women authorship representation in neurological CPGs is low. There is a paucity of literature on trends of women authorship representation in Neurocritical Care Society (NCS). Our primary aim is to determine the pattern of authorship by gender in published NCS guidelines and consensus statements and evaluate the trend through 10 years since the first NCS sponsored guideline.

This study is a retrospective analysis of readily available online material. We included all guidelines and consensus statements primarily sponsored by the NCS for the years 2011-2020. For each guideline, we gathered author gender characteristics and publication year. Utilizing descriptive statistics, we reported the median number of authors by gender. We also determined if there was a significant change in the proportion of women authors through the years 2011-2020.

A total of 12 guidelines with a total of 170 authors were published and sponsored primarily by the NCS were included from 2011 to 2020. Only 31% (n=52) of the authors were women. The median proportion (IQR) of women authors was significantly lower than men (34.8, 0-87.5 vs 65.2, 12.5-100; p<0.03). There was a trend towards improvement in women representation through 2011 to 2020, however it was not statistically significant.

Women representation in authorship guidelines in the NCS is significantly low with no significant change in the trend during the study years. However, improved adherence to NCS bylaws and policies is allowing for improved gender representation in several aspects of guideline development and dissemination. The information in this study may be used to further improve on current guideline authorship diversity processes in the NCS.

COVID-19, being a landmark disease in modern medicine, is changing current ICU practices. More than ever, ICU providers need to be effective and efficient, while minimizing potential exposures. The ABCDEF bundle (AFB) is a standardized bundle of best practices aimed to decrease harmful symptoms in ICU patients. By analyzing AFB utilization and patient outcomes before COVID-19 and during the COVID-19 era, we attempt to find trends in ICU care and outcomes in times of a global pandemic.

An academic NSICU implemented the AFB in March 2019 after systematic multidisciplinary training. Baseline data, total bundle compliance, and individual element compliance was measured and analyzed using previously verified algorithms. Patient outcomes included ventilator duration, delirium, length of stay (LOS), and mortality. We compared data prior to March 2020 (pre-COVID) to March -May 2020 data (COVID era) in a PUI-accepting NSICU.

Overall bundle compliance was 34.8% in the COVID era compared to 30.1% pre-COVID, a statistically significant improvement (p<0.05). Individually, family engagement decreased during COVID-19 (82.4% vs 68.7%) likely due to limited visitations, while delirium assessment improved in compliance (61.7% vs 90.1%). Outcomes, such as total vent days, ICU LOS, delirium free days, and mortality, showed statistically insignificant increases in the COVID era. Longer LOS is possibly linked to delayed COVID results in the early weeks of testing.

The COVID era necessitates efficient utilization of time spent in patient rooms in order to minimize unnecessary exposure. Although the AFB requires detailed assessments of clinical cues, which can be especially challenging to implement during times of new care routines, our data demonstrates AFB's sustainability during the COVID era. As pandemic conditions continue, more longitudinal data needs to be collected with the inclusion of COVID units to best formulate ICU practices that are pandemic sensitive whilst maintaining quality of care.

Emerging evidence suggests COVID-19 patients may be at increased risk for intracranial vasculopathy and microvascular dysregeulation. Hemodynamic monitoring with Transcranial Doppler (TCD) may be a useful tool in unconscious COVID19 who cannot otherwise be clinically assessed. Here we evaluate a preliminary population of COVID patients using a fully automated TCD system, minimizing exposure to medical personnel.

Fourteen patients with COVID-induced acute respiratory distress syndrome were screened at Mount Sinai Hospital. Each subject's left/right middle cerebral arteries were continuously monitored at depths spanning 45-65 mm for up to 30 minutes. Continuous cerebral blood flow velocity envelopes were processed offline to identify individual heart beats, and reject outliers contaminated with movement artifact. Each set of 20 temporally adjacent beats were aligned and averaged to obtain average waveforms for each interval, from which Pulsatility Index (PI) and Velocity Curvature Index (VCI) were computed. PI and VCI were averaged across waveforms and compared against those observed for a previously acquired population of 52 presumed COVID-negative in-hospital controls (IHC).

Nonparametric bootstrap tests revealed VCI to be significantly smaller for COVID relative to IHC subjects (d2=1.41, p<<0.001), with confidence intervals on the means of (2.98, 4.72), and (5.54, 6.14) for COVID and IHC subjects, respectively. For PI, observed differences did not reach significance (d2=.49, p=0.07), with confidence intervals on the means of (0.81, 1.22), and (0.80, 0.90) for COVID and IHC subjects, respectively. PI and VCI distributions among those who survived versus died were not significantly different.

VCI captures morphological information present in the TCD waveforms of COVID patients not as readily discernible using traditional measures such as PI. Future work is needed to determine the degree to which the observed morphologies are specific to COVID relative to other currently uncontrolled factors such as positive pressure ventilation.

Patients with traumatic brain injury (TBI) are at high risk for agitated delirium and behavioral outbursts placing themselves and staff at risk of harm. One "traditional" approach to addressing these issues includes restraints and observational "sitters". However, this care model comes with known negative effects: immobility and increased costs, length of stay and adverse outcomes (e.g. pressure wounds). Additionally, sitters lack specialized training limiting them in their ability to contribute to therapeutic recovery. Therefore, we provided targeted skill development to sitters in the trauma progressive care unit (PCU), reclassified their position to Rehabilitation Care Partners (RCP), and evaluated the effect of this RCP program on outcomes.

RCPs received 24hr of total training from: 1) psychiatry, 2) physical, occupational, and speech therapy, and 3) the Crisis Prevention Institute to include de-escalation, safe mobilization, and dysphagia management. Beginning in March 2019, RCPs were deployed to support TBI patients requiring 1:1 observation. For this project, we compared outcomes-restraint use, adverse events, sitter hours, length of stay (LOS), and security activations-from the first six months of our RCP model by reviewing medical records and incident logs in patients and comparing with a matched patient population from the pre-implementation "traditional care model" period.

Average restraint hours for TBI PCU patients decreased from 63.2 to 13.7 hours (p=0.005); mostrestrictive restraint type was abated. Average indwelling catheter days and hospital-associated pressure injuries significantly decreased. Average sitter hours and average LOS both decreased with a trend towards significance. Security incident calls and alarm activations decreased by 46% and 85%, respectively.

Trauma PCU RCPs have led to significant improvements in quality and safety outcomes for patients with TBI, as well as for our staff. Further analysis will examine RCP impacts on psychotropic medication trends and healthcare cost reductions.

Cerebral venous sinus thrombosis (CVST) is an increasingly recognized complication of traumatic brain injury (TBI). However, management of trauma-related CVST remains controversial, with uncertainty as to the type of anticoagulation to use, or whether it should be used at all. This study looks at the outcomes for TBI patients treated with Aspirin (ASA) versus conservative management.

Retrospective study of patients admitted to the Neurosurgical intensive care unit (NSICU) with diagnosis of TBI and related CVST. Comparison was made between patients that received ASA 81mg and patients who did not receive any antithrombotic. Primary outcome measured was development of venous infarction. Secondary outcomes included new intracranial hemorrhage (ICH) in the post-acute phase, or worsening of the initial ICH, with any subsequent need for surgery, withholding of ASA, reversal agents or other measures to control intracranial pressure (ICP).

14 patients were admitted between April 2017 and April 2020 with trauma-related CVST. 4 were started on ASA 81mg at the time of diagnosis. Primary Outcome: No venous infarcts observed in either patient group. Secondary outcomes: There were no new ICHs seen in either patient group; two patients developed worsening of their initial ICH, one of which required DDAVP. These patients did not receive ASA.

The use of ASA as a treatment for trauma-related CVST was not associated with the development of new ICH, worsening of initial ICH or need for further surgical intervention. However, ASA also did not have any impact on development of venous infarcts, and similar outcomes were observed in both patient groups. Further study is required to evaluate the benefit of ASA in trauma-related CVST.

Traumatic brain injury (TBI) is frequently accompanied by disorders of coagulation, and is associated with poor outcomes. The traditional tools to diagnose coagulopathy include Conventional Coagulation Assays (CCA) which often yield delayed results and can have limited predictive value. Viscoelastic hemostatic assay (VHA) testing, including thromboelastography (TEG) and rotational thromboelastography (ROTEM), provides a useful alternative for identifying coagulopathy and can potentially improve outcomes in TBI patients. We performed a systematic review to evaluate these assays and their impact in the acute management of TBI patients.

A systematic review was performed in accordance with guidelines for the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) to identify studies comparing VHA to CCA in adult TBI patients.

746 article were initially identified after applying search criteria, of which 10 were included for final analysis. With regard to clinical outcomes, 4 studies provided some metric of mortality, 2 studies evaluated the need for neurosurgical intervention, 2 evaluated the progression of TBI, and 2 reported other clinical outcomes. Among the studies, there was variability in the differential utility of VHA compared to CCA. The abnormal VHA parameters R-time, MA, and K value were associated with increased mortality in certain studies but not all studies. An abnormal R-time was the only VHA parameter found to be associated with the need for neurosurgical intervention, but this was only found in one study. An abnormal R time was also the only VHA parameter associated with progression of traumatic intracranial hemorrhage (tICH). Overall, many studies also demonstrated abnormal CCAs, mainly activated partial thromboplastin time (aPPT) to be associated with poor outcomes.

There is no definitive evidence of the comparatively greater utility of VHA over CCA in determining rates of neurosurgical intervention, tICH progression or mortality in TBI patients.

Current traumatic brain injury (TBI) guidelines provide a level III recommendation for resuscitation of hypotensive TBI patients with crystalloid fluids to mean arterial pressure (MAP) of 85-110mmHg based on predominantly class-3 evidence. We previously reported that whole blood (WB) outperforms Lactated Ringer's (LR) in supporting MAP and preventing cerebral hypoxia in a murine model of combined TBI and hemorrhagic shock (HS) at MAPs <70mmHg. Functional correlates of these MAP and cerebral oxygenation improvements remain unestablished.

Anesthetized mice (n=40) underwent controlled cortical impact (5m/s to a depth of 1.2mm) followed by HS (MAP=25-27mmHg). Mice were randomized to five groups for a 90 min "prehospital" resuscitation: LR with MAP target of 60mmHg (LR60), WB60, LR70, WB70, and monitored sham. Mice were bolused 20mL/kg of autologous WB or LR, followed by LR boluses (10ml/kg) every 5 min for MAP below target. Shed blood was reinfused after 90 min. Mice started Morris Water Maze (MWM) testing 14 days after injury. Comparisons were made by repeated measures ANOVA.

Average prehospital MAP in each group was sham: 79±2mmHg, LR60: 57±2mmHg, WB60: 69±3mmHg, LR70: 59±4mmHg, WB70: 72±3mmHg (p<0.05, all comparisons). MWM latency to find the hidden platform was different vs. sham for both LR60 (p<0.002) and WB70 (p<0.007) but not LR70 or WB60. In the probe trial, only WB60 did not significantly differ from sham.

WB resuscitation to a MAP target of 60mmHg produced the greatest mitigation of behavioral deficits after TBI+HS vs. LR at MAP targeting 60 or 70mmHg or WB targeting 70mmHg. Our findings support TBI Guidelines recommending a higher MAP target in LR resuscitation. However, contrary to current guidelines, our findings suggest that WB outperforms LR in resuscitation of TBI+HS and that MAP thresholds lower than those required for crystalloid resuscitation may be tolerated and/or optimal with WB resuscitation.

Advances in MRI technology have led to the development of a low-field (64mT), point-of-care (POC) MRI scanner. The appearance of subdural hematoma (SDH) on POC MRI has not yet been described.

We studied four male patients (ages 65-90) presenting with a radiographic diagnosis of SDH. One patient presented with bilateral SDH, two patients presented with bilateral SDH and underwent left-sided SDH evacuation, and one patient presented with left-sided SDH and underwent left-sided evacuation. T1weighted (T1W), T2-weighted (T2W), and fluid-attenuated inversion recovery (FLAIR) fast-spin-echo POC exams were obtained on all patients. POC MRI exams were analyzed by three raters using a standardized qualitative evaluation for subdural collection appearance, location, and morphology. The three evacuated subdural spaces were evaluated for postoperative changes. Appearance of subdural collections were determined by majority consensus.

POC exams were obtained at approximately six days (n=1), one month (n=1), and two months (n=2) since time of incident. SDH collections were classified as chronic (n=1) and acute-on-chronic (n=3) based on conventional imaging. T1W POC exams demonstrated one chronic SDH collection as a mixed isohypointensity, two acute-on-chronic collections as mixed iso-hypointensities, and one acute-on-chronic collection as an isointensity. T2W and FLAIR POC exams demonstrated one chronic SDH collection as hyperintense, two acute-on-chronic collections as mixed hyperintensities, and one acute-on-chronic collection as a hyperintensity. Pneumocephalus appeared as signal fallout at the site of SDH evacuation on one T1W, two T2W, and two FLAIR exams. Residual blood products were detected at the site of SDH evacuation as mixed iso-hypointensities on two T1W exams, hyperintensities on two T2W exams, and hyperintensities on two FLAIR exams.

These preliminary data detail the appearance of SDH and its postoperative changes on low-field, POC MRI. Further work is needed to detail the serial changes that accompany SDH progression and its surgical intervention.

Penetrating spinal trauma, albeit more common in military settings, has been well reported in civilians. It remains with poorly defined prognostic and management paradigms. Increasing gunshot injuries in civilians has emphasized the need to address this disease state, especially given the associated functional and socioeconomic sequela. We systematically review the available literature on acute management of penetrating spine injury in the civilian population (cvPSI).

We address six questions relevant to the acute management of cvPSI: spinal immobilization, steroid use, infection rate and antibiotic prophylaxis, surgical management, blood pressure goals, and vascular complications. We searched PubMed, Scopus and Cochrane (1979-2019).PRISMA guidelines were followed. Newcastle-Ottawa Scale was used for quality assessment of individual studies and risk of bias was judged using the RTI item bank.

Our literature research identified 1917 papers, of which 53 were analyzed. The rate of spinal instability requiring surgical stabilization following cvPSI is very low. Spinal immobilization might be overutilized without clear evidence and, at times, added risk. There is no evidence that steroid use in cvPSI improves mortality or functional outcome. The rate of spinal and paraspinal infections ranges between 2-49%, with osteomyelitis being the most common infection, and thoracic spine the most involved segment.

There is no evidence that antibiotic prophylaxis is necessary. Surgical decision-making varies and is largely influenced by operator's assessment of spinal column stability and salvageability. MAP goals above 85 were not shown to improve outcome. Use of vasopressors to achieve these goals may be associated with cardiac complications. The rate of vascular injury ranged from 8% in lumbar injury, to 80% in cervical injury.

We reviewed the literature and summarized recommendations in management of cvPBI. The field lacks high quality studies leaving significant knowledge gaps in the management and prognosis. RCTs are needed to better address cvPSI.

Paroxysmal sympathetic hyperactivity(PSH) occurs days to weeks after traumatic brain injury(TBI) and may represent a modifiable risk factor for poor outcome. We hypothesize that, as a quantitative measure of autonomic nervous system function, heart rate variability (HRV) may add to previously described clinical predictors to help identify latent PSH after TBI.

Adult TBI patients admitted to a neurotrauma ICU with continuous vital sign data recorded for at least 14 days were identified from a prospective trauma registry. PSH cases were identified based treatment with bromocriptine/propranolol within the first three weeks of injury and were matched (1:1) with controls by age and post-resuscitation GCS. Electrocardiograph data was assessed for standard HRV parameters over 5-minute epochs for up to 21 days. A generalized estimating equation(GEE) using autoregressive covariation was used to identify clinical and HRV features during the first 36 hours of admission that were associated with PSH, with a significance threshold adjusted to p<0.01 for multiple comparisons. Outcomes measures included discharge GCS, Rancho Los Amigos Score(RLAS), and modified Rankin Score(mRS).

78 PSH cases were matched to 78 TBI control patients with median(IQR) GCS 7(5-8) and mean(SD) age 40 (14). Clinical diagnosis of PSH occurred on mean(SD) post-injury day 7(4). Features independently predictive of PSH by GEE included blunt injury, non-caucasian race, higher post-resuscitation systolic blood pressure, CT presence of DAI, cisternal compression, IVH/SAH, and absence of contusion, and HRV features including lower low frequency to high frequency(LF/HF) ratio and lower detrended fluctuation analysis(DFA) alpha1. All measured outcomes were worse in cases compared to controls despite equivalent admission injury severity.

Quantitative measurement of autonomic nervous system function using HRV derived from early continuous vital sign data after acute TBI may aid in the identification of PSH prior to its clinical diagnosis.

Socioeconomic factors, such as insurance status, have been shown to affect outcomes for patients following traumatic injuries. Traumatic Brain Injury(TBI) patients with either Medicare(MCare) or Medicaid(MCaid) have worse outcomes compared with private insurance(PI) patients. Dual-Eligible beneficiaries (DEB) receive both Medicare(MCare) and Medicaid(MCaid) and constitute an especially vulnerable population. There is no data addressing whether DEB display unique characteristics and outcomes compared to TBI patients with MCare, MCaid, or private insurance.

We conducted a retrospective analysis of 10-years of National Inpatient Sample(NIS) data. Using ICD-9-CM codes, we identified adult patients with known insurance status who were hospitalized for intracranial hemorrhage due to TBI; epidural, subdural, subarachnoid, and intracerebral hemorrhages were included. Patient characteristics and surgical intervention type, if any, were collected. Statistical analysis was performed on SAS using chi-square and T-test, with multivariate logistic regression analysis used to control for gender, age, and race. The primary clinical outcomes of interest included complications, length of stay (LOS), mortality, and discharge disposition.

Among 78,511 patients, DEB(10.1%) were significantly older(average age=73yrs) than MCaid(46yrs), PI(50yrs), or Self-Pay (SP)(42yrs) and more likely female (DEB=59%; MCare=50%; MCaid=34%; PI=38%; SP=21%;p<0.01). Caucasian race was highest among MCare patients(66%). African-American race was highest among MCaid(20%). Among all patients, 3.4% underwent operative intervention, with no difference across insurance groups. LOS was longest in MCaid(11.4d) and shortest in SP patients(7.0d).

Overall, DEB patients were least likely discharged home(35%) and most likely to require skilled nursing facility(49.5%). Mortality was highest among DEB(11.2%) and MCare(12.7%) compared with MCaid(8.9%), PI(8.8%) and SP(9.5%),p<0.05 across groups. Adjusting for confounders, mortality was lower for SP(OR=0.85[95%CI=0.7-0.98]) and PI(OR=0.9[95%CI=0.82-0.97] compared to DEB patients.

DEB are a vulnerable population and health disparities continue to manifestly affect outcomes following neurosurgical emergencies. Adverse outcomes among DEB highlight the need to uncover and address unknown sources of disparities.

Faculty of Medicine, University of Toronto, Toronto, Ontario

Patients with acute traumatic cervical/high thoracic level spinal cord injury (SCI) typically experience severe respiratory complications, necessitating the need for mechanical ventilation (MV). Tracheostomy is preferred in situations where prolonged weaning from MV is predicted. However, it is an invasive procedure with its attendant risks. Randomized controlled trials in general critical care populations found no benefit to early tracheostomy, however these trials did not include acute traumatic SCI. Furthermore, published guidelines for SCI respiratory management focuses on the non-acute care setting providing no consensus recommendations on optimal timing of tracheostomy. We therefore systematically reviewed studies comparing early tracheostomy to late tracheostomy or prolonged intubation in SCI patients to determine effects on mortality.

Studies were identified through a search of databases (MEDLINE, EMBASE, CINAHL, Scopus, Web of Science, CENTRAL, Google Scholar), reference lists of retrieved articles, and abstracts of conference proceedings. Study titles were used for screening, and the full text of the remaining articles were then be examined using pre-determined inclusion/exclusion criteria. Data was obtained by two independent reviewers to ensure accuracy and completeness.

Twenty studies (4,596 patients) met selection criteria, 15 of which were published in the last decade. The majority of studies were single centre retrospective cohort studies. The mean time for early tracheostomy was 6 days across seven studies, and late tracheostomy was performed after 16 days. The remaining studies provided ranges for early and late tracheostomy, most frequently ≤7 days and >7 days respectively. High clinical heterogeneity precluded a meta-analysis. On qualitative analysis, early tracheostomy is not associated with lower ICU and long-term mortality.

Current studies highlight the lack of clinical evidence to guide the optimal timing of tracheostomy in acute traumatic SCI. Future research should seek to understand whether early tracheostomy improves patient comfort, decreases duration of sedation and mechanical ventilation and improves long-term outcomes.

Subdural hematomas (SDHs) are an increasingly common, highly morbid condition. Ischemic brain injury is known to complicate SDHs and leads to worse outcomes. Prior work with radionuclide imaging suggested that SDH-related ischemia primarily results from herniating brain tissue compressing the anterior and posterior cerebral arteries (ACAs/PCAs). No prior work has attempted to characterize SDHrelated ischemia with magnetic resonance imaging (MRI).

We identified SDH patients who received an MRI within 2 weeks of presentation to our institution from 2015-2019. MRIs were reviewed for evidence of acute ischemia on diffusion-weighted imaging. To limit our scope to SDH-related ischemia, we excluded cases that met any of the following criteria: (1) alternative explanation for ischemia (e.g. cardioembolism, dissection, or large-vessel vasospasm); (2) diffusion restriction occurs in the region of significant intraparenchymal/subarachnoid hemorrhage; (3) diffusion restriction occurs in a pattern consistent with diffuse axonal injury.

Out of 2100 SDH patients, 213 patients received an MRI. 16 patients (8% of MRIs) met our criteria for SDH-related ischemia. SDH-related ischemia was found primarily in patients with larger volume SDHs (median 73mL, IQR 50-108mLs) with associated midline shift (median 7mm, IQR 6-15mm). 9 patients had infarcts in the ACA/PCA territories; the remaining 7 patients had a distinct pattern, with predominantly cortical restricted diffusion adjacent to the SDH. The patients with the ACA/PCA ischemia pattern typically presented comatose with a median Glasgow Coma Scale (GCS) score of 6. The peri-SDH ischemia patients often presented with a higher GCS (median 14), but then developed focal deficits in a delayed fashion (5/7 patients developed hemiparesis/aphasia).

SDH-related ischemia is commonly found on MRI. Our work suggests multiple mechanisms account for SDH-related ischemia, including herniation in some cases and potentially a direct toxic effect of the SDH in others. Further research on how SDHs affect the underlying cortex is needed.

Cerebral venous thrombosis (CVT) is a medical condition due to various causes. One of these mechanisms that rarely occur is trauma. Despite numerous reasons, treatment options have been adherent to the conventional approach as those whom we treat in those caused by non-traumatic. The current initial recommended treatment is anticoagulation, and if refractory thrombectomy can be considered. However, trauma-induced CVT (t-CVT) remains challenging as they are associated with preexisting substantial hematoma burden; that treating them with anticoagulation therapy, in theory, may increase the risks of bleeding and ultimately leads to further complication. We performed a retrospect study investigating actual clinical practice and treatment variation among TBI patients.

This is a retrospective study comprised of multiple DHS centers in the Los Angeles county area who admits trauma patients from 2015 to 2020. Charts were reviewed.

We identified a total of 9640 cases identified to have TBI; 37 patients were confirmed t-CVT. There is an overwhelming predominance of male gender (97%). Treatments used in actual practices are also very variable ranging from non-treatment (n=12), aspirin (n=7), Lovenox/heparin (n=10), to new oral anticoagulation (n=8). On subgroup analysis, the non-treatment group demonstrates higher mortality than the treated group (p<0.037). When compared t-CVT to a matched non-CVT TBI patients, there is no statistical significance in mRS (p=0.49), GCS (p=0.63), length of stay (p=0.999), and mortality (p=0.249). Median time in days to start therapy is also different -aspirin (med. = 4), Lovenox/heparin (med.= 8.5), and NOAC (med.= 4).

There are no current guidelines or clinical trials to address the efficacy and treatment of choice in this unique population. From this series, T-CVT may not be a poor prognosticate factor in TBI patients. This clinical dilemma of substantial hematoma burden amongst TBI patients with CVT deserves a closer investigation with a randomized clinical trial of anticoagulation therapy to elucidate safety and appropriate treatment timing.

Eligible patients often fail to receive treatment with intravenous tissue plasminogen activator (tPA) or endovascular therapy. Multidisciplinary acute stroke teams improve acute ischemic stroke management but may hinder trainees' education, which in turn may contribute to poorer outcomes in community hospitals upon graduation. Our goal was to assess individual trainee performance during an acute stroke simulation independent from the stroke team.

In this prospective, observational, single-center simulation-based study, trainees ranging from subinterns to fellows managed a patient with acute ischemic stroke followed by tPA-related hemorrhagic conversion leading to cerebral herniation. Critical actions were developed by a modified Delphi approach and based on the Neurocritical Care Society's Emergency Neurological Life Support (ENLS) protocols. The primary outcome measure was the critical action item sum score. We sought validity evidence to support our findings by comparing novice, intermediate, and advanced trainees' performance using ANOVA.

Thirty-five trainees completed the simulation. The mean sum of critical actions completed by trainees was 11.5/22 (52%). Fifty-one percent of trainees properly administered tPA, 54% immediately stopped tPA infusion following patient deterioration, and only 20% reversed tPA according to guidelines. There was significant effect of training level on critical action sum score (novice mean score [standard deviation (SD)] = 8.2 (2.7) vs. intermediate mean score (SD) = 12.8 (2.8) vs. advanced mean score (SD) 16.8 (3.8), p < .001). The sum scores were positively correlated with multiple choice pre-test scores, r = 0.599, p < 0.01, and acute ischemic stroke experience, r = 0.417, p < 0.05.

Non-adherence to acute ischemic stroke guidelines and errors in the treatment of hemorrhagic transformation after tPA are frequent. Additional training is necessary to prepare trainees for independent practice at hospitals lacking multidisciplinary stroke teams. High-fidelity simulation holds promise as an assessment tool for acute stroke management performance.

Through mutual interest and collaboration, the neuro-critical care and the palliative care teams at a major academic medical center decided to pilot a quality improvement initiative for patients in a 40 bed neuro-ICU. The goal was to increase upstream palliative care inpatient consultations in neuro-critical care patients, emphasizing proactive goals of care discussions and disposition planning for patients and families.

We developed a dynamic process where the neuro-critical care team could send electronic request for consultation to the palliative care team through a shared patient care list in the EMR. The neuro-critical care team would add patients to the shared list in the EMR as they were rounding, which would be checked by the palliative care team and patients would be seen the same day, or in the next 24-48 hours based on the urgency of the request marked as, "stat", "urgent" or "routine". The palliative care provider picking up the consult would call back the neuro-ICU team with updates and care plan recommendations after seeing the patient and continue to follow along closely for support and goals of care discussions regularly.

We observed an increase in upstream palliative care consultation going from prior consult rates of 7% per month, to 83%. On an average, the palliative team received consultation requests by day 2 to day four of the hospitalization in the ICU and active goals of care discussions from day 4 through day 7 resulting in shared decision making and disposition planning. This resulted in significant decrease in length of stay (p value 0.003).

Palliative care provides value based care through its impact on quality and cost. With upstream palliative care consultation for critically ill patients in the neuro-ICU, we were able to impact staffing, resource utilization and nursing education through this project.

Catheter-associated urinary tract infection (CAUTI) accounts for > 560,000 nosocomial infections annually (Gould,2016). The neurocritical care (NCC) population is at risk for CAUTI development related to neurogenic bladder, urinary retention and need for indwelling catheterization. Measures to mitigate and eliminate CAUTIs in the NeuroICU were needed. After correcting catheter utilization and practice related causes, our interdisciplinary team identified inappropriate culturing practices as a remaining cause for CAUTI, where colonization versus true infection was suspected. We developed a pilot program of urine culture stewardship to address this issue.

Our team developed a urine culture (UCx) algorithm to guide the clinician with use of urinalysis (UA) and/or UCx testing for suspected infection in NCC patients. The final decision-making on UCx was left to the discretion of the critical care attending based on pre-defined eligibility criteria. The effectiveness of this algorithm was tested as a unit-based UCx pilot between November 1, 2019-March 1, 2020. Results of UA and UCx were descriptively analyzed and calculated comparing pre-and post-intervention periods.

A total of 49 specimens were evaluated and included UA with/without reflex to culture and/or UCx without UA. 35% (17/49) of UAs were positive. 82% (41/50) of the initially ordered UCxs were cancelled due to ineligibility.22% (11/49) potential CAUTIs were avoided. These patients were identified as clinically asymptomatic for infection and/or suspected colonization of urinary tract but met NHSN criteria (fever >38.1, foley use within infection window) and were later determined to have different etiology of their symptomatology. CAUTI rate was zero during this period.

Implementing an interdisciplinary approach to urinary culturing practices identified non-indicated ordering of UCx in patients with possibly colonized urinary catheters. This significantly improved identification of true infection versus colonization and reduced occurrence of what would have been deemed a CAUTI in absence of true UTI.

In the general Intensive care unit (ICU) setting, incorporation of nurse practitioners (NPs) is shown to improve clinical and fiscal outcomes, decrease length of stay (LOS), improve practice guideline compliance, and improve patient satisfaction. However, the impact of a subspecialty trained, nurse practitioner service on physician/nursing satisfaction and patient outcomes is understudied in the Neurological ICU (NeuroICU).

In summer 2019, we established a primary NeuroICU service staffed by subspecialty-trained NPs at one of two urban academic medical centers within our health system; NPs provide 24/7 care under the supervision of a Board-certified neurointensivist. One year later, we 1) surveyed our NeuroICU bedside critical care nurses (RNs) using Likert methodology to evaluate subjective satisfaction measures, 2) asked our NPs and attending providers to give anonymous pro/con feedback about the service which was subsequently reviewed and categorized, and 3) compared common outcomes measures (mortality, observed/expected mortality, ICU LOS) for this NP service with Vizient metrics.

33 RNs completed the survey (response rate 40.2%). Of respondents, >90% either agreed or strongly agreed that they 1) like that NPs are included in inpatient care of ICU patients, and 2) valued the RN-NP relationship. As perceived NP approachability increased, so did confidence in NP management (R=0.91, p<0.001). Self-reported comfort in managing NeuroICU patients increased as comfort with NP management (R=0.88, P<0.001) increased. Fourteen providers (100%) provided qualitative feedback. Qualitative analysis revealed that the most common "pro" categories included: continuity of care, efficiency, NP fulfilment; the most common "con" categories included NP training variability and fiscal effects. Objectively, ICU LOS, mortality, and observed/expected mortality were superior to the Vizient comparative.

An innovative model of NeuroICU staffing that uses NPs to provide 24/7 care is well-received by RNs, physicians, and NPs, and maintained above-benchmark outcomes when compared with other models of care.

The importance of neurocritical care (NCC) has been recognized. However, to our knowledge, no dedicated educational systems have been established in Japan. We conducted an educational NCC hands-on (NCCHO) seminar and investigated the effects.

This prospective, before-after study used a questionnaire to examine the effect of the NCCHO seminar held in 2018. The learning concept was to identify the methods for maintaining cerebral oxygen balance to prevent secondary brain injury in three skill sessions: intracranial pressure monitoring, trans-cranial color flow image, and EEG, and four scenario sessions: post-cardiac arrest syndrome, subarachnoid hemorrhage, traumatic brain injury, and non-traumatic acute weakness. The primary outcome was the improvement in the examination scores post seminar. The secondary outcomes were increase in the degrees of satisfaction and confidence of the participants. We evaluated the outcomes using Wilcoxon signed rank test. A p-value < 0.05 was considered significant.

Total 156 medical personnel participated in these seminars. We excluded 11 (7.1%) participants who did not complete the seminar or did not submit their questionnaires or examinations. We analyzed 145 (92.9%) participants, including 121 (83.4%) physicians from the field of emergency or intensive care medicine and 21 (14.5%) other medicals professionals. Their median age was the 30s (IQR: 20-30) with median intensive care medicine experience of 3 years (IQR: 1.0-6.0). The median scores in the examination, improved significantly from 21 (IQR: 19-23) to 26 (IQR: 24-27) after the seminar (p < 0.001). Total satisfied participants were 137 (94.5%). A higher number of professionals reported increase confidence levels, i.e., from 1 (6.9%) before the seminar to 64 (44.1%) after the seminar (p < 0.001).

Our seminar improved the NCC knowledge of medical personnel and made them more confident.

University of North Carolina Department of Neurology, Chapel Hill, NC, United States

Hospice provides value-based end-of-life care, yet it is often underutilized, particularly in the inpatient setting. Patients with severe neurologic injury whose preferences and values align with comfort-focused care (CFC) may be appropriate for hospice on discharge. Prominent barriers to hospice referral include limited access, lack of physician training and comfort with end-of-life communication, and operational inefficiencies in transitions of care. We implemented a quality improvement initiative to improve rates of successful hospice referral for eligible NSICU patients.

We convened an interprofessional team of palliative care physicians, performance improvement specialists, NSICU physicians, nurses, and care managers. We utilized Lean methodology to map existent processes and identify targets for improvement. In order to identify targets for improvement, standardized root cause analyses (RCA) were performed on select cases deemed potentially eligible for hospice transfer. Iterative process improvements and educational campaigns were implemented beginning in March 2020 to maximize hospice referral ease and efficiency for frontline clinicians. Successful hospice referral rates per month were evaluated utilizing statistical process control charts.

Thirty-five RCAs were performed. The highest frequency barriers to hospice referral included issues related to information sharing (31%), hospice referral processes (34%), hospice availability (17%), and other (22%). Hospice referral rates significantly increased from a baseline of 23% in the initial preintervention period (July-September 2019) to 57% post-intervention (February -April 2020). Sustained special cause variation demonstrates steady increases in hospice referral rates attributable to nonrandom process improvements.

Hospice care for CFC patients has been previously underutilized in our NSICU, largely driven by insufficient information-sharing regarding hospice and referral process inefficiencies. Improving access to hospice for eligible patients may be meaningful in optimizing end-of-life care, patient-family experience, and hospital resource utilization.

Two neurointensivists in different institutions are currently developing integrated Neurocritical care programs. We sought to identify key metrics and processes to generate data essential for program development.

Manual chart review and consolidation of data from two existing ICU's including admission diagnoses, volume, provider specialty, attending of record, and CPT codes.

(A) We identified the following processes needed for development: 1) Characterize current patient population of the NeuroICU including acuity and percentage of neuroscience and non-neuroscience patients. 2) Characterize baseline staffing models including intensity of ICU provider staffing and primary team attribution. 3) Generate phased program development strategies informed by baseline conditions including presence of existing intensivist programs, advanced practice providers, and physician trainees. (B) Example data of baseline conditions. 1) Hospital 1: 299 NeuroICU patient-days in a sequential 30-day period were analyzed. Based on clinical effort in 32.6% the neurointensivist was the primary provider while 167 patient-days (67.3%) required co-management between neurointensivist and neurosurgeon. Based on diagnosis, physical location of care, and attending-of-record designation, attribution was 78.2% to neurosurgery, 21.7% to a Non-neurocritical intensivist, and 0% to the neurointensivist. Hospital 2: 685 Neuro-ICU patient-days in a non-sequential 20-day period were analyzed. Critical care coverage by 2 separate teams was provided to ~ 60% of the patients to a population consisting of 85% neuroscience patients. Clinical effort attribution to neurosurgery was 60-75% for the neuroscience population. However, primary team attribution in 75% of patients is to a non-neuroscience hospitalist team.

There are many elements to consider when starting a NeuroICU program including the ideal staffing model. Those decisions can be best informed when clinical effort data is accurately captured. Commonly-used metrics for administrative decision-making may not accurately characterize that effort. Clinicians should be familiar with their individual metrics and characteristics necessary to build their programs.

During the COVID-19 pandemic, many healthcare providers were redeployed to intensive care units without critical care training. Having practiced in New York City, the epicenter of the pandemic, we combined our experiences and current data to develop a curriculum for these providers. In anticipation of another wave, we believe this curriculum will increase confidence and clinical knowledge as well as decrease stress level in caring for critically-ill COVID-19 patients.

From May 22 -May 25, 2020, twenty-two health care providers voluntarily participated in a 2.5 hour curriculum consisting of nine system-based lectures. Identical pre-and post-tests assessed selfconfidence, stress level and clinical knowledge regarding care for these critically ill patients. All survey answers were de-identified. We used a paired t-test to analyze the change in confidence, stress level and clinical knowledge.

The majority of providers (13, 59%) had not spent time as a primary ICU provider. 11.7% of the time providers reported feeling confident in the various aspects of care for a critically ill COVID-19 patient. After completing our curriculum, providers reported feeling confident 73.9% of the time. Providers reported feeling "stressed" 79.9% of the time prior to the curriculum and 37.5% of the time after. Test takers significantly improved on the clinical vignette multiple choice portion where the average test score was 61.9% (SD=18.29) before the curriculum compared to 80.1% (SD=18.36) after, t(42)=3.3, p<0.05.

These preliminary results demonstrate a statistically significant improvement in knowledge base and trends toward increased confidence and decreased stress level in providers caring for critically-ill COVID-19 patients. This tool may effectively prepare non-ICU trained providers for future waves of COVID-19 infection.

Early readmissions after transfer from higher to lower level of care units are associated with increased resource utilization and worse clinical outcomes. Little is known about in-hospital readmissions of acute care neurology patients. We evaluated acute care neurology patient characteristics and system factors associated with early readmission (< 72 hours) to the neurocritical care unit (NCCU) and intermediate stepdown care unit (ISCU) following transfer to lower level of care units at an academic hospital.

We performed a retrospective analysis of acute care neurology patients readmitted to higher level of care units (NCCU or ISCU) within 72 hours of transfer to a lower level of care unit between January 2015 and May 2017. Data collected included readmissions during "off hours" (weekend or overnight), critical care interventions administered within 2 days prior to transfer, and primary readmission reason.

92 readmission events occurred in 90 patients. Average time to readmission was 29.5 hours (24% ≤12 hours; 26% >12 to ≤24 hours; 50% >24 to ≤72 hours). Most readmissions occurred during "off hours" (27% weekend; 43% overnight). Primary readmission reasons included: respiratory distress (44%), neurological decline (25%), hemodynamic instability (9%), sepsis (5%), MI (2%), cardiac arrest (1%) and other (14%). 44% of readmitted patients were diagnosed with a comorbidity (infection, seizure, cardiopulmonary complication, renal failure/insufficiency) during their index higher level of care stay; 74% received a critical care intervention within 2 days prior to transfer. Average length of stay in the higher level of care unit after readmission was 142 hours (SD + 147 hours).

The majority of readmissions occurred during "off hours." Respiratory distress and declining neurological status accounted for most readmissions. While not all early readmissions are preventable, understanding modifiable system factors that may predispose patients to greater risk for readmission may shed light on effective strategies for readmission prevention.

Inter-professional collaboration and physician-nursing teamwork is crucial to providing high quality of care to critically ill patients. Implementing an interdisciplinary change poses challenge. Methods used to successfully implement interdisciplinary afternoon nurse-led rounds in a Neuroscience ICU were explored.

A change project was implemented using the Plan Do Study Act model (PDSA) for improvement. Using responses from an initial survey to assess nurses' satisfaction with the current process, perceived value of adding nurse-led afternoon rounds and nurses' feelings of empowerment, our interdisciplinary team created a structured afternoon rounds to ensure plans outlined on morning rounds were enacted. To provide a dedicated "space" to empower nursing, afternoon rounds were designed to be led by nursing. Individual education as well as 1 month of mock rounds preceded the implementation of afternoon rounds. Several PDSA cycles were completed to optimize the workflow and improve nurse participation. An initial and two follow-up surveys to were administered to eligible nursing staff. Their responses were anonymous.

41% of eligible staff participated in the initial pre-implementation survey. At 4-and 8-months post implementation, follow-up surveys were completed by 37% and 67% of eligible staff, respectively. At both follow-up timepoints, approximately 93% of nurses reported finding positive value in afternoon rounds. Pre-implementation, only 35% of nurses reported thorough follow-up of plans initiated on morning rounds; at 8-months post-implementation 78% agreed. At 8-months, 63% of nurses reported a positive impact on their practice.

Nursing is empowered to promote change within their discipline but feel limited in the degree of change they can enact without the support of their physician partners. Physician-nursing teamwork is crucial to providing high quality care of critically ill patients and has been associated with improved patient outcomes. Collaboration between physicians and nursing facilitated the development and successful implementation of nurse-led afternoon rounds in a Neuroscience ICU.

The ABCDEF Bundle (AFB) is an evidence-based organizational tool aimed at minimizing the physical and mental sequelae of ICU care. Best practices incorporated in the AFB address elements of immobility, pain, agitation, and delirium, all of which are particularly challenging to identify and treat in neurologically injured ICU patients. While the AFB has demonstrated improvement in patient outcomes, its impact on neurocritically ill patients has yet to be elucidated. This study assesses AFB's feasibility and effect on quality metrics in the Neurosciences ICU (NSICU).

Systematic training on the AFB was performed across inter-professional teams and implemented at an academic NSICU. Baseline data, total bundle compliance, and individual element compliance was measured and analyzed using calculations from the SCCM ICU liberation campaign. Patient outcomes included ventilator hours, delirium, length of stay (LOS), and mortality. We compared data from March 2018 -February 2019 (pre-AFB) to March 2019 -February 2020 (post-AFB) to evaluate AFB's impact in the NSICU.

Overall bundle compliance rates increased to 30.1% post-AFB compared to only 3.4% pre-AFB (p< 0.001). AFB implementation resulted in a 9.5% reduction in the number of ventilator hours (118.6 vs 107.3, p< 0.05) and a 12.8% increase in the number of delirium free days (72.6% vs 85.4%, p< 0.0001). While there was a statistically significant reduction in total ICU LOS from 5.13 days to 4.70 days (adjusted for by acuity, p< 0.05), there was no demonstrable change in mortality rates (5% vs 5%).

Systematic implementation of the AFB in the NSICU led to statistically significant improvements in quality of care by reducing ventilator hours, delirium days, and LOS. Although there was no effect on mortality, our submaximal total bundle compliance rates show that further studies are needed to evaluate the limitations and challenges associated with executing the AFB in the NSICU population.

Case-mix index (CMI) is a quality metric that measures complications and comorbidities of a hospital's inpatient population and is used in allocation of financial resources. This project aimed to utilize Clinical Documentation Integrity (CDI) specialists and providers in a focused capacity to improve documentation accuracy and better capture quality metrics in the Neuro ICU.

Education was given to Neuro ICU providers on proper documentation required to capture specific patient diagnoses present on admission. This included education on diagnoses related to the 2018 AMC Risk Model's Top 10 Mortality Risk Variables for Neurology/Neurosurgery, such as breaking down the Glasgow Coma Scale into individual components. Two CDI specialists were assigned to review medical records from the Neuro ICU, a change from nine CDI specialists who originally reviewed these patients along with patients on other units. Total queries, provider response rate (measured by the number of 'Agreed and Documented' queries), and CMI were calculated to compare the effects of the interventions.

Total queries increased 2-fold. Provider response rate increased by 1.4%. Analysis of CMI showed that before interventions, CMI increased by 0.1685 as a result of CDI queries. After interventions, queries sent by CDI increased CMI by 0.2559. This is a 1.5-fold increase in CMI.

The increase in total queries combined with increase in provider response rate shows that providers remained engaged and involved despite being queried more. The CMI increase shows that illness severity was found to be higher with the more specific documentation identified by the CDI specialistprovider interaction. Focused CDI specialists allowed for a more efficient documentation review specific to the common errors made in that unit. Continued CDI and unit-based physician relationships are anticipated to sustain documentation improvements as providers will understand the 'why' behind queries sent to capture quality metrics.

Andexanet alfa is a specific factor Xa reversal agent used to treat uncontrolled or life-threatening bleeding. Effectiveness of andexanet alfa was assessed in managing intracerebral hemorrhage (ICH) in oral factor Xa inhibitor (oFXaI) recipients.

We included consecutive patients admitted to Hartford Healthcare from 8/7/18 to 12/7/19 with computed tomography (CT)-confirmed ICH, presenting <24 hours from their last dose of oFXaI and treated with andexanet alfa. We assessed hemostatic effectiveness (Sarode et al. criteria); median (25%, 75%) change in hematoma volume (initial-to-closest to 12-hour CT, but within 3-24 hours postandexanet alfa infusion); and the need for additional hemostatic treatments. CT results were independently adjudicated by two investigators.

Twelve ICH patients (median age 86 years, 66.7% men, 66.7% received apixaban, 33.3% rivaroxaban and 91.7% had atrial fibrillation) were included. Nine (75%) of ICHs were spontaneous and median baseline hematoma volume was 13.2 mL (7.8, 29.5). Median time since last oFXaI dose-to-arrival was 7 (4, 20) hours and arrival-to-andexanet alfa administration was 3 (2, 4) hours. Eight patients received low-and four high-dose andexanet alfa. Ten patients (83.3%) achieved excellent and 2 (16.7%) poor hemostatic effectiveness. Median hematoma volume reduced by -0.08 mL (-3.1, 0.8). Two patients received hemostatic management <48 hours after andexanet alfa infusion (one underwent craniotomy after repeat CT and received 4-factor prothrombin complex concentrate (4F-PCC); one received 4F-PCC). No additional hemostatic management was required.

In oFXaI patients experiencing ICH, andexanet alfa resulted in a high rate of hemostatic effectiveness and decreased hematoma volumes.

Coagulation Factor Xa (recombinant) (Andexxa®) was approved as a specific reversal agent against direct-acting oral factor Xa inhibitors (DOACs) for life-threatening bleeding, based on a single-arm prospective open-label study measuring hemostatic efficacy. Prior to Andexxa® availability, 4F-PCC was used at our institution for DOAC associated intracranial hemorrhage (ICH).

Retrospective single-center review of patients with DOAC-associated ICH reversed with 4F-PCC or Andexxa®. Primary endpoint of hemostatic efficacy was determined by change in intraparenchymal hemorrhage (IPH) volume or thickness of subdural hemorrhage (SDH) or subarachnoid hemorrhage (SAH) on CT by 2 independent raters adopting the same criteria as the pivotal ANNEXA-4 study.

Of 54 patients (mean age 80, 48% female), 23 received Andexxa® and 83% received the low dose. 78% had traumatic ICH (SAH=24, IPH=15, SDH=25). SDH thickness was larger in the Andexxa® group (13.7 vs 5.1 mm, p=0.04) but ICH type, IPH volume (8 vs 7.75 cc), SAH thickness were similar. The Andexxa® group had lower percent expansion of any ICH (1.1% vs 18.7%, p=0.07) with significantly higher "excellent" or "good" hemostatic efficacy (100% vs 73%, p=0.033). Seven patients in the Andexxa® group had neurosurgery (EVD=3, SDH evacuation=4, craniectomy=1) compared to none in the 4F-PCC group (p=0.001). In-hospital mortality/hospice (n=5 vs 5) and home/acute rehabilitation (n=11 vs 15) were similar. One thrombotic event occurred in each group (DVT, stroke). Estimated median cost (AWP) for 4F-PCC is $11,151 vs $33,000 for low-dose Andexxa®.

Andexxa® shows significantly better hemostatic efficacy compared to 4F-PCC with similar thrombotic risk. Strengths include efficacy endpoints similar to the ANNEXA-4 study, adjudication by two independent raters and a cohort reflective of usual clinical practice. It remains unclear if improved hemostatic efficacy is relevant to clinical outcomes given no significant difference in mortality or discharge disposition. Larger studies and cost-analyses are warranted.

Direct-acting oral anticoagulants (DOACs) are used to reduce stroke risk of patients with atrial fibrillation (AF); however, these medications can increase the risk for major bleeding. Objective: To examine the healthcare economic burden of AF patients treated with DOACs who were hospitalized for intracranial hemorrhage (ICH).

Adult patients treated with DOACs (rivaroxaban, apixaban, edoxaban) who were hospitalized for ICH (1/1/2015-4/30/2018) were extracted from the MarketScan claims databases. Demographics and patient clinical characteristics were evaluated during a 6-month baseline period. Healthcare resource utilization and costs were determined for index hospitalizations and during a variable follow-up (1-12 months). Costs are reported in 2019 USD and were annualized for the follow-up period.

Among the study population of AF patients treated with DOACs (n=7,577), 9.9% (n=753) had a hospitalization for ICH; 58.3% (n=439) were non-trauma related, while 41.7% (n=314) were trauma related. Among the patients hospitalized for ICH, mean age was 77.9 years, 83% were ≥65 years, and 42% were female. Mean Charlson Comorbidity Index, CHA2DS2-VASc, and HAS-BLED scores were 4.3, 4.9, and 3.8, respectively. For index hospitalizations, the mean hospital LOS was 6.8 days (standard deviation [SD]: 7.3 days) and mean cost was $54,163 (SD: $111,883). During the follow-up (mean: 7.3 months), all-cause inpatient costs were on average $45,277 per patient (SD: $126,875), outpatient medical costs were on average $44,631 per patient (SD: $72,972), and total healthcare costs were on average $95,879 per patient (SD: $157,959).

Among AF patients treated with DOACs who had an ICH, initial hospitalizations were costly; the healthcare economic burden of patients with ICH extended well into the follow-up period and was nearly double that of initial hospitalizations. These findings indicate there is a significant unmet need in the management and follow-up care of patients treated with DOACs who are hospitalized with ICH.

MRI advances have led to the development of a low-field (64mT), point-of-care (POC) MRI device. The appearance of intracerebral hemorrhage (ICH) on POC MRI has not been characterized. We aim to describe ICH appearance on POC MRI.

We studied 15 patients (47% female, ages 43-89 years) with a diagnosis of ICH confirmed by conventional imaging. T1-weighted (T1W), T2-weighted (T2W), and fluid-attenuated inversion recovery (FLAIR) fast-spin-echo exams were obtained on 12, 15, and 14 patients, respectively. Two patients were studied at two and four serial timepoints. Three raters analyzed 13 T1W, 19 T2W, and 18 FLAIR POC exams. Signal intensity ratios (SIR) were computed by dividing the mean signal intensity of the lesion on a single slice by the mean signal intensity of the contralateral hemisphere. Averaged SIR and standard deviations were computed across raters. 

These preliminary data inform appearance of ICH on POC MRI. Further work is needed to differentiate signal intensities across ICH progression and clarify the sensitivity and specificity of POC MRI.

Because long-term uncontrolled hypertension (HTN) is associated with intracerebral hemorrhage (ICH) recurrence and post-ICH cognitive decline, it is an attractive opportunity for intervention. We investigated the prevalence and predictors of uncontrolled HTN at hospital discharge for patients admitted with ICH.

Subjects were patients aged ≥ 18 years presenting with CT-confirmed ICH who were prospectively recruited for a single-center longitudinal study from April 2016 to December 2019. ICH patients who did not survive to discharge were excluded. Two-sample t-tests and chi-squared tests of independence were performed to identify univariate relationships between covariates and uncontrolled HTN at discharge.

Stepwise regression utilizing Bayesian information criterion was performed to determine the final model.

345 patients were included in the analysis. 81.7% were hypertensive at presentation to the ED (BP≥140/90), and 54% had uncontrolled HTN at discharge (BP≥130/80). On presentation, 36.5% were taking 0 antihypertensives, 27.8% taking 1, 18.8% taking 2, 11.0% taking 3, and 5.9% taking ≥4. ICH location was deep in 48.7%, lobar in 42.3%, cerebellar in 5.5%, intraventricular in 2.9%, and multifocal in 0.6%. In multivariable analysis controlling for age; sex; pre-existing hypertension, hypercholesterolemia, diabetes mellitus, or coronary artery disease; renal insufficiency; previous ICH or ischemic stroke; number of home medications and antihypertensives; hematoma location; EVD placement; hematoma evacuation; presenting Glasgow coma score; and hospital length of stay, only HTN on presentation predicted uncontrolled HTN at discharge (p<0.001, 95% CI 0.49-1.64, OR 2.86).

Hypertension was uncontrolled at discharge for 54% of ICH survivors. This may represent a missed opportunity for the healthcare system to provide an acute intervention with substantial downstream benefits. Further studies are needed to determine the contributions to risk of uncontrolled HTN at discharge from the following: patient factors including genetics and medication resistance or intolerance; pharmokinetic limitations; treatment inadequacy; socioeconomic factors; and systems issues.

Portable, low-field MRI allows for neuroimaging at the point-of-care (POC). We aim to assess the accuracy of hematoma volume measurements in POC MRI and its relationship with clinical outcome in patients with intracerebral hemorrhage (ICH).

We studied 27 ICH patients (ages 21-82 years, 41% females). T2-weighted (T2W) and fluid-attenuated inversion recovery (FLAIR) exams were obtained on a 64mT, portable MRI system. Three raters used the ABC/2 method to measure hematoma volumes on POC exams and the closest (14.5±11.4 hours) standard-of-care (SOC) exam (8 CT; 16 1.5/3T MRI) within 36 hours. Intra-class correlation coefficients (ICC) were computed to assess inter-rater agreement of measurements and accuracy of POC volumes compared to SOC volumes. POC volumes were correlated against National Institutes of Health Stroke Scale (NIHSS) scores at the time of scan, length of hospital stays, and discharge clinical outcomes (modified Rankin Score). 

POC MRI volumes significantly correlated with SOC volumes, and there was a significant association between hemorrhage size on POC MRI images and discharge clinical outcome. Further work is needed to validate these findings and demonstrate the role of POC MRI in predicting long-term outcome after ICH.

Observational evidence from small, single-center studies suggests that lower admission hemoglobin (Hb) levels are associated with poor outcome in spontaneous intracerebral hemorrhage (ICH). Our aim was to test for association between admission Hb levels and poor outcome in multiple studies of ICH.

We performed an individual patient data meta-analysis of three studies of ICH, including two randomized clinical trials and one multi-ethnic observational study. Our exposure of interest was admission hemoglobin levels and the primary outcome was 3-month post-ICH dichotomized modified Rankin Scale (mRS) (0-3 versus 4-6). Intermediate outcomes were admission hematoma volume and hematoma expansion defined as 6 mL or 33% increase in hemorrhage size on repeat CT.

A total of 4,172 ICH patients were included in the analysis (mean age 63 [SD 14]; female sex 1,668 [40%]). Each additional g/dL of admission Hb was associated with 14% (OR 0.86, 95%CI 0.82-0.91) and 7% (OR 0.93, 95%CI 0.88-0.98) reductions in the risk of poor outcome in unadjusted and adjusted analyses, respectively. Dose-response analyses indicated a direct linear relationship between Hb levels and poor outcome across the entire evaluated range (test-for-trend p<0.001). No associations were found between admission Hb and hematoma volume or hematoma expansion.

Lower Hb levels are associated with poor outcome in ICH. Further research is needed to evaluate admission Hb levels as both a therapeutic target and predictor of outcome.

Uncontrolled hypertension (HTN) is associated with ICH recurrence and post-stroke dementia. Nonetheless, a substantial proportion of ICH survivors display uncontrolled HTN (uHTN) at hospital discharge and in outpatient follow-up. We hypothesized that transitions of care are a vulnerable period for HTN control and investigated whether HTN control suffered following transition from ICU to inpatient floor (transition).

Subjects were consecutive ICH patients ≥ 18 years in 2016-2019 at a single academic medical center. All survived hospitalization, spent > 24 hours in both an ICU and an inpatient floor post-ICU, and had ≥ 2 BP readings in the 24 hours prior to and following transition. uHTN was defined as SBP ≥ 130 mmHg or DBP ≥ 80 mmHg. Paired t-tests compared subjects' median SBP and DBP during 24 hours pre-and posttransition. Risers were patients whose median SBP and DBP rose and were hypertensive post-transition.

Median age (for n=195) was 71 (IQR 61-80) with uHTN present in178 (92%) on admission, in 150 (77%) at transition, and 100 (51%) at discharge. Among the 45 subjects who had achieved blood pressure control at transition, 17 (38%) had uHTN at discharge. Compared with pre-transition, post-transition median SBP was 5.8 mmHg higher (p < .001, 95% CI: [3.7-7.8]) and median DBP was 3.0 mmHg higher (p < .001, 95% CI: [1.9, 4.1]). There were 94 risers. In multivariable models, ICU stay was the only independent predictor for risers (p < .001, OR: 0.89/day).

For the average ICH patient, BP increased rather than decreased following transition from ICU to floor. Shorter ICU stay associated with BP increase following transition. >1/3 of patients who had achieved control at transition had uHTN at discharge. Interventions to maintain HTN control following transition might help ensure better BP control at discharge and long term.

We have previously identified clinical outcome heterogeneity amongst patients with different primary lobar intracerebral hemorrhage (ICH) etiologies defined by modified Boston Criteria. We sought to evaluate whether there are functional coagulation differences amongst these lobar ICH etiologies using whole blood viscoelastic coagulation assay (Rotational Thromboelastometry [ROTEM]).

Spontaneous, primary lobar ICH patients admitted to a single, tertiary referral center between 2016-2019 who received baseline ROTEM testing were included for analysis. Patients with preceding anticoagulant use or baseline evidence of coagulopathy using traditional coagulation testing were excluded. Cerebral amyloid angiopathy (CAA) ICH etiology was the primary exposure defined as probable/definite CAA ICH (vs possible/non CAA ICH) using modified Boston Criteria. Multivariable linear regression analyses evaluated the association of CAA ICH exposure on admission ROTEM data after adjusting for age.

Of 31 lobar ICH patients included for analysis, 12 (39%) were probable/definite CAA ICH. Probable/definite CAA ICH patients were older compared to possible/non-CAA ICH patients (79 +/-6 vs 52 +/-18 years old). Multivariable linear regression revealed that probable/definite CAA etiology was associated with faster extrinsic (EXTEM) pathway coagulation kinetics on ROTEM testing compared to possible/non-CAA ICH etiologies (EXTEM coagulation time: 63.0 +/-14.3 seconds in probable/definite CAA vs 95.3 +/-76.9 seconds in possible/non CAA ICH patients; stand coeff B: -0.53; 95%CI: -129.27 to -3.38; p=0.04 ). We did not identify any other intergroup differences in functional coagulation detectable using ROTEM or traditional coagulation tests.

We identified faster coagulation kinetics specific to the extrinsic pathway in lobar ICH patients with probable/definite CAA compared to possible/non-CAA ICH etiologies. Further work is required to see if these results are replicable, to investigate mechanisms for these findings, and to discern whether faster coagulation kinetics translates to more optimal hemostasis in lobar CAA ICH compared to non-CAA ICH patients.

Patients with hydrocephalus due to intracerebral hemorrhage (ICH) often require an extraventricular drain (EVD) for CSF diversion. Prediction of dependence on permanent cerebrospinal fluid (CSF) shunting remains challenging. Prior studies have shown a CSF inflammatory response after ICH, but the role of inflammation in shunt dependence is unknown. We evaluated the role of CSF cell index (CI) in ventriculoperitoneal shunt (VPS) dependence.

We performed a single-center retrospective cohort study using prospectively collected data from consecutive patients with ICH. We included patients with an EVD who had available CSF laboratory data without culture-proven ventriculitis. Peak CSF leukocytes, peak CI, and days to peak cell counts were calculated. Our primary outcome was VPS placement. Mann-Whitney U test was used to study the relationship between CSF leukocytes and CI and VPS placement. K shape clustering was performed to evaluate differences in CI trends. 

Higher cell index and longer time to peak cell index are associated with VPS placement. Our study suggests that CI may be a better predictor of VPS placement than CSF WBC count, and as such may strengthen the value of a predictive model for VPS placement in ICH.

Volumetric assessment is key to define treatment decisions & prognosis. Prior studies have described algorithms to derive volumetric analysis of intraparenchymal hemorrhage (IPH). We aimed to develop a software scheme to calculate total ICH volume with delineation of IPH and IVH component.

In this study, we developed a proprietary automated interactive computer-aided detection (ICAD) software scheme to perform segmentation of CT brain imaging and compute volumetric analysis of ICH volume via a 3-dimensional image. ICAD uses a combination of CT-Hounsfield unit of each voxel and image thresholding techniques to automatically estimate & segment the hemorrhage. Further, each CT image was reviewed slice by slice by two observers & the boundaries of total intracerebral hemorrhage volume (ICH), IPH & intraventricular hemorrhage volume (IVH) were demarcated by a semi-automated method. Each observer analyzed a randomly selected independent set of 100 patients. Further, machine learning algorithm was used to differentiate IPH and IVH in the automated segmentation boundaries of ICH defined by the ICAD tool. For this purpose, the observers' boundaries of ICH and IVH were used as the ground truth to label the automated segmentation results using the nearest neighboring principle. Initial admission CT imaging for 200 patients out of the 411 patients admitted with a diagnosis of ICH between years 2012-2015 at OU Medical Center were analyzed. Finally, the spatial overlap and validation of the ICAD tool was performed by calculating dice similarity coefficient (0-1) (DSC) between automated and semi-automated boundaries for ICH, IVH, and IPH. Moreover, the values of DSC were compared between the two independent subsets reviewed by observers. 

This is an internal validation of ICAD in this patient cohort of ICH irrespective of location of hemorrhage. Further steps involve using deep learning methods to perform fully automated volumetric analysis of ICH and markers of secondary brain injury in ICH.

Spontaneous intracerebral hemorrhage (sICH) is a major cause of morbidity and mortality worldwide. There is limited data regarding predictors of successful extubation among sICH patients.

We performed a retrospective study to compare the demographic, clinical and radiographic characteristics of "successfully extubated" and "unsuccessful extubated" patients. We define "all-time" successful extubation as requiring no tracheostomy during the duration of hospitalization. We then performed a multivariate logistic regression model to determine the independent predictors of successful extubation among these patients. Our secondary outcome include hospital and intensive care unit length of stay.

A total of 71 patients with sICH were included for analysis. They were on average 57±12 years old, mostly African American ( 59,75% ) and males ( 39, 49% ) with and admission SOFA score of 4±2, GCS of 8 (IQR 7). The most common ICH location was the basal ganglia (31, 39%). Majority of the patients were successfully extubated (54, 68%).The demographic, clinical and radiographic characteristics of the "successful" versus "unsuccessful" cohorts were comparable except for age (57±13vs56±12, p=0.025), absence of brainstem component (45, 83% vs 11, 44% p<0.001), absence of external ventricular drain (38, 70% vs 10,40% p=0.014) and time from admission to intubation (days) (0.06±0.32vs0.27±0.70, p=0.002). In the multinomial logistic regression model, including the covariates of age, absence of brainstem component, time to intubate and absence of EVD, it was found that the absence of brainstem component increases the odds of "all-time" successful extubation (OR 1.9, 95%CI 2.002-22.23). Patients successfully extubated had a significantly shorter Hospital lengths of stay (19.51±12.13vs21.13±20.24, p=0.045) compared to the "unsuccessful" cohort.

Successful extubation in our cohort was 68%. We showed that the absence of brainstem involvement in ICH is an independent predictor of all-time successful extubation for patients with ICH. This information can provide as a guide decision making regarding extubation, tracheostomy and goals of care.

Chronic subdural hematoma (cSDH) is predicted to become the most common cranial surgical condition by 2030. Recurrence is estimated between 5-15%, with higher rates associated with anti-coagulant use, bilateral disease, baseline comorbidities, and advanced age. Use of a surgical drain after evacuation has been associated with lower recurrence rates. We previously reported the first use of an irrigating surgical drain placed after craniotomy in the United States. Here we provide a look at the first grouped retrospective analysis of the use of an irrigating surgical drain (IRRAS, Stolkholm, Sweden) after craniotomy for treatment of subdural hematoma in 6 patients.

Retrospective review of 6 cases of the use of an irrigating surgical drain after craniotomy for treatment of subdural hematoma at a single institution. Goal is to highlight the learning curve of utilizing a new device and assessing settings and duration of an irrigating drain, hospital stay, neurological status at follow-up and hematoma recurrence.

The learning curve associated with a new device was witnessed in the evolution of placing the drain intraoperatively, educating nurses on proper care and documentation, and establishing comfort with elevating drainage rates from minimal to aggressive (3ml/hr to 100ml/hr). The average length of stay was 3.83 days, almost half the average length of stay in large cohort studies. Each case had no complications in the placement or removal of the irrigating drain, no spontaneous recurrences to date, and improved neurologic exam from pre-operative exam.

The largest review to date of the use of an irrigating surgical drain after craniotomy shows a safe and effective method of treating subdural hematoma. Use of this drain warrants further investigation and establishment of a standard protocol to compare against current treatment regimens to evaluate impact on patient outcomes, recurrence, hospital stay and cost.

Glucose variability (GV) has been reportedly a risk factor for poor outcome after ischemic stroke. However, its effect on hemorrhagic transformation after endovascular recanalization therapy (ERT) remains to be elucidated. In the current study relationships study the relationship between GV and symptomatic intracerebral hemorrhage (sICH) after successful recanalization with ERT were investigated.

A total of 176 patients with acute ischemic stroke who had undergone successful recanalization after ERT (Thrombolysis in Cerebral Infarction 2b or 3) were enrolled between January 2013 and November 2019. Glucose data after ERT were obtained during the first 36 hours, and parameters of glucose including mean, maximum, minimum, difference between maximum and minimum, standard deviation, coefficient of variation, successive variations, median absolute deviation, average consecutive absolute change, median consecutive absolute change, and time rate (TR) of glucose variation were assessed. The outcomes were sICH and unfavorable outcome at 3 months after ERT. The sICH was defined as parenchymal hemorrhage type 2 with neurological deterioration of 4 points of more on the National Institute of Health Stroke Scale. Moreover, modified Rankin scale 3-6 at 3 months was considered as an unfavorable outcome.

Among all patients (n = 176), sICH developed after successful ERT in 16 (9.1%) patients. Parameters associated with glucose fluctuation over time were significantly related to sICH and unfavorable outcome. After adjusting for potentially confounding variables, the TR of glucose (per 1 mg/dL/h increase) variation was independently associated with sICH (odds ratio, 1.17; 95% confidence interval [CI], 1.012-1.343) and 3-months unfavorable outcome (OR 1.14, 95% CI, 1.000-1.297).

Time-related GV during the first 36 hours after successful ERT was correlated more strongly with sICH and short term outcomes than any other glucose parameters. This suggests that maintaining stable glucose may be an important factor in prevention of sICH after successful ERT.

COVID-19 is a global pandemic with major economic and health risks to patients and medical staff at hospitals. We hypothesized that we could protocolize acute stroke workflow to triage COVID-19 patients in the prehospital phase and intrahospital phase while using telemedicine to minimize the amount of personal protective equipment (PPE) and personnel required from the stroke team members, as well as reducing potential donning/doffing requirements and COVID-19 exposures from intubation.

We describe a progressive modular acute stroke workflow involving a tiered approach with treating every stroke patient as a person under investigation (PUI) for COVID-19 for maximum protection of staff while minimizing the use of personal protective equipment (PPE) by using prehospital and intrahospital telemedicine to prevent unnecessary exposures to stroke team staff.

These step-by-step workflow measures instituted at our facility helped to maintain delivery of time sensitive treatment for acute stroke while minimizing PPE and staff exposure to COVID-19. Optimizing patient care lead to conservation of PPE as well as adaption of protocol as inpatient census of COVID-19 positive patients changed our coordinated team response.

A multidisciplinary group of stroke and COVID-19 experts from neurology, neurocritical care, the interventional practice, emergency medicine, and anesthesia built upon an already very efficient patient flow algorithm to create a streamlined process for quickly preparing stroke patients for intervention while maintaining maximum provider safety. Healthcare providers and hospitals are urged to acclimate our protocol to best enhance current institutional protocols, bearing in mind existing best practice recommendations and limited resource availability to ensure high quality care and continued safety of patients, providers and staff.

The Destiny II trial showed that patients older than 60 with large middle cerebral (MCA) infarcts have decreased mortality with decompressive hemicraniectomy (DHC) when compared to medical treatment, yet remain substantially disabled. We hypothesize that maximal medical management in those patients has comparable outcome as DHC.

We performed a retrospective study of patients with large MCA infarcts, who were older than 60 and admitted between April 2017 and December 2019 to a single, tertiary care, academic center. Large MCA infarcts were defined as >2/3 of the MCA territory. Demographic data, clinical characteristics and medical treatment were collected via chart review. Primary endpoint was a dichotomized modified ranking score (mRS 0-5 versus mRS 6) 12 months after the stroke and was obtained via standardized phone interviews. The DHC arm of the DESTINY II trial served as the control group using the Chi-Squared test.

We analyzed 26 patients with large MCA infarcts who received maximal medical treatment only. Median age was 71.5 (range 61-96). 54% were male. Median NIHSS on admission was 16.5 (4-31). 69% had right hemispheric infarcts. 35% had an M1, 19% had an ICA and 12% had a tandem occlusion. 92% had MCA infarcts while the rest had combined MCA/ACA infarcts. 31% received tPA and 27% underwent thrombectomy. All patients were admitted to the NICU. 38% received hypertonic saline treatment. Median hospital stay was 10.5 days (4-54) and median ICU stay was 4.5 days (1-33). We did not find a significant difference in 12-months survival rate between our cohort (54%) when compared to the DHC arm of the DESTINY II trial (57%). However, mRS scores in our group of surviving patients were significantly smaller (mean: 3.44 vs. 4.22 / p=0.02).

Maximal medical management could have similar clinical outcome as DHC in patients older than 60 with large MCA infarcts.

Stroke is a leading cause of death and disability in sub-Saharan Africa. With limited capacity for acute stroke interventions in low-resourced healthcare settings, strategies to prevent post-stroke complications such as aspiration pneumonia (APNA) may hold the greatest promise for reducing critical illness and death after stroke. We investigated the incidence and clinical impact of post-stroke APNA in a prospective cohort in Zambia.

We conducted a prospective cohort study of adults with stroke admitted to University Teaching Hospital (Lusaka, Zambia) between 12/2019-3/2020. Demographic and clinical parameters were captured serially over the hospital course, including exam findings, NIH Stroke Scale (NIHSS), Glasgow Coma Scale (GCS), Modified Rankin Scale (mRS), and nine indicators of possible APNA (e.g. fever, cough, radiographic pneumonia, etc). Possible APNA was defined as ≥2 indicators present, and likely APNA as ≥4 present. Ttests and chi-square tests were used to compare clinical parameters between participants with and without likely APNA.

125 participants with stroke were enrolled: 60.8% female, mean age 60+/-16 years, 54% ischemic stroke. Seventy (56%) had possible APNA, and 39 (31%) had likely APNA. Swallow screening was documented for only 4 participants (3%). Participants with likely APNA were older (67.7 vs 56.1 years, p<0.001), had higher admission NIHSS (25.9 vs 16.7, p<0.001), lower admission GCS (8.1 vs 12.1, p<0.001), and higher admission mRS (4.96 vs 4.32, p<0.001). Likely APNA was associated with significantly increased risk of in-hospital mortality (84% vs 32%, p<0.001). There were no between-group differences in rate of head-of-bed elevation or oral feeding.

APNA is a frequent hospital-acquired complication of acute stroke in Zambia, especially among older patients with more severe presentations. Importantly, APNA is associated with significantly increased mortality, suggesting that interventions to mitigate APNA may improve stroke outcomes in Zambia and other low-resourced settings.

Acute stroke care has led to an increase in the number of first-time ischemic stroke survivors, some of whom may face sequelae such as the development of seizures. Previous work suggests that thrombectomy and thrombolysis may be associated with increased risk of post-ischemic stroke seizures. We utilized a multi-state analysis of a large administrative claims data to investigate the effect of stroke treatment on the risk of post-ischemic stroke epilepsy (PISE) after ischemic stroke.

Using claims data from California, New York, and Florida, we identified adult survivors of a first time acute ischemic stroke (ICD-9-CM code 434.X1). Patients without a longitudinal identifier, out-of-state residents, and patients with history of epilepsy, brain trauma, infections, or metastases were excluded. Included subjects were followed for a primary outcome of any readmission or ED visit (available in California only) for seizure (ICD-9-CM code 345.X). Cox proportional hazards regression was used to identify covariates associated with the development of PISE.

Of 563,608 included patients (mean age 74 [SD 14], 52% female), 6,278 (1%) developed PISE during a median follow-up time of 4.0 years (IQR 2.2-6.0). Mechanical thrombectomy was associated with development of PISE (HR 2.33, 95%CI 1.69-3.20, p<0.001) in univariable models only. In multivariable models adjusting for age, sex, and race, we discovered that pre-existing atrial fibrillation (HR 1.93, 95%CI 1.80-2.07, p<0.001) and administration of tPA (HR 1.39, 95%CI 1.24-1.56, p<0.001) were significantly associated with development of epilepsy. In the same models, patients who underwent decompressive hemicraniectomy experienced a nearly 5-fold increase in odds for developing epilepsy (HR 4.71, 95%CI 3.47-6.39, p<0.001).

We provide evidence that survivors of ischemic stroke with atrial fibrillation and those who received tPA or underwent hemicraniectomy experience significantly increased risk for PISE.

Epilepsy is a known complication of ischemic stroke with an incidence rate of 2-4% (Camilo et al 2004) .

Prior studies concluded that stroke patients have a higher risk of epilepsy after undergoing acute stroke therapies (IV-tPA and/or thrombectomy) (Naylor et al 2018). However, it is unclear whether this increased risk is due to increased stroke survival or a direct impact of the intervention. In this study, we sought to identify the effect of acute stroke therapies on post ischemic stroke epilepsy (PISE) by evaluating TICI score, change in NIHSS from admission to discharge, and reperfusion-related complications during hospitalization.

We identified patients from the Yale Acute Brain Injury and Tissue Repository who developed PISE (≥1 seizure in month 2-12 post-stroke). Patients were matched with non-PISE stroke controls 3:1 based on admission NIHSS, age, and sex. We included patient demographics, past medical history, and hospital interventions in our analysis. We collected variables from the patient's hospital stay, including admission and discharge NIHSS, admission and discharge modified Rankin score, IV-tPA administration, thrombectomy, TICI score, hemorrhagic transformation, and decompressive hemicraniectomy. We performed univariate and multivariate logistic regression analysis.

We identified 34 PISE patients out of 1471. Our univariate analysis revealed that stroke patients are more likely to develop PISE if they have TICI 0 (vs. 3) (p=0.03), higher discharge mRS (p=0.06), little/no improvement in NIHSS from admission to discharge (p=0.006), hemorrhagic transformation (p=0.01) and decompressive hemicraniectomy (p=0.02). In our multivariate regression, independent predictors included little/no improvement in NIHSS (p=0.006), hemorrhagic transformation (p=0.02), and decompressive hemicraniectomy (p=0.04).

These data suggest that persistently high NIHSS assessments, irrespective of acute interventions, and complications of ischemic stroke increase PISE risk. Both predictors are proxies for severe stroke burden, suggesting that injury severity drives increased PISE risk rather than acute stroke interventions directly.

Clinical predictors of functional and safety outcomes in IV thrombolysis (IVT) for acute ischemic stroke, have been poorly studied in Latin America. Our aim was to assess predictive factors for in-hospital functional outcomes in patients with acute ischemic stroke treated with IVT alone.

This is a retrospective study of consecutive ischemic stroke patients, treated with IVT in a Public Tertiary Hospital from Costa Rica, from February 2012 to December 2019. Demographic, clinical and imaging variables were recorded. Univariate and adjusted multivariate linear and logistic regression models were elaborated in order to establish independent predictors of poor functional outcomes. In-hospital poor functional outcomes were defined as a modified Rankin score (mRs) of 3-6 and post IVT symptomatic hemorrhage.

We included 446 patients with acute ischemic stroke treated with IVT. Median NIHSS reported was 13 (IQR 9-18) with a mean door-to-needle time of 43 ± 17.1 minutes. Thirty two (7.1%) patients died and twenty-seven (6.1%) developed symptomatic hemorrhage. Multivariate regression analyses established that having a smoking history (OR: 2.1, 95%CI 1.2-3.8), presentation NIHSS (OR: 1.2, 95%CI 1.1-1.2) and total anterior circulation infarction (TACI, Oxfordshire classification) (OR 3.1, 95%CI 1.6-5.7) were independently associated with a discharge mRs of 3-6. Moreover, severe brain edema (COED3, SITS-ISTR) (OR: 8.1, 95%CI 1.4-47) and baseline NIHSS (OR:1.1, 95%CI 1.0-1.2) significantly correlated with death. Finally, patients with a prior use of antiplatelet therapy (OR: 2.5, 95%CI 1.1-5.6) and higher baseline NIHSS (OR:1.1, 95%CI 1.0-1-2) had a higher likelihood of developing symptomatic hemorrhage.

Our study showed that presentation NIHSS, smoking history and large stroke were independent predictors of poor outcome (mRS 3-6), while prior antiplatelet therapy and presentation NIHSS were independent predictors of developing symptomatic hemorrhage following thrombolysis.

Improvement in door-to-needle (DTN) times in recombinant tissue plasminogen activator (rtPA) administration for acute ischemic stroke is considered an important aspect in improving functional outcomes of acute ischemic stroke patients. The objective of our study was to analyze the impact of improving DTN on discharge functional outcomes.

Retrospective data from consecutive acute ischemic stroke patients treated with rtPA from February 2012 to December 2019 in a Tertiary Hospital were included. Demographic, clinical and imaging variables were recorded. Hospital acute stroke protocol was reviewed in three periods (2012-2014, 2015-2017 and 2018-2019) , with subsequent analysis of DTN and functional outcomes as indexed by the modified Rankin score (mRS, poor functional outcome=3-6), death and symptomatic post rtPA hemorrhage.

There were 446 patients during the time period (52.2% male, mean age 67.1 ± 14.4 years, and median baseline NIHSS of 13 (IQR 9-18)). Anterior circulation strokes accounted for 46.4% of cases, with an large vessel etiology as the most frequent subtype (44.1%). There was a significant trend in the improvement of DTN median time across the three periods analyzed, from 2012-2014 (45 minutes, IQR 38-56), to 2015-2017 (40 minutes, IQR 34-50) and 2018-2019 (36 minutes, IQR 25-45) (p˂0.001). 59.4% of cases presented with mRs=3-6, the mortality rate was 7.1% (32 patients) and 27 patients (6.1%) developed symptomatic hemorrhage. Poor functional outcome (mRS=3.6) (p=0.98), death (p=0.74) and symptomatic hemorrhage (p=0.14) showed no significant differences across the three periods of time.

Our results showed that improving DTN did not have major impact on hospital discharge mortality and functional outcome, however caution should be used in interpreting such results given single center retrospective nature of this study. Further various randomized control trials have shown strong impact of DTN in improving functional outcome at 90 days.

Large hemispheric infarct(LHI) patients are at risk for secondary brain injury(SBI) characterized by early neurological decline(END) and malignant cerebral edema(MCE). We hypothesized that admission characteristics and time-and frequency-domains of heart rate variability(HRV) can predict both END and MCE.

LHI patients, defined as acute anterior circulation ischemic stroke with an admission NIHSS score of ≥ 15, were selected for retrospective analysis. Demographics, acute stroke intervention characteristics, and modified Rankin score(mRS) at 90 days were collected. END was a delta NIHSS≥ 4 at 24 hours from admission. MCE was clinical and/or radiographic herniation or need for craniectomy within 48 hours of injury. Continuous ECG was collected over the first 3 hours of NCCU admission to calculate time-and frequency-domain HRV features. Multivariable logistic regression was performed to determine factors independently associated with END, MCE, and poor outcome at 90 days(mRS≥ 4).

452 consecutive patients met study criteria(age: 66+/-15 years, 52% women, NIHSS score 20+/-4) between January 2016 and December 2019. END was identified in 12%, MCE in 18%. Higher admission NIHSS score(p = 0.035) and low frequency to high frequency(LF:HF) ratio(p = 0.004) were independent predictors of END. Higher admission glucose(p = 0.045), younger age(p = 0.01), tPA administration(p = 0.021), and mean ECG-NN interval(p=0.036) were independent predictors of MCE. Factors associated with poor outcome at 90 days included higher NIHSS score on admission(p = 0.001), no intervention with mechanical thrombectomy(p = 0.04), higher admission glucose(p = 0.011), older age(p = 0.00), END(p = 0.009), and MCE(p = 0.011).

Time-and frequency-domain features of HRV in the first 3 hours after admission and baseline clinical factors are associated with END and MCE. Both END and MCE are contributors of poor outcome at 90 days in LHI patients. Further analyses exploring the relationship between HRV and SBI after LHI are warranted.

Neurocritical care is routinely offered to patients post neurothrombectomy of anterior large vessel occlusion (LVO) strokes. We aim to study the relationship between immediate improvement in NIHSS score on outcomes post thrombectomy and potential implications for requiring neurocritical care.

We performed a retrospective review of anterior LVO (internal carotid/proximal middle cerebral artery) patients with undergoing neurothrombectomy between January 2015-December 2018. Demographic, clinical (NIHSS score on admission and within 30 minutes post recanalization, time last know well-TLKW), and imaging (ASPECTS, TICI, intracranial hemorrhage) information was analyzed. Ultra-early functional improvement (Ultra-EFI) was defined as NIHSS score <6 within 30 minutes of successful recanalization. We analyzed the incidence and predictors of ultra-EFI and explored reasons for neurological decline post ultra-EFI.

Of the 343 anterior LVO patients undergoing neurothrombectomy, mean age was 71 ±15 and 46% were males. Mean NIHSS was 17±6 and TLKW to arrival was 9±11 hours. Ultra-EFI was observed in 23% (79/343) of patients. Independent predictors of ultra-EFI include lower pre-treatment NIHSS (p=0.003), favorable ASPECTS (p=0.02), and lower systolic blood pressure (0.01). Rates of 90-day-mRS 0-2 were higher (71% VS 33%, P<0.01) and the rate of mortality (8% VS 28%, P<0.01) was lower in the ultra-EFI group compared to the non-ultra-EFI group. Amongst patients with ultra-EFI, 1.3% (1/79) experienced increase in NIHSS by ≥4 points. This patient received thrombolysis, achieved TICI-2B recanalization, and follow-up neuroimaging revealed a parenchymal hemorrhage and an infarct volume of 44 ml. None of the patients required continuous antihypertensive infusions.

Approximately 23% of anterior LVO stroke patients undergoing neurothrombectomy have a NIHSS score of <6 within 30 minutes of successful recanalization. Approximately 1% of them experience significant decline in neurological status within 24 hours of the procedure with the majority achieving functional independence at 90 days. Need for advance neurocritical care should be re-evaluated in these patients.

The American Heart Association's Get with the Guidelines (GWTG) Stroke program has changed stroke care delivery in the USA since its establishment in 2003. GWTG is a voluntary registry and continuous quality improvement initiative that collects data on patient characteristics, hospital adherence to guidelines and inpatient outcomes. Studies have suggested that women may receive lower quality of care (QOC) than men.

We retrospectively analyzed consecutive patients with HCA CLM from 2016 to 2019 among men and women. Demographics and clinical data were collected from the Get-With-The-Guideline (GWTG)-Stroke registry and electronic medical records. Quality benchmarks for stroke education, door-to-needle (DTN) times, dysphagia screening, NIHSS, evaluation for rehab, LDL screening were analyzed. We analyzed our GWTG scores between men and women using Fisher's exact test with statically significant set at p < 0.05.

In the year 2019, women with ischemic stroke and atrial fibrillation were anticoagulated at a statically lower rate than men (p < 0.0001). And, in the year 2019, women with ischemic stroke were prescribed anti-thrombotics at a statically lower rates than men on discharge (p = 0.0263).

Women had a lower overall stroke QOC, although absolute differences in most individual GWTG endpoints were small. Further investigation into the factors contributing to the gender disparity in stroke treatment should be pursued.

Approach to acute cerebrovascular disease management has evolved in the past few months to accommodate the rising needs of the 2019 novel coronavirus (COVID-19) pandemic. In this study, we investigated the changes in practices and policies related to stroke care through an online survey.

A 12 question, cross-sectional survey targeting practitioners involved in acute stroke care in the US was distributed electronically through national society surveys, social media and personal communication.

Respondants from 39 states completed 206 surveys with the majority (82.5%) from comprehensive stroke centers. Approximately half stated some change in transport practices with 14 (7%) reporting significant reduction in transfers. Common strategies to limit healthcare provider exposure included using personal protective equipment (PPE) for all patients (127; 63.5%) as well as limiting the number of practitioners in the room (129; 64.5%). Most respondents (81%) noted an overall decrease in stroke volume. Many (34%) felt that the outcome or care of acute stroke patients had been impacted by COVID-19. This was associated with a change in hospital transport guidelines (OR 1.325, P=0.047, 95% CI: 1.004-1.748), change in eligibility criteria for IV-tPA or mechanical thrombectomy (MT) (OR 3.146, P=0.052, 95% CI: 0.988-10.017), and modified admission practices for post IV-tPA or MT patients (OR 2.141, P=0.023, 95% CI: 1.110-4.132).

There is a reported reduction in stroke volume across the nation. Updates in hospital transport guidelines and practices related to IV-tPA and MT may affect the perceived care and outcome of acute stroke patients.

One of the most serious complications of acute ischemic stroke is malignant cerebral edema (MCE). Many biomarkers have been investigated to predict MCE, but currently none are used to guide management. We assessed if patients with ≤50% collateral flow (compared to the contralateral side) on the presenting CT angiogram were more likely to develop MCE than those with >50% collateral flow. We also assessed if the speed of collateral loss further predicted the development or timing of MCE.

We retrospectively analyzed all large vessel acute ischemic strokes at Mayo Clinic Florida from January 2018 to June 2019. We obtained basic demographics and presenting symptoms, including NIHSS. Two independent and blinded investigators scored all images by assigning Alberta Stroke Program Early CT Score (ASPECTs) and Modified-TANs scores, as well as confirming the location of the occlusion on CT angiogram. We recorded the presence, timing, and management of MCE. The study was approved by the local IRB.

There were 46 patients with ICA, M1, or M2 ischemic strokes. 11 patients (24.4% developed MCE. There was no difference in age, sex, NIHSS, GCS, ASPECTS score, or successful reperfusion between patients with MCE and those without. Patients who developed MCE were 6.42 times more likely to have had <=50% collaterals on initial CT angiogram than patients who did not MCE (CI 1.20, 34.40, p=0.03). There was no association between collaterals and timing of MCE and no association between fast or slow progressors in the development or timing of MCE. No slow progressor died while 31.2% of all fast and normal progressors (p = 0.04).

Collateral blood flow as measured by the Modified-TAN score may be a simple way to predict the future development of MCE. Accurate prediction may lead to focused initiation of prophylactic or therapeutic treatments.

Acute Ischemic stroke (AIS) patients are frequently admitted to neuroscience intensive care unit (NICU) after receiving tissue plasminogen activator (TPA). AIS patients who are admitted to NICU undergoes frequent hourly neurological examinations to quickly detect neurological deterioration. The hourly neurological examination requires the patient to be awakened and cause disruption of the sleep-wake cycle. There is an association between poor sleep quality and delirium. Delirium increases the risk for mortality, morbidity, length of hospitalization, and institutionalization in patients with AIS.

We retrospectively chart reviewed patients admitted to our NICU with AIS after TPA during a one-year period. Delirium incidence during hospitalization was assessed through confusion assessment method for the ICU (CAM-ICU) scale. Antipsychotic medications for behavioral modification was assessed through medical administration record. Post thrombolytic neurological complications were studied by assessing the need for emergent head computed tomography during the first twenty-four hours.

The mean age of our patients were 66.5 years. The average national institute health stroke scale was 10.58. The average modified Rankin Scale at discharge was 2. One patient out of fifty was diagnosed with delirium and was discharged home on antipsychotic medication. One patient out of fifty had a post thrombolytic intracranial hemorrhage, however, this was not captured in the hourly neurological examination and was found on the next day post TPA twenty-four head CT.

Frequent neurological examinations are not associated with increased risk for delirium during the hospitalization in the AIS patients who received thrombolytic therapy. The study also did not find an increased incidence of prescription of antipsychotics secondary to delirium at discharge. Frequent hourly neurological examinations did not increase the detection of neurological deterioration. Further studies looking at stroke characteristics and patient comorbidities are needed to understand who will benefit the most from hourly neurological examination.

Basilar artery occlusion (BAO) increases risk of dysphagia and recurrent aspiration. Tracheostomy and percutaneous endoscopic gastrostomy (PEG) tube placements are important decision points in poststroke care. In patients with BAO, higher NIHSS have been associated with a higher degree of disability as measured by the modified Rankin Scale (mRS). However, there is currently no data on predictors of tracheostomy and PEG placement in this population.

Seventy-seven patients with a BAO treated at a comprehensive stroke center (70 mechanical thrombectomies, 7 maximal medical management) were included in the study. Retrospective data was collected on general demographics, NIHSS on days 0, 1 and 5, TICI score, pre-stroke and 90-day mRS score, and presence of atherosclerosis in the posterior circulation. Primary outcome was defined as the need for tracheostomy or PEG tube during the stroke hospitalization. Medians were compared with Mann-Whitney U Test and categorical variables with Fisher's exact test. Logistic regression was used to analyze which variables may predict need for tracheostomy or PEG. SPSS was used for all statistical analyses.

Seventy-five out of 77 patients had data available for the primary outcome. 19/75 (25%) required a tracheostomy and 23/75 (30.3%) required PEG tube. The need for trach/PEG was not associated with NIHSS on days 0, 1, or 5, TICI reperfusion score, or the presence of atherosclerotic disease in the posterior circulation. While there was an association between history of hyperlipidemia and reduced need for PEG placement, this did not hold up in multivariate analysis.

While in previous studies higher NIHSS was associated with a worse functional outcome in posterior circulation stroke, it did not correlate with the need for tracheostomy or PEG in our case series of patients with BAO. Further research is needed to identify prognostic factors for recovery in these patients, which may help guide family and clinician decision-making.

Calcium is known to be a major mediator in ischemic neuronal cell death as well as a mediator of endothelial-derived vasodilation. Elevated serum calcium levels at admission in patients with ischemic stroke have been associated with milder clinical deficits, better clinical outcomes and smaller infarct sizes. However, the relationship between serum calcium levels and penumbral volumes on acute neuroimaging scans has not been studied previously.

We conducted a retrospective review of all patients with primary diagnosis of acute middle cerebral artery strokes between 08/01/2019-04/01/2020 (109 subjects total), who were assessed with CT or MRI perfusion scans that were processed with RAPID imaging software. Cases with no evidence of infarct and penumbral volumes were excluded. Data collected include demographics, stroke etiology subtype, comorbidities, laterality of middle cerebral artery stroke, admission serum calcium levels, initial NIHSS, and MR and CT perfusion data. Serum calcium levels were collapsed into quartiles initially and then dichotomized into the lowest quartile versus all others combined. Baseline demographics and clinical characteristics were compared across quartiles using chi-square test for percentages and Kruskal-Wallis rank sum test for medians.

Our analyzed cohort has a median age of 69 years, and 72.7% of the cohort was male. Serum calcium levels were collapsed into < 8.7 mg/dL and ≥ 8.7 mg/dL. 

The lowest quartile of serum calcium levels was significantly associated with the largest penumbral volumes (TMax > 6 second threshold) as compared to all other quartiles combined. Serum calcium levels may influence penumbral burden. Neurointensivists should consider serum calcium levels in their initial evaluation of penumbral burden in ischemic stroke cases.

Apollo Hospitals, Comprehensive Stroke Unit, Chennai, India

Intravenous Tenecteplase is the standard of care for Acute Ischemic Stroke patients presenting within 3 hours of symptom onset. Mechanical Thrombectomy is indicated in large vessel occlusions (LVO) in those with thromboembolism in the Terminal ICA, Carotid T or Proximal M1 territories.

We included consecutive patients admitted for Acute Ischemic Stroke, who presented within 3 hours of symptom onset and were evaluated with CT scan of Brain with Angiography of Carotid and Intracranial Vessels. Patients were treated with a bolus dose of Intravenous Tenecteplase (0.2mg/Kg). LVO subjects were subjected to Mechanical thrombectomy using a stent retriever with a time window of up to 24 hours. All patients were evaluated using the NIHSS, GCS, mRS, and Barthel Index. Standard Stroke Risk Stratification was performed. The outcomes were measured at 24 hours, 7 days, 1 month, and 3 months. The rate of recanalization, hemorrhagic changes requiring decompressive craniotomy and mortality were studied.

26 patients were studied. Male : Female = 20:6, Mean age = 62.4 (35-87). Mean Door-to-Groin time was 94 mins (45-120mins). Immediate improvement in recanalization rates was observed in 22/26 (84.61%) patients. Favorable outcomes of mRS 0-2 at 90 days were seen in 16 patients (61%). Bad outcomes of mRS 3-5 were seen in 9 patients (34.62%). 1 patient experienced in-hospital mortality. Decompressive craniotomy was performed in 6/26 patients (23%). Diabetes Mellitus and Hypertension were seen in 80% of patients.

Intravenous Tenecteplase followed by Mechanical Thrombectomy resulted in improved recanalization rates and functional independence at the end of 90 days. Our rate of favorable outcomes are significantly higher when compared to treatment with Intravenous Alteplase and Mechanical Thrombectomy, or only Mechanical Thrombectomy in patients presenting within 4.5 hours of symptom onset.

Desmopressin (DDAVP) has been recommended for antiplatelet medication reversal in patients with intracranial hemorrhage (ICH) but there are limited data describing its side effects especially pertaining to hyponatremia and thrombosis. The purpose of this study was to evaluate the side effects of DDAVP, namely hyponatremia and thrombosis, in patients with traumatic ICH who were prescribed pre-injury antiplatelet medications.

Data were collected on patients with traumatic ICH, who were prescribed pre-injury single (SAT) or dual (DAT) anti-platelet therapy and received single dose DDAVP (0.4 mcg/Kg). Data collected included serum Na before DDAVP (Na Pre), first serum Na after DDAVP (Na Post), lowest serum Na within 24 hours after DDAVP (Na Lowest), and thrombotic events (venous or arterial) within 5 days after DDAVP dose.

33 patients were included in the study. 26 patients (79 %) were on SAT and 7 (21 %) were on DAT. Compared to Na Pre (140 +/-6), Na Post (140 +/-6) and Na Lowest (139 +/-6) were not statistically different (p = 0.5, p = 1.0 respectively). One patient (3 %) developed DVT within 5 days after DDAVP. There were no other thrombotic events ( DVTs, MIs, ischemic strokes, or pulmonary emboli).

Single dose DDAVP given for traumatic ICH patients prescribed SAT or DAT was not associated with hyponatremia. Risk of DVT was low at 3 %. These results need to be evaluated in a randomized controlled trial to further elucidate the side effects of single dose DDAVP for this indication.

Poor bowel function is a risk for patients in the Neuro Critical Care Unit (NCC). Many medications are used to manage poor bowel function. We retrospectively compared scheduled once-daily polyethylene glycol 3350 (PEG) to scheduled twice-daily docusate sodium (DS) as part of an NCC bowel protocol. The lack of evidence to support the efficacy of DS in primary literature prompted a change in the bowel regimen in December of 2016 to improve patient bowel function and decrease constipation related complications.

We included inpatient neurosurgical patients age 18 or older who underwent a surgical procedure and spent at least 24 hours in the NCC. A total of 166 patients were included, 83 using docusate before the regimen change, and 83 using PEG after the change. Data collected included baseline demographics/characteristics, time-to-first bowel movement (primary outcome), use of rescue therapies, number of bowel movements, ileus, bowel perforation, and fecal management system (FMS) placement. Statistical analyses included time-to-event analysis (Kaplan Meir and Cox Regression), Poisson Regression (count data), and logistic regression (categorical outcomes). Differences between baseline characteristics were assessed using χ2 for categorical variables and Mann-Whitney for continuous variables.

Baseline characteristics were almost similar between groups. There was no statistical or clinically significant difference in the primary outcome. The odds of receiving rescue therapy were 71% less (95% CI 0.14 to 0.59; p=0.001) in PEG patients as compared to their counterparts. PEG patients also needed an FMS less often [OR 0.28, (95% CI 0.87 to 0.92; p=0.037)]. Those receiving PEG had a higher frequency of BMs than those receiving DS [OR 1.21, (95% CI 1.10 to 1.33; p<0.001)].

PEG did not decrease time-to-first bowel movement relative to DS but showed a statistically significant reduction in the use of rescue therapy and FMS placement, which we feel justifies the change in bowel regimen.

There is limited literature assessing dosing strategies of 4F-PCC as a reversal agent for DOACs; especially in patients with an intracranial hemorrhage. This study sought to compare low-dose (25 units/kg) versus high-dose (50 units/kg) 4F-PCC for the management of DOAC-associated intracranial hemorrhage.

This was a single-center, retrospective, observational study of adult patients admitted at a Level 1 Trauma Center who received 4F-PCC from August 1, 2015 to August 31, 2019 for DOACassociated intracranial hemorrhage. The primary outcome was the assessment of hematoma expansion within 48 hours of receiving 4F-PCC which we defined as an increase of 6ml or 33% of hematoma size. The secondary outcomes included thromboembolic events within 14 days, hospital length of stay, mortality and cost analysis.

Fifty-nine patients met our inclusion criteria with 32 patients (54%) in the low-dose group and 27 patients (46%) in the high-dose group. Hemostasis was achieved in 78.1% of patients receiving the lowdose and 74.1% of patients receiving the high-dose (p=0.716). No thromboembolic events were reported within 14 days after receiving 4F-PCC. Mortality occurred in 18.8% and 18.5% in the low-dose and high-dose groups, respectively. The median (IQR) hospital length of stay was 8.5 (3-15.5) days in the low-dose group and 6 (3-11) days in the high-dose group. The average cost difference between each dosing strategies is $2,749.

There was no statistically significant difference observed with achieving hemostasis between patients who received 25 units/kg versus 50 units/kg of 4F-PCC. No thrombotic events were reported and there were no observable differences with the hospital length of stay and mortality. However, there was a considerable cost difference between the two dosing strategies with potential cost savings of $170,082 if we used the lower dosing regimen. Therefore, our study suggests that low-dose 4F-PCC is costeffective with achieving hemostasis.

Dual antiplatelet therapy is used to prevent stent thrombosis in patients who receive intracranial stents, but the optimal dose of aspirin is unknown. A recent meta-analysis raised the question of whether the degree of platelet inhibition by aspirin is affected by bodyweight. This study sought to determine if this can be observed through a platelet reactivity assay.

This is a retrospective review of patients who underwent neurovascular stent placement and had a VerifyNow Aspirin assay result. The primary outcome was the correlation between VerifyNow Aspirin result, bodyweight, and initial dose of aspirin. Secondary outcomes included the impact of VerifyNow P2Y12 result and the impact of weight on the incidence of bleeding or a thrombotic event within 6 months.

Of the 142 included patients, 62.7% weighed ≥70 kg and 88.7% were initiated on aspirin 300-325 mg daily. 83.8% achieved a therapeutic ARU. There was minimal correlation between VerifyNow Aspirin result, bodyweight, and aspirin dose (R2=0.020). There was no difference in the mean ARU between patients who weighed <70 kg versus ≥70 kg (449 vs 435 ARU, p=0.32). There was no difference in the incidence of bleeding (28 vs 17.1%, p=0.14) or a thrombotic event (4 vs 5.3%, p=0.59) in patients weighing <70 kg and ≥70 kg, respectively. In a multivariate analysis that included VerifyNow Aspirin result, bodyweight, sex, and VerifyNow P2Y12 result, only the VerifyNow P2Y12 result predicted the development of either complication (p<0.01).

Bodyweight did not influence the likelihood of obtaining a therapeutic VerifyNow Aspirin result. Only the VerifyNow P2Y12 result predicted clinical outcomes. The clinical utility of obtaining VerifyNow Aspirin assays for this patient population is unknown.

Hypertonic saline (HTS) is used in management of cerebral edema of various etiologies in brain injured patients. Preliminary evidence suggests that use of HTS may be associated with acute kidney injury (AKI). Incidence and risk factors associated with AKI in patients receiving HTS are not well defined, which is what we aimed to assess in our study.

IRB-approved prospective, observational study of brain injured patients hospitalized in the NICU of Thomas Jefferson Hospital. We included adult patients (≥18 years) who received HTS, excluding those receiving renal replacement therapy prior to admission. AKI was defined using KDIGO criteria in the 1-14 days following admission. Analysis compared those with and without AKI. Continuous variables were analyzed using one-way analysis of variance and categorical variables were analyzed using Pearson's chisquared test.

Of the 100 patients enrolled over 5 months, AKI was observed in 19%. 

This retrospective cohort study enrolled 144 adult patients in the NSICU (n=72) or the MSICU (n=72) with a positive respiratory culture drawn at least 48 hours after admission. Patients culturing corynebacterium or rare quantitative growth were excluded. The primary outcome compared rates of ceftriaxone-resistance between the two ICU locations. Secondary outcomes identified independent risk factors for ceftriaxone-resistance (multivariate regression), and resistance patterns between ICU services (neurocritical care, medical ICU, surgical ICU). Statistics were performed using Sigmaplot for nominal variables (chi-squared) and continuous variables (t-test/Mann-Whitney test based on normality).

Ceftriaxone-resistance rates were identical between the two locations (39% v 39%, p=0.88). Patients admitted to neurocritical or surgical ICU teams were less likely to culture ceftriaxone-resistant organisms compared with the medical ICU team (36% v 28% v 55%, respectively, p=0.02). History of resistant infection (OR-9.8, 95% CI-1.9-49.3), chronic obstructive pulmonary disorder (OR-2.2, CI-1.1-4.7), and hospital length of stay > 5 days (OR-2.7, CI-1.2-6.2) were independently associated with resistance. A subgroup analysis of neurocritical care patients suggested ceftriaxone-resistance was significantly more common after 5 days of admission (n=64) compared to cultures drawn before day 5 (n=18), (48% v 16%, p=0.04).

Physical location had no impact on ceftriaxone-resistant nosocomial pneumonia. Incidence of ceftriaxone-resistance differed between admitting ICU teams. In neurocritical care patients, ceftriaxoneresistance is significantly more common 5 days after admission.

Tirofiban is a GPIIb/IIIa inhibitor extensively studied in cardiac intervention to decrease risk of arterial thrombosis. It has an expanding role in neurointervention, though the optimal dose in this population is unknown. Furthermore, tirofiban is renally eliminated and requires dose adjustment in renally-impaired patients (creatinine clearance (CrCl) below 60 mL/min). Major bleeding events (MBEs) can be a fatal side effect of antiplatelet medication after neurointervention, and understanding tirofiban dosing is critical for optimal management. This study was designed to determine the safety of two tirofiban dose strategies during neurointervention in patients with renal impairment.

This retrospective cohort study enrolled adults receiving tirofiban for elective or emergent neurointerventional procedures. The primary outcome compared MBEs between patients with renal impairment and patients with CrCl >60 mL/min. MBEs were defined using TIMI criteria which included new or expanding intracranial hemorrhage. Dosing was determined by the provider but an order panel suggested a standard dose (0.1 mcg/kg/min maintenance infusion) if CrCl>30 mL/min or a renal adjustment (0.05 mcg/kg/min maintenance infusion). Secondary outcomes analyzed MBEs by dose. Chisquared/Fisher's exact testing compared proportions using SigmaPlot.

Ninety-nine patients were enrolled in the study. Twenty-three patients had renal impairment and MBEs were more common in these patients (26.1% v 9.2%, p=0.086). Specifically, non-intracranial hemorrhage (ICH) MBEs were significantly more frequent in the renally-impaired group (17.4% v 1.3%, p=0.01).

Retrospectively, patients were adjudicated as receiving an appropriate dose or inappropriately high dose based on CrCl<60 mL/min using our dosing protocol. Of 17 patients receiving an inappropriately high dose, 5 experienced a MBE. This was more common but not significantly higher than appropriatelydosed patients (29.4% v 11.0%, p=0.062).

A higher dose of tirofiban was associated with greater non-ICH MBEs in renally-impaired patients. The optimal dose is unknown, however, a maintenance infusion of 0.05 mcg/kg/min may reduce risk of MBEs.

Management of pain in Neuro Critical Care is complex. Opioids are often initiated as a first-line approach to controlling pain, potentially compromising critical neurological assessment. Opportunities exist for pre-emptive, non-narcotic strategies to improve pain while limiting risk of overuse and adverse effects associated with opioids.

Prospective observational study in non-ventilated, closed Neuro ICU patients receiving opioids at Comprehensive Stroke Center (N=1685). A pre-intervention survey was administered to assess staff and physician knowledge regarding safe and adequate pain management. An anxiety scale for patient was assessed on admission and daily. Baseline data was collected. Pain was evaluated using 0-10 pain scale, delirium prevalence was assessed using CAM-ICU, total opioid was calculated using Morphine Milligram Equivalents (MME), and data was reviewed for opioid related morbidities. Integrative interventions are implemented: Aromatherapy, Music therapy, Therapeutic Mindfulness Practice, and Movement with physical therapy as adjunct modalities to current standard of care.

Baseline data suggests overall 80.2% of patients spend their time in mild pain category (0-3), required an average of 11.8 MMEs (Morphine Milligram Equivalents) during their average 3.25 days in the Neuro ICU, and 22.6% screened positive for delirium during the hospitalization. When patient pain scores were greater than 3, MMEs use increase (154 vs 13 mg), foley days increase (2.3 vs 1.6 days), required additional bowel treatment (31% vs 16%), and dexamethasone usage increased (21 vs 9 mg).

Study is ongoing; however, baseline data suggests that studied non-ventilated Neuro Critical Care patients reported mild pain scores (80.2%) with 22.6% of positive for delirium. Patients with pain score of >3, had 12 fold increases in MMEs, 0.5 additional days of Foley, double the use of dexamethasone, and required additional bowel treatment. Further intervention using integrative modalities adjunction to current standard of care are currently in place.

Neuroendocrine dysfunction (NED) has been reported in the literature as a complication following blunt traumatic brain injury (TBI). Despite the prevalence of penetrating brain injury (PBI), data on NED in the acute setting of PBI is scarce, and the current clinical approach is extrapolated from literature on closed TBI.

This systematic review is presented in accordance with the PRISMA guidelines. PubMed, Scopus, and Cochrane were searched. Studies were narrowed to human studies in adults. Risk of bias was computed using the Newcastle-Ottawa Scale (NOS), or the Methodological quality and synthesis of case series and case reports for case reports, as indicated.

421 studies were screened, six relevant studies with 481 total subjects were included. Two studies were prospective cohort analyses, and four articles were case reports. The onset of NED was acute in all studies, by post-injury day one. Risk factors for NED were minimally reported but included worse injury severity and cerebral edema on imaging. Anterior pituitary dysfunction was diagnosed through serum hormonal level profile. The most common hormonal deficiency was thyroid and gonadal hormones, treated with hormone replacement, followed by hypocortisolism, treated with hydrocortisone. The prevalence of central diabetes insipidus (DI) was up to 41% in one study and was treated with DDAVP. Most patients demonstrated neuroendocrine dysfunction months after injury. In separate reports, DI and hypocortisolism showed an association with higher mortality. The available literature for this review is poor, and the studies included had overall moderate quality with moderate risk of bias.

NED appears to be a prevalent in the acute phase of PBI. Despite a potential association between post-PBI NED and mortality, data on the optimal management of NED is limited. This defines the need for prospective studies to better characterize the clinical features and optimal therapeutic interventions for NED in PBI.

Cerebral venous thrombosis (CVT) is an uncommon but life-threatening neurological condition that disproportionately afflicts younger patients and females. Patients with CVT have traditionally been treated with unfractionated heparin infusion (UFH) or low molecular weight heparin (LMWH) as a bridge to anticoagulation with warfarin since the establishment of UFH efficacy in a randomized clinical trial in 1991. There is paucity of data for the use of apixaban, factor Xa inhibitor, for this patient population.

We performed electronic chart review and patient interview to collect data for case series of 8 patients with CVT who were anticoagulated initially with UFH infusion followed by long-term anticoagulation with apixaban at an academic university hospital with a specialized Neurocritical Care Unit.

Patients (n = 8, female = 6, age 40 [range 27 -60]) ultimately treated with apixaban presented with typical CVT symptoms. One patient had history of CVT and was already on anticoagulation with rivaroxaban . The diagnosis was confirmed with imaging, which also showed venous infarct and/or intracranial hemorrhage in most of these patients. Almost all patients were initially anticoagulated with UFH infusion. They were all transitioned to apixaban prior to discharge. On long-term follow up (minimum 3, maximum18 months), 88% (n=7) patients were on apixaban without any reported adverse effects or progression of CVT on follow-up imaging. Among four patients with follow-up imaging, three patients (75%) demonstrated partial or complete recanalization (partial n = 1 [25%], complete n = 2 [50%]). The rest four patients still have pending follow-up imaging. None of the patients had any bleeding or thromboembolic complications.

In our series of patients with CVT receiving apixaban met safety outcomes including lack of recurrent thromboembolic events or bleeding complications. Larger randomized trials are warranted to evaluate further safety and efficacy of apixaban in this patient population.

The Coronavirus Disease 2019 (COVID-19) is a global pandemic affecting multiple organ systems, including the brain. Neuroimaging of patients who are in a COVID-19 intensive care unit (ICU) may be difficult to acquire given the safety concerns and challenges involved in moving these critically ill patients. We report on the safety and clinical findings of a portable magnetic resonance imaging (MRI) scanner in a cohort of ICU patients with suspected neurologic injury from COVID-19.

This is a prospective, non-randomized, observational study at one institution utilizing portable MRI in patients with laboratory confirmed severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. Patients selected for imaging had any of the following: 1) unexplained encephalopathy or coma, 2) seizures, 3) focal neurologic deficit, and 4) abnormal head CT. Imaging was performed in each patient's ICU room with a portable, self-shielding, 0·064 Tesla (T) MRI.

Among 19 patients, a total of 20 MRI scans in seven ICUs were acquired between April 13 and April 23, 2020. No adverse events to patients or staff from MRI acquisition were reported. In 12 patients, abnormal findings were seen, which included increased fluid attenuated inversion recovery (FLAIR) signal (n=12), hemorrhage (n=3), and diffusion-weighted imaging (DWI) positivity (n=3). Imaging led to a change in clinical management in five patients.

In critically ill COVID-19 patients, the use of portable MRI is safe, feasible, and leads to changes in clinical management. This technique can be applied to any ICU patients whose care requires imaging of the brain.

This pilot study explored whether the less invasive NIRS FORE-SIGHT absolute cerebral oximeter, used for measuring brain oxygenation, was accurate compared to the standard Licox monitor in a neurocritical ICU population. Could it be reliably used for clinical decisions: set a priori at least 90% sensitivity and 70% specificity.

Patients, aged 18-80, with a neurological diagnosis (stroke, SAH, TBI) and a GCS<8 who clinically required a Licox monitor from 6/1/17-10/31/19 were enrolled with an IRB waiver of consent. Following insertion of Licox, two NIRS sensors were placed bilaterally. Data from each system was collected hourly for 24 hours. Simultaneous NIRS and Licox readings were compared to calculate sensitivity and specificity of NIRS predicting if Licox indicated oxygen insufficiency (<15). NIRS was analyzed separately for higher/lower, injured/contralateral side, with cut points at 60 (as used for cardiac patients), 70 and 75. Data for both Licox and NIRS were aggregated over all readings for each patient.

24 patients included with up to 2184 measurement pairs. The lower NIRS value showed the best sensitivity with 8.6% using 60 as cut point, rising to 47.9 for 70 and 67.5 for 75; specificity declined from 84.6 to 38.4 to 11.9 respectively. In the aggregated data, only 13 of the 24 had any Licox reading below 15; that was predicted by any NIRS value on contralateral side under 70 with 100% sensitivity and 30% specificity.

The results offer no support for using NIRS as a less invasive substitute for Licox. However, the aggregated data findings raise another question: Can the less invasive NIRS be used as a pre-screen for whether Licox is required? Looking only at NIRS before the first low Licox the same contralateral measure retains good sensitivity at 87.5. More investigation with a larger sample is needed.

Practicing Neurocritical care remains a challenge during the Coronavirus Disease 2019 pandemic. With limited Personal protective equipments (PPE) and drastic increase in patient numbers, recording Electroencephalograms (EEG) on confirmed or suspected patients was challenging while trying to ensure staff safety and avoiding contamination of EEG equipment.

We performed a prospective cohort of EEGs using a new FDA cleared technology (ZETO) as the first dry electrodes EEG helmet, which is cloud-based and wireless. Also does not require a technologist for placement. This is the first time ZETO is used and tested in a large inpatient cohort. The goals of this study is to evaluate the feasibility and efficacy of ZETO to reduce EEG placement time by > 90% and ability to record > 90% of EEGs ordered in the ICU, ER and inpatient units with at least good quality score > = 3 on a predefined score of (1-5); 1 is uninterpretable EEG and 5 is excellent recording with minimal artifacts without using the artifact reduction software.

We recorded 20 EEGs out of planned 100 total for the trial duration. 3 EEGs were repeated duo to lack of ability to place the helmet, significant artifacts, or patient extreme agitation. We did notice that there is a learning curve for EEG techs and providers using this system, and those who used it more were able to achieve better studies. The average time from arrival to patient room to the view of the first readable EEG page is 6.5 minutes. 53% of those EEG successfully recorded had a quality rating of 4 or better.

ZETO EEG is a feasible, effective and a quick innovative technology that can be implemented in the inpatient setting and it's a very useful tool to provide neurological care for suspected or confirmed COVID-19 patients.

The neurological exam is crucial when evaluating patients who have sustained a neurological injury. However, the exam may be difficult to interpret in patients placed in a pharmacologically induced coma, as signs of early clinical deterioration can be masked by medications such as pentobarbital due to suppression of cranial nerve function and spinal reflexes. The use of pupillary reactivity has been vital in evaluation of such patients. Automated pupillometry and use of the Neurological Pupil Index (NPI) is shown to increase sensitivity and consistency when compared to neurological examination alone. It remains unknown if pupillary reactivity is: 1) preserved at various doses of pentobarbital or 2) affected by the depth of coma as assessed with continuous EEG patterns.

We performed a retrospective chart review and gathered pupillometry and EEG data on 3 consecutive patients in a pentobarbital coma admitted to a Neuroscience Intensive Care Unit at an academic center. While in the pentobarbital-induced coma, the patients were monitored with serial neurological exams, hourly pupillometry, and continuous EEG for the duration of the pentobarbital infusion.

We found that the pupillary reactivity, as assessed by NPI, 1) is preserved, though diminished, at different pentobarbital doses and 2) persists even in the presence of near-isoelectric EEG patterns. 3) There was a temporary cessation of the pupillary light reflex after loading doses of pentobarbital.

Pupillary reactivity is a robust part of the neurological exam and independent of pharmacological influences. Automated pupillometry improves researchers' ability to correlate the temporal relationships among pupillary reactivity, physiologic parameters, medication administration, and catastrophic neurological conditions.

Cerebral edema is a common cause of mortality in acute liver failure (ALF). Because invasive monitors are often precluded by comorbid coagulopathy, we instead use continuous EEG (cEEG) and pupillometry to monitor for neurological deterioration. We have often observed transient attenuations in the normal pupillary index (NPI) in patients with hepatic encephalopathy (HE). Here, we aim to characterize pathologic changes in cEEG and pupillometry as non-invasive tools to guide management in ALF.

ALF patients with at least 12 contiguous hours of cEEG and hourly NPI measurements were analyzed from a single-center retrospective observational cohort. Pupilometer readings were dichotomized to high and low NPI responsiveness, with a threshold NPI of 3.5. A threshold of 20% of total NPI recordings below 3.5 was used as a cutoff for group comparison. Fischer exact tests were used to compare patients with moderate to severe slowing on cEEG versus those with electrographic burst suppression.

A total of 12 patients were included, with acetaminophen toxicity causing ALF in 46% of cases. Rhythmic periodic patterns were common, but no seizures were observed in any patients. 8 patients had low NPI responsiveness in <20% of recordings. In these patients, cEEG was notable for variable levels of slowing, none demonstrating electrographic burst suppression. Conversely, in the 4 patients with >20% low NPI responsiveness, all had evidence of electrographic burst suppression, none of whom were induced with high-dose anesthetics or cerebral edema. The increased frequency of low NPI responsiveness appeared to correlate with the occurrence of burst suppression (p<0.05).

Frequent (>20%) NPI attenuation correlates with the presence of burst suppression on cEEG in patients with ALF. In the absence of structural abnormalities or pharmacologic sedation, these changes likely reflect metabolic suppression from HE. These findings demonstrate the potential utility of NPI variability as a quantitative approach to monitor HE severity.

EEG monitoring provides crucial information in the management of patients with suspected status epilepticus. However, their availability is limited particularly in community hospitals. The lack of EEG capability may necessitate transfer of patients to a higher level of care. Here, we describe the clinical and financial impact of deploying a portable EEG (Rapid Response EEG) in the community hospital setting.

We retrospectively chart reviewed the use of the Rapid Response EEG device in a community hospital setting. EEG recordings were performed in the emergency department and ICU when clinically indicated.

Outcomes included length-of-stay, the detection of seizures or status, disposition and resource utilization, and transfer to the affiliated flagship hospital. The financial impact was modeled for various scenarios using an estimated cost of hospitalization of $3000-6000/day and transfer cost of $4000-20,000.

Rapid Response EEG was used in 75 patients of which 34 (45%) patients presented with a clinical event concerning for a seizure. Transfer to main tertiary hospital occurred in 4 cases (5%), more frequently among patients found to have seizure (14%) or highly epileptiform, patterns (13%) compared to slow activity (4%) or burst suppression (0%). Ceribell avoided transfer for EEG monitoring in 31% of patients. We estimate the cost of Rapid Response EEG infrastructure and disposables at <$30,000, compared to ~$100,000-140,000 for conventional EEG systems and ~$420,000-562,000 for EEG technologists. Our preliminary analysis of forty patients revealed an estimated cost savings of $20,350 per patient and $814,000 per year compared to transferring a patient for conventional EEG monitoring.

Rapid Response EEG System increased access to EEG among high-acuity patients in the community hospital setting. The result was improved resource utilization and relatively few transfers to the flagship hospital resulting in significant cost savings.

EVD insertion is the most commonly performed neurosurgical procedure in the neuroscience ICU. It is used to provide real-time intracranial pressure monitoring and CSF diversion, most frequently employed during the treatment of intracranial hemorrhage. Although the procedure is considered relatively safe, complications such as hemorrhage and infections do occur. The rate of EVD placement associated hemorrhage is reported in the range from 10-44%, and recent publications have suggested possible risk factors associated with higher risks of hemorrhage, such as age, lower platelets count, and antiplatelet agents use within 96 hours prior to admission. However, there is a paucity of data published regarding the effects of platelet transfusion prior to EVD placement.

A retrospective review was performed on patients who were admitted to the neuroscience ICU with intracerebral hemorrhage and EVD placement from Jan 1st, 2014 to Dec 31st, 2019, at our tertiaryreferral hospital.

After screening, 299 patients were enrolled. Eighteen out of 47 patients (38.3%) who received platelet transfusion prior to EVD insertion developed EVD-related hemorrhage versus 84 out of 251 patients (33.5%) who did not receive platelet transfusion (P=0.51). Ninety-three patients took antiplatelet agents before admission, and 38 (40.9%) developed EVD hemorrhage versus 60 out of 180 patients (33.3%) who did not take antiplatelet agents before admission (P=0.22). There was no statistically significant difference in EVD hemorrhage rates between subarachnoid hemorrhage (37.1%) and intracerebral hemorrhage (29.2%, P=0.16). The relationship between EVD placement scores and hemorrhage rates did not reach statistical significance either (P=0.053).

In our retrospective study, platelet transfusion prior to EVD placement does not reduce the risk of EVDrelated hemorrhage. The accuracy of EVD placement does not predict the chance of hemorrhage either. Like other studies, antiplatelet agents seemed to increase the chance of EVD hemorrhage, though not statistically significant in our study.

There are no evidence-based resources guiding the perioperative management of Moyamoya patients undergoing revascularization surgery. We investigated practice patterns among a transdisciplinary group aiming at identifying possible heterogeneity of practices on key components of care that warrant prospective studies.

We administered a web-based survey comprising 62 questions pertaining to demographics of respondents and several aspects of pre-, intra-, and post-operative care to physician members of the Neurocritical Care Society, Society of Critical Care Medicine, American Academy of Neurology, Society for Neuroscience in Anesthesiology and Critical Care, and American Association of Neurological Surgeons.

reliminary analysis of 79 responses from 7 countries (90% US) captured input from overlapping disciplines [anesthesiology (17.4%), critical care medicine (27.5%), neurology (62.3%), neurosurgery (18.8%)], mainly from experienced providers (51.6% > 5 years beyond training) practicing in academic centers (86.4%). Arterial line for continuous systolic blood pressure (SBP) and mean arterial pressure (MAP) recording is the most commonly used method of monitoring perioperatively. Significant variability exists on selected level for transducing pressure: phlebostatic axis/right atrium is most commonly used in the pre-(59.4%) and post-operative (52.9%) phases, whereas responses were split between circle of Willis/tragus (35.7%) and phlebostatic axis (32.1%) intra-operatively. Neurosurgeon preference is the main contributor to the selection of perioperative hemodynamic goals. The median minimum and maximum intra-and post-operative SBP and MAP goals were 100-140 mmHg and 70-100 mmHg, respectively. Crystalloid infusion was the preferred method to achieve hemodynamic goals (93.3%), followed by phenylephrine and norepinephrine tied at 68.9%; however, these vasopressors are considered contraindicated in this setting by 14.9% of respondents.

Our results illustrate great heterogeneity of Moyamoya perioperative practices among different stakeholders. Surgeons' preference is the main driver of hemodynamic goals and heterogeneous hemodynamic augmentation practices is evident, which constitute equipoise for prospective studies targeting optimal hemodynamic management strategies.

This study was designed to determine patient and disease-related factors known proximate to the presentation of SE associated with developing refractory status epilepticus (RSE).

This was a propensity-matched observational case control study using age and sex matched pediatric patients with prolonged convulsive seizures and RSE. These matched sub-populations were obtained from patients enrolled in the pediatric Status Epilepticus Research Group (pSERG) study. Patients were between 1 month -21 years who had convulsive seizures at onset. The prolonged convulsive status epilepticus (non-RSE) group were those requiring only benzodiazepines and a single second-line antiseizure medication (ASM) to stop their seizures. The RSE group were patients requiring more than benzodiazepines and a single second-line ASM to treat their seizures. Variables that could be acquired early after presentation were explored with univariate analysis of the raw data. Variables with p<0.1 were retained for multivariable analysis. Multivariable logistic models were fitted to the age and sex matched data to obtain variables associated with RSE.

We enrolled 398 patients in the RSE group and 197 patients in the non-RSE group. Univariate analysis demonstrated no differences in the age (RSE 4.3 years; non-RSE 4.6 years; p=0.59) and sex (RSE and non-RSE groups 56% male) of the patients. On multivariate analysis, a family history of seizures (OR 0.37; 95% CI 0.20 -0.70, p = 0.0022) and a prescription for rectal diazepam (OR 0.21; 95% CI 0.078 -0.53, p = 0.0012) were associated with decreased odds of RSE, while intermittent seizures increased the odds of having RSE (OR 3.16; 95% CI 1.2 -8.5, p = 0.022).

This study shows that having a family history of seizures and having a prescription for rectal diazepam are associated with decreased odds of RSE, while an intermittent seizure phenotype is associated with 3fold increase odds for developing RSE.

MRI is commonly obtained acutely after pediatric TBI but its prognostic value remains unclear. Diffuse axonal injury (DAI) grade predicts functional outcome among adults with TBI, but studies on pediatric patients remain limited. We conducted a retrospective observational study of early neuroimaging in children with TBI to determine if Rotterdam CT score and DAI grade correlate with functional outcome.

Children with TBI were identified through retrospective chart review from two regional level one trauma centers in Northern California. Patients who underwent both CT and MRI obtained during acute hospitalization were included. Rotterdam CT scores and DAI grade were calculated. Primary outcome was long term functional outcome at 6-12 months after TBI based on the Pediatric Cerebral Performance Category Scale (PCPCS). Secondary outcome was hospital discharge disposition (home versus inpatient rehabilitation).

Of 123 total patients with TBI, 53 (43%) were classified as severe TBI, defined as the initial GCS score less than 9. The most common mechanism of injury was motor vehicle collision (52%). Initial head CT showed acute hemorrhage in 69% of patients, with a mean Rotterdam CT score of 2. MRI was acquired at 5 +/-8 days after injury. Hemorrhagic DAI lesions were detected in 69% of all patients. Moderate and severe TBI patients with unfavorable primary outcome were more likely to have a lower GCS score, have longer intensive care unit (ICU) length of stay, higher Rotterdam CT Score, and higher DAI grade. GCS score and DAI grade were independent predictors for both primary and secondary outcomes.

Higher DAI Grade correlates with worse short term and long term outcomes in children with acute TBI. Depth of lesion analysis of MRI may be useful in advising health care providers and families regarding need for acute rehabilitation and probability of long term neurological sequelae in this population.

Traumatic brain injury (TBI) is a leading cause of death and disability in pediatrics. Limited data exits regarding the presence of multiple organ dysfunction syndrome (MODS) and its impact on mortality in this population. We sought to assess the occurrence of MODS in pediatric patients with severe TBI and investigate its association with mortality.

Retrospective observational cohort study of pediatric patients with a severe TBI (GCS≤8) 

In this cohort, non-survivors were more likely to have MODS particularly with higher rates of cardiovascular, respiratory and hematologic failure. When more specific criteria were utilized to better examine the occurrence of respiratory and renal failure, non-survivors were more likely to develop PARDS and AKI. PARDS was independently associated with increased mortality.

Phoenix Children's Hospital, Department of Neurology, Phoenix, AZ, United States

Post-traumatic seizures (PTS) after severe traumatic brain injury have been associated with poor outcome. Early recognition that PTS is about to occur would aid in clinical management. The goal of this study was to implement a machine learning algorithm to predict PTS from continuous electroencephalography (EEG) data before they occur.

The ability to retrospectively predict PTS events was assessed from continuous EEG data in 4 pediatric patients with severe TBI. To do this, for each patient, the frequency power (0-19 Hz, 1 Hz bins) of 15-26 EEG channels was extracted as input time series (i.e. variates) to a single-layer long short-term memory (LSTM) artificial neural network with a dual regional attention model. The role of the LSTM model is to discover predictive temporal patterns, whereas the regional attention improves model accuracy in sparse data by focusing the learning process on parts of the data that are likely to be relevant. In order to support the attention mechanism, we extracted robust temporal features in a data driven manner by considering the temporal characteristics of the EEG frequency content. The accuracy to predict PTS events 5.1 minutes before onset was quantified for each patient with a 2x5-fold cross-validation.

Across the 4 patients, 20 PTS events were detected, of which 94±13% were correctly predicted. Conversely, false prediction of PTS occurred only 0.03±0.05% of non-seizure times. Results have shown that the proposed dual regional attention technique significantly improves predictive accuracy of the artificial neural network based model.

Using individualized models, we were able to predict PTS from continuous EEG 5 minutes before onset in 4 children with PTS using a LSTM with a regional attentional model. Future work will explore this algorithm in a larger cohort and investigate whether additional multimodal neurologic monitoring data improves prediction.

University of Colorado School of Medicine -Child Neurology, Aurora, CO, United States

Introduction FIRES (Febrile Infection-Related Epilepsy Syndrome) is a subcategory of new-onset refractory status epilepticus (NORSE) that occurs in children with preceding febrile illness 1 . The acute refractory status epilepticus in children with FIRES is followed by chronic refractory epilepsy and cognitive deficits in nearly all patients1-10. Recent research suggests a relation to increased inflammatory cytokines for some patients,3,5,8-9 prompting acute treatment with the IL-1 inhibitor anakinra. However, long-term neurocognitive outcomes in patients treated with anakinra have not been examined. We aim to describe the long-term neurocognitive outcomes in a cohort of patients with FIRES treated acutely with anakinra.

Retrospective case series of two children with FIRES treated acutely with anakinra at 14 days of illness with > 2 years follow-up, including description of standard clinical neuropsychological evaluations. 

Despite improvement in seizures acutely, long-term cognitive outcomes of our patients were similar to those described in the literature 2,7,10. Patients had intellectual/learning disabilities but were able to engage in some academic activities. Currently collecting data from other institutions for larger case series to better understand determinants of long-term outcome in FIRES patients treated with anakinra.

Spreading depolarizations (SDs) are associated with subarachnoid hemorrhage (SAH) and Traumatic Brain Injury (TBI) outcomes. We sought to determine the association between clinical measures on continuous EEG (cEEG) and intracranially quantified SDs.

We included SAH and TBI patients undergoing >24 hours of interpretable intracranial monitoring via 8contact depth or 6-contact subdural strip platinum electrodes, or both. DC or near-DC electrocorticography (ECoG) recordings acquired using the CNS Monitor (Moberg, Inc.) were reviewed in bipolar and reference montages superimposing high-pass (0.5-30 Hz) and low-pass (0.005-0.5 Hz) activity. SDs were rated according to established consensus criteria and compared to synchronized cEEG rated according to ACNS Critical Care EEG Monitoring Consensus Criteria for: lateralized rhythmic delta activity (LRDA), generalized rhythmic delta activity (GRDA), lateralized periodic discharges (LPDs), generalized periodic discharges (GPDs), any IICA, or the composite 2HELPS2B score. Among SAH patients, cEEG was ascertained for validated delayed cerebral ischemia (DCI) biomarkers: new or worsening epileptiform abnormalities (new/worsening EAs) and new background deterioration (BD).

35 of 37 monitored patients met eligibility criteria (20 post-SAH including 5 with strip electrodes; 15 post-TBI including 5 with strip electrodes). 5 (25%) of 20 SAH and 1 (6.66%) of 15 TBI patients developed SDs, including 4 (40%) of 10 recorded with strip and 5 (17.24%) of 29 with depth electrodes. There was no significant association between occurrence of SDs and day 1 ACNS scalp cEEG main terms (LPD, GPD, LRDA or seizure activity, individually or in combination), or cEEG biomarkers of DCI among post-SAH patients.

In this single-center cohort, there was no significant association between SDs and scalp EEG IICA measured in clinical practice. Limitations include sample size and overall SD rates. Identifying scalp EEG correlates of SD may benefit from composite clinical features including power or machine-learning derived tools quantitative cEEG features.

While treatments to prevent human posttraumatic epilepsy (PTE), or reduce its severity, are lacking, evidence suggests SV2A protein as a therapeutic target. Levetiracetam binds SV2A and demonstrates anti-inflammatory, antioxidant, neuroprotective and antiepileptogenic effects in brain-injury and seizure/epilepsy models, and a phase 2 clinical trial signaled possible antiepileptogenesis. Brivaracetam shows much greater affinity and specificity for SV2A, binding and modulating it differently from levetiracetam. It also possesses a broader preclinical antiepileptic spectrum, and a better safety profile for head injury patients. Thus, we assessed brivaracetam's antiepileptogenic potential in rats using an etiologically realistic PTE model.

We used the optimized fluid percussion injury (FPI) model in young (5-week-old) male Sprague-Dawley rats, for which detailed power analyses are available to guide its use. Brivaracetam was delivered with dosing protocols designed to maintain plasma levels of 3-5 ug/ml for four weeks. Different doses and latencies to treatment were rapidly screened (62 animals) with small group sizes, in short 1-month-long protocols, to eliminate futile treatment protocols. The antiepileptogenic effect was measured in a blind, randomized, 16-week-long validation study testing four weeks of treatment initiated 30min (n=18), 4hr (n=18) and 8hr (n=22) after FPI, compared with vehicle-treated FPIs (n=20). Perilesional seizures and their spread were sensitively detected by video-EEG using a 5-electrode epidural montage. Endpoint measures included incidence, frequency, duration and spread of seizures.

Twelve weeks after treatment ended, treatment starting 4hr post-FPI (the best-performing protocol) induced decreases in seizure incidence (38%; p=0.05), seizure frequency (59%; p=0.01), time spent seizing (67%; p=0.005), and proportion of spreading seizures (45%; p=0.02), that were all independent from duration-based seizure definitions. Statistics included one-tail Fisher's exact and Mann-Whitney tests.

Brivaracetam persistently reduces the incidence and frequency of seizures, the time spent seizing, and further modifies PTE by decreasing the spread of focal seizures from the neocortical focus.

A previous study has shown the efficacy of scalp electroencephalogram (EEG) with a time constant of 2 s in detecting direct current (DC) shifts. However, its feasibility in the field of neurocritical care has not been assessed. Thus, this study aimed to examine the use of scalp EEG in a neuro intensive care unit (ICU).

This is a retrospective study conducted in a community teaching hospital from January 2019 to December 2019. All consecutive patients who underwent continuous EEG monitoring in the neuro ICU were included in our study. Patients diagnosed with status epilepticus (SE) were finally included in the analysis, and the data recorded with scalp EEG with a frequency filter of 0.08 Hz (time constant of 2 s) using the international 10-20 system were analyzed.

In total, 391 patients were admitted in the neuro ICU. Of these patients, 96 (24.6%) underwent continuous EEG monitoring. Moreover, 18 (18.8%) were diagnosed with SE, of whom six presented with convulsive SE (CSE), three with nonconvulsive SE (NCSE), and nine with both CSE and NCSE. The EEG data of 18 patients were then examined, and five patients were considered as candidates. However, there were no available methods that can accurately identify whether slow waves formed artifacts in four patients. Ensuring artifact-free condition, a patient with post-cardiac arrest who is taking neuromuscular blocking agents, presented with passive DC shifts at C3 and P3 with high-frequency oscillations.

DC shifts were detected in a case who is taking neuromuscular blocking agents using scalp EEG with a time constant of 2 s. However, further studies must be conducted to assess the feasibility of this procedure.

Serotonin Syndrome, a constellation of symptoms described by Hunter's criteria is often triggered by the addition of opioids to a patient who was previously stable on a serotonergic agent. We report a rare case of Serotonin Syndrome manifesting with decerebrate posturing associated with two distinct video EEG patterns. Previous studies have reported persistent rhythmic slow (theta) wave activity with occasional alpha waves on the EEG of a patient with serotonin syndrome. Another study on rats found different EEG patterns correlating with mild and severe syndrome as reduced amplitudes and seizurelike activity, respectively.

A 73-year-old Caucasian male with chronic pain syndrome on fentanyl patches came to the Emergency Room with opiate withdrawal. He had no refills for five days prior to presentation. Patient was given diazepam, diphenhydramine, hydromorphone, lorazepam, and haloperidol for agitation which led to respiratory failure and intubation. He was admitted to ICU and sedated on propofol and fentanyl. Vital signs were significant for fever and hypertension. On physical exam, brisk reflexes, bilateral clonus, and up-going toes on the right were present.

Labs were significant for low GFR and elevated CPK. He had unremarkable CT head and MRI brain. Video EEG monitoring revealed a rhythmic mid-range theta diffuse and intermittent pattern and a flat background with a diffuse 1 hertz low-voltage delta equal in both hemispheres. He had decerebrate posturing tonic flexion of his legs and extension of both arms without internal or external rotation at the transition between these 2 patterns.

Propofol was switched to Dexmedetomidine and Serotonin Syndrome was diagnosed. 48 hours after discontinuation of all the offending agents, his symptoms improved, he was extubated and discharged. Though further studies are needed, this shows the unique correlation of involuntary abnormal movements and EEG patterns in the use of Video EEG in evaluating a patient with Serotonin Syndrome.

The NeuroICU is a multidisciplinary specialty for the complex management of status epilepticus (SE) patients on antiepileptic drugs (AED). There is a growing amount of literature investigating pharmacogenomics (PGx) adverse drug reactions (ADR) and potential drug-drug interactions (DDI), including Stevens-Johnson syndrome (SJS), toxic-epidermal necrolysis (TEN) and DRESS. We describe an illustrative case and review of literature.

Case Report and Literature review among known AED, PGx, ADR, and DDI in the Neuro/ICU population.

A 33-year-old female developed a progressive increase in her generalized tonic-clonic seizures. Her local neurologist increased her levetiracetam from 500mg BID to 1g BID, and 1.5g BID over 8 weeks. At each increase the family noted new psychiatric manifestations including agitation, violence, and she was eventually arrested by police. The patient developed status epilepticus and was later intubated, sedated, and transferred to our NeuroICU for continuous EEG. The patient developed fever and developed a diffuse maculopapular rash. Due to suspicion of HLA 11:01, LVM was discontinued and placed on lacosamide and lamotrigine (LMT) after a focused PGx panel revealed no genes prone to LMT-SJS. The patient was extubated and later discharged home. Since 1969, 172 articles resulted from search terms "pharmacogenomics", "antiepileptic drugs", and "drug reaction." Only 17 high quality review articles discussed severe ADR of SJS, TEN, DRESS and neuropsychiatric manifestations of levetiracetam. Ethnic diversity was best described for carbamazepine for HLA-B15:02 (Han Chinese) and HLA-A31:01 (Europeans and Koreans) alleles. HLA 11:01 described for levetiracetam, with valproate being a common ADR/DDI drug and aromatic AEDs.

AEDs used in the NeuroICU have common, ethnically diverse, and major potentially unnoticed reactions that may masquerade as neurogenic fever, non-specific rashes, or agitated delirium. We plan to create a multicenter PGx registry among NeuroICU teams to help determine larger prevalence of severe ADRs and DDIs.

Current guidelines recommend introduction of traditional anesthetic medications like propofol or benzodiazepine for both generalized and focal forms of refractory status epilepticus (RSE). These patients require prolonged mechanical ventilation and secondary complications like hypotension and infections tend to cause poor outcome. Subanesthetic dosages of ketamine can be an effective therapy for focal RSE without need of mechanical ventilation.

The subanesthetic ketamine infusion protocol was defined as 0.25 mg/kg bolus dosing followed by infusion starting at 2mcg/kg/min with titration allowed up to 15 mcg/kg/min. Focal RSE was defined as failures to respond to at least 2 antiepileptic drugs. Non-intubated adult patients diagnosed with focal RSE between 10/1/2019 and 03/31/2020 were treated with ketamine infusion per protocol. A retrospective case-series review was performed upon institutional review board approval.

Seven patients with focal RSE were treated with ketamine infusion. Six of them had prior history of epilepsy. Mean age was 68.5 years and 4/7 were males. The median (Md) time from onset of seizures to initiation of ketamine was 33.5 hours (range 25-49, n=6). All patients were on multiple AEDs before ketamine (range 2-4, Md=3). Only 3 of 7 patients required titration of maintenance infusion to 15 mcg/kg/min, maximum rate pre-determined in the protocol. Six patients (86%, n=7) had resolution of status epilepticus within 24 hours of ketamine initiation without any recurrence. Patients received ketamine infusion ranging 24-130 hours (Md=70h). None of the patients required mechanical ventilation during ketamine therapy and most common side effects observed were hypertension and urinary retention.

Our study is the first to report utility and safety of subanesthetic dosages of ketamine for treatment of focal RSE. Ketamine can be a preferred agent over benzodiazepines with its high efficacy and added benefit of preserved respiratory drive. A randomized controlled trial is needed to test this hypothesis in larger population.

Formal interpretation of electroencephalography (EEG) in acute-care units is often performed by trained experts, including neurophysiologists and epileptologists. Owing to the rapid expansion of EEG, delays in formal interpretation are often encountered, with possible implications for patient management. Numerous studies have investigated the role of educational programs to enable non-experts (e.g. nurses and residents) to recognize EEG patterns at the bedside. However, the overall content, structure, duration, and efficacy of these EEG training programs remains unknown.

We conducted a systematic review of EEG educational programs for non-experts in adult and pediatric/neonatal acute-care settings (emergency departments (ED) and intensive care units (ICU). Our primary outcome was to describe important similarities, differences, and overarching themes among training programs. We searched for studies published on MEDLINE, Embase, Cochrane central, CINAHL, and Web of science. To be considered for inclusion, studies were required to present sufficient details of their educational program, including (to the extent possible) content covered, program structure and duration, assessment methods, and trainee experience. Randomized control trials, cohort studies, and descriptive studies were all considered for inclusion. Data was presented in a qualitative manner and organized by theme.

Our search yielded a total of 7,034 studies, of which 5,626 were screened. Twenty-six full-length studies met our predefined inclusion criteria. Studies were all published in English and included 23 cohort studies, two descriptive studies, and one RCT. The majority of studies were single center in design and originated in the USA. One study was performed in the ED, 15 in adult ICUs, four in pediatric ICUs, and six in neonatal ICUs. The majority of EEG training programs were geared towards ICU nurses (16 studies), followed by ICU physicians, and ICU fellows. Most training programs focused on quantitative or processed forms of EEG rather than short/intermittent EEG. By far the most common training program was for amplitude-integrated EEG (14 studies) followed by color density specral array (6 studies). EEG education programs involved a mix of large group didactic lectures, small group sessions, and selflearning or one-on-one review with EEG experts. In 16 studies, the overall length of the training program was one day or shorter. Trainee response was positive in the subset of studies reporting this variable.

The majority of EEG training programs involve ICU nurses, quantitative EEG, and are relatively short in duration. Such data could inform future EEG curriculum design for non-experts.

Recent studies have reported on the relationships between the gut microbiome and seizure disorders. Certain gut flora and their metabolites have been shown to exert antiepileptic and neuroprotective effects. While many gastrointestinal (GI) diseases are known to cause disruption of the normal gut microbiome in humans, the clinical impact of many GI diseases on subsequent risk for status epilepticus remains unknown. We conducted an exploratory analysis evaluating the relationship between categories of GI diseases and status epilepticus.

We performed a retrospective cohort study using claims between 2008-2015 from a nationally representative 5% sample of Medicare beneficiaries ≥66 years of age. We used previously validated diagnosis codes to ascertain our primary outcome of status epilepticus. In an exploratory manner, we categorized GI disorders by anatomic location, disease chronicity, and disease mechanism. We used Cox proportional hazards models to examine associations of GI disorder categories and status epileptics with adjustment for demographics.

In a mean of 1,719,890 beneficiaries in each analysis, several categories of GI disorders were associated with an increased risk of status epilepticus. 

Numerous categories of GI disorders were associated with an increased risk of future status epilepticus after statistical adjustment for demographics.

Rush University Medical Center, Neurology, Chicago, IL, United States

Coronaviruses have tropism for neural cells and there is evidence for neurologic manifestations of COVID-19. There is currently a paucity of data on EEG changes in patients infected with SARS-CoV-2.

We performed a retrospective chart review of 20 SARS-CoV-2 positive patients admitted at our institution who underwent video EEG monitoring between 3/31-5/30/2020. 

Although there is compelling evidence that COVID-19 affects neurologic function and all recorded EEGs were abnormal, we did not detect any seizures or unique EEG pattern in this sample. The abnormal background/generalized slowing noted in every EEG suggests a non-specific encephalopathy in this critically ill group, with focal dysfunction noted in 5/20 patients. A variety of abnormal movements were captured in 4/20 patients with no EEG correlate, which could represent an infectious or autoimmune movement disorder.

Appreciation for non-convulsive status epilepticus (NCSE) has led to an increase in admissions/transfers to neurocritical care units (NCCU) for continuous electroencephalography (cEEG). We sought to identify the yield of cEEG and factors that predict escalation of anti-epileptic therapy in patients admitted to the NCCU for suspected NCSE.

We performed a retrospective chart review of consecutive patients admitted to the NCCU for possible NCSE. Patients were admitted through the emergency department or transferred from other hospitals. We excluded patients with ongoing clinical SE at the time of admission/transfer or confirmed NCSE based on cEEG prior to admission. We collected demographics, medical and seizure history, cEEG results and medication management. We used logistic regression to explore predictors of need for escalation of anti-epileptic therapy based on cEEG.

We identified 218 admissions (54.4±17.9 years-old, 53% female, GCS 7.5±3.6) to the NCCU for management of suspected NCSE. Fifty-four patients (25%) were confirmed to be in status epilepticus (67% electrographic). Eighty-two percent of patients had changes in anti-epileptic therapy (78% antiepileptic drug, 37% anesthetic drips) based on cEEG. Absence of epilepsy history (Odds Ratio [OR] 0.23, 95% Confidence interval (CI) 0.08-0.66)), younger age (OR 0.97, 95% CI 0.93-0.99), lower GCS (OR 0.89, 95% CI 0.79-0.99), lower mean arterial pressure (OR 0.98, 95% CI 0.96-0.99), and fewer antiepileptic drugs prior to NCCU admission (OR 0.52, 95% CI 0.35-0.78) were associated with escalation of antiepileptic therapy based on cEEG.

Only one quarter of patients accepted to our NCCU for possible NCSE had confirmed status epilepticus, although eighty-two percent required escalation of anti-epileptic therapies. cEEG is a high yield monitor in patients with suspected NCSE.

Prior studies have shown anti-seizure drug (ASD) prophylaxis is associated with worse outcomes in patients with intracerebral hemorrhage (ICH). Here we sought to determine ASD treatment patterns, safety and association with outcomes in acute ICH.

We performed a retrospective analysis of patients undergoing EEG during admission for ICH. Presence of seizures/IIC patterns on EEG was recorded. Frequency of ASD use, indications for ASD, types of ASD, and ASD related adverse effects were recorded. We defined poor outcomes as MRS of 5-6 at 3-months. (n=24)); seizure/IIC patterns on EEG (28.42% (n=27)); history of epilepsy (6.32% (n=6)). Patients with temporal ICH were more likely to receive ASDs (OR=3.07, CI=1.13-8.38). The most commonly prescribed ASDs were: Levetiracetam (62.1% (n=59)), Lacosamide (9.47% (n=9)), Phenytoin (8.42% (n=8)), and Valproic acid (3.16% (n=3)). Among patients treated with ASDs, treatment related adverse effects were seen in 10.4%(n=7) patients. In 7.5% (n=5), ASDs were discontinued secondary to adverse effects. After excluding patients that received ASDs for a history of epilepsy, and adjusting for baseline ICH score, clinical seizure on admission, indication for ASD use, and seizures/IIC patterns on EEG, there was no significant association between ASD treatment and 3-month outcomes (MRS 0-4 vs. MRS 5-6); (OR=.13, CI =.011-1.41, p=0.093).

Patients with ICH frequently receive ASD treatment in response to clinical seizures and EEG findings. We did not find a significant association between ASD treatment and outcomes. Future randomized studies are indicated to determine whether ASD treatment in response to clinical seizures or EEG findings improve outcomes in patients with acute ICH.

Increasing burden of seizures and IIC patterns on EEG monitoring, is associated with worse discharge outcomes in acute ischemic stroke (AIS) patients. Here we sought to determine 1) anti-seizure drug (ASD) prescription patterns, and 2) ASD safety and effectiveness in patients with acute ischemic stroke (AIS).

We performed a retrospective analysis of patients with AIS who underwent EEG monitoring. Stroke location, etiology, NIHSS and baseline comorbidities were recorded. Burden of seizures/IIC patterns on EEG was recorded. We examined frequency of ASD prescription, indications for ASD use, and ASDrelated adverse effects. Primary outcome measure was 3-month mRS. 

Increasing seizure/IIC burden is associated with worse 3-month outcomes in AIS, and these patterns are frequently treated with ASDs. We found that ASD treatment was not significantly associated with outcomes. Our sample may be underpowered to detect a treatment effect, and prospective studies are indicated to determine whether ASD treatment in response to EEG findings improve outcomes in patients with AIS.

The current definition of delayed cerebral ischemia (DCI) is based on clinical characteristics precluding its use in poor-grade subarachnoid hemorrhage (SAH) patients. Additional concepts to evaluate the unconscious patient are required. Invasive neuromonitoring may allow timely detection of metabolic and oxygenation crises before irreversible damage has occurred.

We present a cohort analysis of all consecutive SAH patients referred to a single tertiary care center between 2010 and 2018. The cohort (n=190) was split into two groups, one before (n=96) and one after (n=94) the introduction of invasive neuromonitoring in 2014. A total of 55 poor-grade SAH patients were prospectively monitored using parenchymal oxygen saturation measurement (ptiO2) and cerebral microdialysis (CMD). The primary outcome was the Extended Glasgow Outcome Scale (GOS-E) after 12 months.

With neuromonitoring, the first DCI event was detected earlier (mean 2.2 days, p=0.002 

In this cohort analysis of poor-grade SAH patients, the introduction of invasive neuromonitoring and the extension of the classical DCI definition towards a functional dimension resulted in an earlier detection and treatment of DCI events. This led to an overall decrease in DCI related infarctions and an improvement in outcome.

Aneurysmal subarachnoid hemorrhage (aSAH) initiates a deleterious cascade activating multiple inflammatory systems, which can contribute to delayed cerebral ischemia (DCI). Procalcitonin is an established marker for sepsis treatment monitoring, and its time course in the context of DCI after aSAH patients remains unclear.

All patients admitted to our institution with aneurysmal aSAH from 2014 to 2018 were prospectively screened for eligibility. Daily procalcitonin levels were recorded alongside aSAH relevant characteristics. The predictive and confirmative value of procalcitonin levels was assessed using a receiver operating curve (ROC) and area under the curve (AUC) analysis. The course of PCT around the DCI event was evaluated in infection-free patients.

A total of 132 aSAH patients were included. Early procalcitonin levels (first 3 days post-aSAH) had low predictive value for the development of DCI (AUC = 0.661; p = 0.023) and unfavorable (GOS-E1-4) longterm outcome (AUC = 0.674; p = 0.003). In a subgroup analysis of infection-free patients (n=72), procalcitonin levels were higher in patients developing DCI (p = 0.001) and DCI-related cerebral infarction (p = 0.002). Procalcitonin concentrations increased gradually after DCI and decreased upon successful treatment. However, in refractory cases progressing to cerebral infarction, procalcitonine levels showed a secondary increase.

Early procalcitonin values were associated with consecutive DCI development, and the occurrence of unfavorable outcome. An additional rise was observed in patients progressing to cerebral infarction. Procalcitonin levels can contribute to DCI risk assessment in aSAH patients.

Early Brain Injury (EBI) after aneurysmal subarachnoid hemorrhage (aSAH) is hypothesized to play a significant role in mediating neurological injury. We hypothesized that the inflammatory mediator soluble ST2 (sST2) is associated with markers of EBI.

A total of 190 aSAH patients were prospectively enrolled. Clinical markers of EBI included early loss of consciousness (admission GCS ≤ 14), poor clinical status after treatment of hydrocephalus (24-hour GCS ≤ 8), and early neuroworsening (a decrease in GCS ≥ 2 within 72 hours of symptom onset). Radiographic markers included early hydrocephalus requiring external ventricular drainage (EVD) and global cerebral edema (Subarachnoid Hemorrhage Brain Edema Score (SEBES) ≥ 3). Plasma sST2 level was measured at early (3.5 ± 1.2 days post-hemorrhage), intermediate (7.8 ± 1.3 days post-hemorrhage), and late (13 ± 2.3 days post-hemorrhage) time points. The relationships between EBI markers and poor functional outcome (90-day mRS ≥ 3) were assessed using multivariable logistic regressions. The association between sST2 level over time and EBI markers was evaluated using repeated measure linear mixedeffects models.

Elevated sST2 

EBI after aSAH is a syndrome with multiple independent mechanisms which can each lead to poor functional outcome. Plasma sST2 is an independent predictor of clinical severity and hydrocephalus, but not global edema or early neuroworsening.

To mitigate initial brain injury, we introduced targeted temperature management (TTM) at 34°C to patients with WFNS grade V subarachnoid hemorrhage (SAH). We sought to study the feasibility and effect of endovascular cooling after rewarming during a high-risk period for delayed cerebral ischemia.

SAH patients before (July 2010 to May 2014) and during (June 2014 to February 2020) endovascular cooling period were investigated. Thirty-one patients in each group were treated with TTM at 34°C immediately after arrival by conventional surface cooling, together with emergency aneurysm repair. After rewarmed to 36°C around 7 days after the onset, patients in the study group underwent TTM to maintain 36-38°C for 7 days with Thermogard XP Temperature Management System (Asahi Kasei ZOLL Medical Co., Ltd.) using a two-balloon endovascular cooling catheter (CoolLine Catheter) through the internal jugular vein. The Control group was treated with antipyretics against fever.

There were no significant differences in gender (female, 58% vs. 64%), age (59.4 vs. 62.5 years), the Glasgow Coma Scale (4.1 vs. 4.0), modified Fisher CT classification, aneurysm location (anterior circulation, 81% vs. 87%), and treatment (clipping, 87% vs. 94%) between the endovascular cooling group and control. Significant difference was detected between the endovascular group and control in: fever above 38 °C after rewarming, 11 vs. 22 patients (p = 0.0046); mean duration above 38 °C, 4.6 vs. 21.5 hours (p = 0.0062); vasospasm-related cerebral infarction, 1 vs. 13 (p <0.0001). There was a trend toward favorable outcome (good recovery or moderate disability on the Glasgow Outcome Scale) at 3 months in the intravascular cooling group (55% vs. 32%, p=0.0716).

Fever was controlled better with an endovascular cooling device after TTM at 34°C. Elimination of fever was associated with a lower incidence of vasospasm-related infarction and may improve outcomes in poor-grade SAH

Delayed cerebral ischemia (DCI) and vasospasm are leading causes of secondary brain injury and morbidity for patients who survive the initial stages of aneurysmal subarachnoid hemorrhage (aSAH). Intraventricular nicardipine (IVN) is reported in the literature as a "rescue" therapy for aSAH patients with refractory vasospasm or DCI, but typical dosing may raise safety concerns. We report a case series of three patients who safely received repeated doses of concentrated IVN without complications.

This is a retrospective case series of three patients with modified fisher grade 4 aSAH with cerebral angiogram confirmed vasospasm, who received concentrated IVN of 2.5mg/ml rather than with the usual concentration of 1mg/ml.

A series of patients (n=3) with modified fisher grade 4 aSAH were treated with a range of 13-23 consecutive IVN doses for refractory vasospasm. The standard 1mg/ml IVN dose was replaced with a concentrated 2.5mg/ml dose, a reduction of 1.5 to 2 ml, in order to prevent large aspirations and fluctuations of cerebral spinal fluid and to maintain isovolumic technique. The concentrated IVN injections were well tolerated with no complications, including no subsequent ventriculitis or meningitis, or immediate safety concerns such as changes in intracranial pressure (ICP) or cerebral perfusion pressure (CPP).

We report that treatment with concentrated IVN for SAH-related DCI appears safe and tolerable. These three cases suggest that concentrated IVN should be considered in select circumstances when large (6-8mL) volumes of CSF cannot be safely aspirated without compromising isovolumic technique, ICP, or CPP. Further randomized controlled trials are warranted to assess safety and tolerability of concentrated IVN for aSAH related DCI.

Pupillometry allows a standardized measure of the pupillary light reflex (PLR). Acute hydrocephalus (HCP) is a common complication in approximately 66% of patients with aneurysmal subarachnoid hemorrhage (aSAH). HCP may affect the PLR due to increased intracranial pressure or dilation of the rostral aqueduct. The association between the PLR and HCP in aSAH patients has not been clearly established. The objective of this study is to analyze the correlation between the neurological pupil index (NPi) and the degree of HCP in aSAH patients.

The ENDPANIC registry is a prospectively collected database of pupillometry readings in patients admitted to four different neurological intensive care units. The degree of HCP was quantified using the hydrocephalus score. We analyzed data from aSAH patients in this registry who had pupillometry assessments within 6 hours of a brain CT. We investigated the correlation between the NPI and hydrocephalus followed by regression model to control for confounders.

A total of 46 patients were analyzed (mean age: 54 ± 7 years, 54.4% males, mean HCP score 5.5 ± 4.1) Fisher grade 4 aSAH was identified in 47.6% of the population. Mean NPI for right eye was 3.94 (±1.3) and 3.6 (±1.7) for the left eye. After adjusting for age, sex, race and sedation, there was no significant correlation between HCP and NPi (right eye r2= 0.26, p= 0.29, left eye r2= 0.21, p= 0.53)

In patients with aSAH, NPi was not correlated with HCP score. A small sample size could be a limitation of this study. Further studies are needed to characterize the clinical significance of pupillometry in the evaluation of patients with aSAH and HCP.

Non-traumatic subarachnoid hemorrhage (nSAH) is frequently complicated by DCI. Careful management of intravascular volume status (IVS) is a key factor in nSAH management, yet IVS can be challenging to reliably clinically assess without invasive monitoring. One emerging non-invasive option is to use radiotracer-tagged albumin nuclear imaging to assess blood volume (BVA). Our object was to quantify if/how BVA changed management during the high-risk periods for delayed cerebral ischemia (DCI).

Sequential nSAH patients were identified from October 2018 to December 2019 from a customdesigned electronic health record query based on problem lists and populated into a database. Demographics, hospital course, nSAH details were obtained from this database, in conjunction with retrospective chart review to obtain clinical IVS assessments. BVA was ordered at the discretion of the treating team without reference to a standard protocol. Data is presented as counts, percentages, means or medians with [IQR] as appropriate. Descriptive statistics were used to compare patients who had BVA to those that did not, with significance set at p=0.05. .47) scale between those who received BVA and those that did not. In 6/13 patients, management was changed by the BVA, including 1 case transfusion of packed red cells, 4 cases decrease in fluid administration/conversion to concentrated fluids, 1 case bolus of fluid.

BVA may inform fluid management in the high-risk DCI period in nSAH; protocolized, systematic study is needed.

The quick sequential organ failure assessment (qSOFA) score and systemic inflammatory response syndrome (SIRS) criteria are commonly used as screening tools for sepsis. Following subarachnoid hemorrhage (SAH), mediators of inflammation are released, which could lead to a falsely positive SIRS or qSOFA.. We hypothesized that the qSOFA score and SIRS are nonspecific in SAHs, and their use in the ED may lead to false positive sepsis screens and inappropriate antibiotic use.

This was a retrospective review of subjects from our institutional prospective SAH database. The initial SIRS and qSOFA scores were calculated retrospectively while patients werein the ER. The sensitivity, specificity and positive predictive value of these criteria for sepsis and septic shock was determined, using sepsis-III criteria.

Out of 104 patients analyzed, 37 had positive qSOFA scores and 61 were positive for two of four SIRS criteria. Five patients were treated for sepsis and two were treated for septic shock, all of whom screened positive for qSOFA and SIRS For sepsis, a positive qSOFA score had a sensitivity of 1, specificity of 0.68, and positive predictive value of 0.14 (PPV=0.05 for septic shock.) A positive SIRS score had a sensitivity of 1.0, specificity of 0.5, and positive predictive value of 0.08 (PPV=0.03 for septic shock.)

The early identification and treatment of sepsis is important, and is frequently used to evaluate quality of care in emergency departments. This study suggests that screening with qSOFA or SIRS is poorly predictive of sepsis and septic shock in patients with SAH. Overreliance on these screening tools, which have not been validated in patients with severe brain injury, may lead to inappropriate diagnosis and treatment of sepsis.

Headache after subarachnoid hemorrhage (SAH) is common, severe, and often treatment refractory. It negatively impacts patients' rehabilitation, quality of life, increases hospital readmissions, and can increase intracranial pressure by neural activation. There are no evidence-based guidelines for management of SAH-related headaches. Current standard of care includes prescription of multiple medications, often including opiates, until the pain is mitigated. Peripheral nerve blocks (PNBs) may be an effective therapeutic option for SAH-related headaches. We conducted a small pilot trial of PNBs to determine utility in headache pain management.

Data for 5 patients in a prospective interventional PNB group and 5 patients in a retrospective control group over a 12-month period was collected. All patients were followed for at least one week following their initial SAH event. All patients received a standard treatment of medications including acetaminophen, magnesium, gabapentin, dexamethasone and anti-spasmodics or anti-emetics as needed. Those in the PNB group additionally received bilateral greater occipital, lesser occipital, and supraorbital peripheral nerve blocks. The primary outcome was severity of pain, measured by Numeric pain rating scale (NPRS).

The mean ages in the PNB group and control group were 58.6 and 57.4, respectively. No patients in either group had vasospasm. Mean Hunt and Hess score in the PNB and control groups were 2.4 and 2.2, respectively. 3 patients in both groups had radiographic hydrocephalus and IVH, requiring external ventricular drain (EVD) placement. The PNB group had an average reduction in mean raw pain score of 2.8 (1.92, 4.68, p=0.024), and relative pain score of 0.3 (0.22, 0.48, p=0.026), compared to the control group. There was no difference in other aspects of management of these patients during the trial period.

Peripheral nerve blocks can be an effective treatment for headache management in the subarachnoid hemorrhage population. Further investigations with a larger sample size are warranted.

Hypovolemia following aneurysmal subarachnoid hemorrhage(aSAH) may worsen secondary brain injury. We aim to evaluate clinical characteristics and outcomes associated with admission volume status in aSAH patients using bedside ultrasonography.

We retrospectively reviewed 44 aSAH patients with lactate serially measured and bedside ultrasonography performed at admission. Volume status was assessed using Inferior Vena Cava(IVC) indices; Caval Index(CI) if breathing independently, Distensibility Index(DI) if on a ventilator. Patients were classified as hypovolemic if CI>50% or DI>18%. Univariate analyses were conducted by independent t-test or Wilcoxon rank sum test depending on data normality. Mixed effects model was used to test the global difference between groups across all time points. Three post-hoc contrast tests with Bonferroni adjustment were used for the comparison at 3 different time points.

Sixteen of 44(36%) aSAH patients had IVC indices indicating hypovolemia. Seventeen(39%) were intubated. Demographic characteristics were comparable between hypovolemics vs. non-hypovolemics. Glasgow Coma Scale, modified Fisher, and Hunt & Hess were also similar. While lactate level < 6H from admission did not differ between hypovolemics vs. non-hypovolemics (2.78±0.29 vs. 2.43±0.22, p=0.25), the level from 6-18H was significantly higher in the hypovolemic group after Bonferroni adjustment (2.79±0.29 vs. 1.89±0.21, p=0.04). Wilcoxon rank sum test showed significantly higher median procalcitonin level < 6H from admission in hypovolemics(0.14, IQR 0.1-0.62) vs. non-hypovolemics (0.1, IQR 0.06-0.16, p=0.02), but it was not significantly different in 18-30H. IVC indices did not predict vasospasm, delayed cerebral ischemia, or hospital mortality. The hypovolemic group had significantly fewer ventilator-free days(p<0.01), while EVD days, ICU days, and hospital days were similar.

Lactic acidemia was common in our sample. While both groups had lactic acidemia initially, the hypovolemic group had more persistent lactate elevations and higher procalcitonin levels early. IVC indices suggesting hypovolemia are associated with fewer ventilator-free days, but not with severity of clinical presentation or outcomes.

Predictors for cerebral vasospasm (CVS) are still under research in current literature. Previous studies have suggested an association between blood pressure fluctuations and the development of CVS and delayed cerebral ischemia. The aim of this study was to establish the possible correlation between blood pressure variability (BPV) and CVS.

We conducted a retrospective review of consecutive patients with aneurysmal subarachnoid hemorrhage (aSAH) admitted to the Baylor St. Luke's Medical Center Neurointensive Care Unit (NICU) between January 2013 and December 2019. Blood pressure variability (BPV), determined as the difference between the initial 3 days maximum and minimum of diastolic and systolic BP independently (Δ DBP and Δ SBP), serum sodium levels and evidence of vasospasm detected by transcranial Doppler (TCD), CT Angiogram (CTA), MR angiogram (MRA) and digital subtracted angiography (DSA) were collected. The association between BPV, hyponatremia and cerebral vasospasm (CVS) were investigated using univariate and multivariate logistic regression models.

114 patients with aSAH were included. The multivariate logistic regression model showed that patients with increased SBPV had higher odds of a developing CVS, OR 1.026, 95% CI 1.006-1.047. Furthermore, patient with hyponatremia also showed higher odds of developing CVS, OR, 5.02, 95 % CI 1.717-14.925. These increased odds were significant after controlling for gender, ethnicity, and admission neurological scales including; modified fisher scale (mFS), Glasgow Coma Scale (GCS) and World Federation of Neurological Surgeons Scale (WFNS).

We were able to demonstrate that patients with aSAH and early SBPV presented a higher likelihood of developing CVS. This is consistent with previous studies mentioning how SBPV impacts on cerebral autoregulation rendering vulnerable to CVS. Furthermore, multiple logistic regression model established hyponatremia as an independent predictor of CVS.

Optimal fluid management after aneurysmal subarachnoid hemorrhage is still a matter of debate in literature. Previous studies have suggested an association of an increased fluid intake with higher incidence of hospital complications, increased disease severity and poor functional outcomes. We aim to examine the relationship between fluid balance and survival after aSAH.

We conducted a retrospective review of consecutive patients with aneurysmal subarachnoid hemorrhage (aSAH) admitted to the Baylor St. Luke's Medical Center Neurointensive Care Unit (NICU) between January 2013 and December 2019. Total daily amount of fluid intake, output and fluid balance were calculated daily during the entire NICU stay. The association of fluid balance with the survival after aSAH was investigated using survival analysis with cox proportional hazards regression analysis.

114 patients with aSAH were included. A positive fluid balance during NICU stay was identified in 51.82 % of patients whereas 48.18 % had a neutral/negative fluid balance. Median survival in our sample was of 130.5 days (IQR 18-523). Patients with a positive fluid balance showed a statistically significant decreased survival (HR 1.56, 95% CI 1.01-2.39, p=0.04) compared to patients with a negative fluid balance after adjusting by age and admission WFNS.

We were able to demonstrate that a positive fluid balance significantly decreased survival after adjusting for age and admission WFNS. Our findings are consistent with prior literature concerning the detrimental effects of fluid overload after aSAH. Further prospective studies are needed for confirmation and interpretation of our results.

Blood pressure variability (BPV) has been suggested as an important measure of brain hemodynamics after aSAH. Previous studies have suggested an association between BPV and worse clinical outcomes. We aim to examine the relationship between BPV and short-term neurological outcomes in aSAH.

We conducted a retrospective review of consecutive patients with aneurysmal subarachnoid hemorrhage (aSAH) admitted to the Baylor St. Luke's Medical Center Neurointensive Care Unit (NICU) between January 2013 and December 2019. Blood pressure values were continuously recorded during the initial 3 days of NICU stay. BPV was determined as the difference between the initial 3 days maximum and minimum of diastolic and systolic BP independently (Δ DBP and Δ SBP). The association between BPV and clinical outcome were investigated using univariate and multivariate logistic regression models.

114 patients with aSAH were included. The multivariate logistic regression model showed that patients with increased SBPV had higher odds of a poor neurological outcome at hospital discharge (GOS ≤ 3), OR 1.110, 95 % CI 1.024-1.202. This increased risk was significant after controlling for gender, ethnicity, and admission neurological scales including; modified fisher scale (mFS), Glasgow Coma Scale (GCS) and World Federation of Neurological Surgeons Scale (WFNS). There was no association between DBPV and outcomes for our sample.

We were able to demonstrate that an increased SBPV during the first three days of admission to the NICU was independently associated with a poor neurological outcome at hospital discharge. Our findings are consistent with prior studies that suggest BPV impacts brain autoregulation after aSAH, therefore determining a worse prognosis in this set of patients. Further prospective studies are needed to confirm our results.

The management of headache pain following subarachnoid hemorrhage (SAH) can be challenging and often requires the sustained use of high dosage opiates (1). We investigated predictors of opiate utilization in SAH and their impact on 3 months outcomes.

A retrospective chart review was performed on sequential patients admitted with SAH and a Hunt Hess Grade (HHG) of 1-3 between January 1, 2017 and May 31, 2019. Opiate utilization was determined by calculating the mean daily morphine equivalence dosage (DMED) over the first 14 days of admission. Spearman correlation coefficient and multiple regression analyses were performed to investigate predictors of DMED and to test its association with clinical outcomes at 3 months.

83 aneurysmal and 49 peri-mesencephalic SAH were identified, of which 90% received opiates for headache. The DMED used was 18.7mg [0-37.5]. Age (r=-0.19, p=0.02), modified fisher scale (mFS) (r=0.26, p=0.002) and HHS (r= 0.28, p= 0.0009) were significantly associated with DMED. The DMED was found to be less among non-smokers compared to smokers (184mg vs 353mg, p=0.01), perimesencephalic SAH compared to aneurysmal SAH (195mg vs 283mg, p=0.004), and aneurysm clipping compared to coiling (152mg vs 306mg, p=0.002). In multivariate analysis, only HHG was statistically significant in predicting DMED (p =0.015). The DMED was significantly associated with hospital length of stay (r= 0.43, p<0.001) and the presence of headache at 3 months (194 vs 356 mg, p<0.05). While 48 (42%) reported headache, only 6 (5 %) remained on opiates at 3 months.

We found that opiate utilization following SAH was significantly associated with length of stay and headaches at 3 months. Future studies should focus on alternative treatments for headache pain in this patient population.

Chronic opioid use can usually be traced to an initial valid healthcare source. Severe, refractory pain and higher opioid doses increase the risk for chronic opioid use disorder. Little is known about the prevalence and risk factors for chronic opioid use following subarachnoid hemorrhage (SAH).

We performed a retrospective chart review of patients with SAH from November 2015 through September 2019. Severity and duration of pain were addressed as a single unit of pain burden defined as area under the curve when pain (Numeric Rating Scale) is graphed against time. Opioid doses were converted to standard morphine equivalents. Continued opioid use was defined by medication reconciliation at first follow up. Descriptive statistics were performed to compare outcome groups and logistic regression modeling to determine risk factors for continued opioid use.

We identified 238 consecutive SAH patients (mean age 54, 67% female, median HH grade 3) with a median follow-up of 64 days (Interquartile Range [IQR] 49-96) following admission. Forty-seven percent continued to use opioids at follow up. No significant difference was found in mean daily opioid dose (43mg vs 38mg, p=0.47) between groups. In multivariate analysis, a discharge opioid prescription was the strongest predictor for continued use at follow up (Odds Ratio [OR] 74.8, 95% Confidence Interval [CI] 28.0-200.0) with depression also demonstrating a robust predictive nature (OR 6.1, 95% CI 1.3-27.9). When discharge opioid prescriptions were removed from the model, non-white race (OR 1.9, 95% CI 1.1-3.5), private insurance (OR 2.0, 95% CI 1.1-3.6), depression (OR 3.2, 95% CI 1.2-8.7), and ICU Pain Burden (OR 1.003, 95% CI 1.002-1.004) were associated with continued opioid use.

Nearly half of all SAH patients continue to use opioids long after hospital discharge. Decreasing ICU pain burden through opioid-sparing therapeutics should be studied as a strategy to reduce chronic opioid use following SAH.

Subarachnoid hemorrhage (SAH) represents a devastating form of neurological emergency. Recent release of National Inpatient Sample (NIS) data for 2017 provides the opportunity to explore recent variations in hospitalizations, mortality and outcomes of SAH in the United States.

Analysis of US National Inpatient Sample of hospitalizations with SAH from January 1, 2017 to December 31, 2017. Baseline demographics and hospital-level characteristics associated with outcomes of SAH were analyzed. Hospital-level characteristics included hospital size (small, medium and large); hospital type (rural, urban community and urban teaching); and hospital geographic region (Northeast, Midwest, South, West). Outcomes included mortality and favorable discharge disposition defined as discharge to home.

Overall, a total of 41,460 hospitalizations with SAH were identified (57% women; median age, 63 [IQR 22.5, 

In 2017 in US, about 1 in 4.5 died following hospitalizations for SAH, relatively lower than prior reports from a mortality rate of 26.3% reported from 1996-2001. Larger hospital size and urban teaching hospital settings were associated with better discharge outcomes as compared to small or rural hospitals.

Perimesencephalic subarachnoid hemorrhages (panSAH) represent an atypical stroke mechanism with an unclear etiology. A presumed venous blood is often invoked, although the mechanism by which it is provoked remains unclear. Given that these hemorrhages often have low-lying blood at the cervicomedullary junction, the question arises whether structural pathology in the high cervical spine could precipitate this atypical hemorrhage. Here we aim to characterize the correlation of cervical spine disease with specific subtypes of angiogram-negative SAH, with a prediction that only panSAH will correlate with high cervical spine pathology.

Using an observational cohort study of patients treated at a single center from 2012 -2017, we identified 239 patients with angiogram-negative subarachnoid hemorrhages who underwent cervical spine MRI. Patients were stratified by bleed patterns: 73 patients had perimesencephalic-pattern (panSAH); 139 aneurysmal -pattern (aanSAH); and 27 were convexity (conSAH). The radiological features of MRI cervical spine included the severity of cervical spinal cord stenosis, osteophytes, degenerative disc disease, intrinsic cord signal and leptomeningeal signal.

There was no statistically significant relationship between the presence of spinal cord stenosis, degenerative disc disease, or osteophyte complexes between from C1-C3 (p>0.05). Similar results were found when comparing panSAH against a pooled analysis of aanSAH and conSAH. Extending the region of interest down to C7 did not alter the above comparisons (p > 0.05). Consistent with prior reports, cervical spine imaging also had low yield in identifying an occult vascular abnormality; 0 vascular lesions were found in 239 studies.

The yield of cervical spine imaging in angiogram negative SAH is limited. We find no evidence to support occult, structural pathology in the cervical spine as a mechanism of panSAH.

Preclinical evidence of the effects of albumin in models of subarachnoid hemorrhage has led to theorized mechanisms of clinical benefit that include attenuation of neuroinflammatory signaling and maintenance of adequate intravascular volume. We conducted this study to assess its effect on the rate of complications and mortality during ICU stay.

This was a quasi-experimental study with historic controls (before and after). We reviewed clinical records for all patients diagnosed with aneurysmal subarachnoid hemorrhage and treated in the ICU at our institution since 2011. We conformed two comparison groups based on the time of implementation of a continuous albumin infusion protocol for SAH in our institution. Since 2014, patients receive 5% albumin as fluid therapy during the first 5 days after bleeding. We recorded baseline characteristics, laboratory findings and occurrence of delayed cerebral ischemia (DCI, primary outcome), vasospasm, hyponatremia, pulmonary edema (PE), and death.

We included 189 patients for analysis (126 received albumin and 63 historic controls). Patients were similar in baseline characteristics, with exception of an increased proportion of patients with high grade SAH (WFNS >4) in the albumin group (34.9% vs 19%). DCI incidence was 39.7%, with lower frequency in the intervention group (35%) than in the control group (49%)(R=0.71, 95%CI=0.50 ;1.00). A finding further confirmed in multivariate analysis (MA) controlling for differences in severity between groups: OR= 0.43, 95%CI=0.22 ; 0.85. We found no differences in hospital mortality. We also report a reduced frequency of hyponatremia (27.0% vs. 49.2%, RR= 0.71, 95%CI= 0.55 -0.91), consistently lower administered fluid volumes and increased incidence of PE (12.7% vs 4.8%) in the albumin group.

Albumin is safe in SAH patients and appears to be associated with reduced rate of DCI, hyponatremia and less administered fluids during the first days of ICU stay. A phase III prospective study is warranted.

We had already reported that increased glucose variability was an independent predictor of unfavorable neurological outcome in patients with subarachnoid hemorrhage (SAH) (Okazaki 2018 189). However, the variability of cerebral glucose is unknown. We characterized the variability of cerebral glucose after SAH and its association with markers of physiologic distress.

We retrospectively analyzed cerebral microdialysis data (over 4 days) from a cohort of 13 SAH patients: 4 males with median age 63: IQR 56-70 and H-K grade 3-5, WFNS grade 2-5, Fisher group 3-4. Hourly values of cerebral glucose, lactate, pyruvate, and glycerol as well as continuous intracranial pressure and arterial blood pressure were recorded. Glucose and lactate variability are calculated by coefficient of variation (CV). The relationship between cerebral parameter variability and blood one was analyzed using the Wilcoxon rank sum test. 

Our data suggest that cerebral glucose variability may greater than blood one. In SAH, cerebral hypoglycemia has been shown to occur despite the absence of blood hypoglycemia during insulin therapy (Schlenk 2008 R9, Zetterling 2011 . In addition to cerebral hypoglycemia, the evaluation of cerebral glucose variability after SAH depends on further study.

Mechanical ventilation has not been well evaluated in subarachnoid hemorrhage (SAH) patients. In a study of traumatic brain injury, positive end-expiratory pressure (PEEP) had a significant effect on intracranial pressure (ICP), but within a safe ICP range.1 We expect the same relationship of PEEP and ICP in SAH patients, and seek to explore the impact of cerebrovascular reactivity (CVR). We hypothesized that higher PEEP would be associated with elevated ICP, and this effect would be more pronounced with abnormal CVR.

We performed a retrospective study of SAH patients admitted to our Neurocritical-Care Unit between June-2006 and January-2013 with multimodal neuromonitoring. CVR was evaluated by PRx, the correlation-coefficient between blood pressure and ICP. Patients were divided into two groups, those with intact (PRx≤0.20) and those with abnormal (PRx≥0.20) CVR.2 We analyzed an average of 11 days of data per patient. Linear regression was used to evaluate the relationship between PEEP and ICP.

55 SAH patients were included (37 female, age 54+/-15), with an average Hunt-Hess 4, Modified Fisher Score 2, and Glasgow Coma Scale 7. Patients with intact CVR (n=43) required a lower PEEP (median 5 IQR 5-7 vs median 5 IQR 5-10, p<0.001) and had lower mortality (22% vs 67% p<0.001) compared to those with abnormal CVR. We found no significant association between PEEP and ICP overall (r -0.05, p=0.230), nor in patients with intact CVR (r -0.034, p=0.460). In patients with abnormal CVR, higher PEEP correlated with lower ICP (r -0.417, p=0.026).

Contrary to our hypothesis, in patients with abnormal CVR, higher PEEP was actually associated with lower ICP. This pilot data suggests that PEEPs in the range of 5-10 are not associated with elevated ICP in SAH patients, regardless of CVR. The unexpected inverse relationship of PEEP and ICP in SAH patients is worth further exploration.

Opioid use is the mainstay for acute treatment of headaches after aneurysmal subarachnoid hemorrhage (aSAH); limited data exist capturing modern neurocritical care practices on headache and opioid use trends. We evaluated time trends related to pain severity and analgesic strategies, hypothesizing a decline with the increasing recognition of the impact of the opioid crisis.

We conducted a single center retrospective review of aSAH subjects with Hunt Hess ≤ 3, who were able to verbalize pain scores between 2012-2016 and 2017-2019. We collected hourly pain scores using visual analogue pain scale 0-10 and details on analgesic regimens. Descriptive statistics methods were used.

The cohort comprised 114 subjects (median 56 years, female 75.4%, Caucasian 71.9%) with a median ICU length-of-stay of 13 days (IQR 9-15). Severe headache (≥7) Opioids were less likely to be prescribed at discharge after 2017 (Z= -2.87; p=0.004).

Narcotics remain the mainstay of treatment for acute headache in aSAH, though pain control remains suboptimal despite high dosing. Increased use of alternative analgesics and decreased opioid prescription at discharge may reflect efforts to reduce longitudinal opioid use. Evidence-based therapeutic approaches to headache, including opioid-sparing strategies, are urgently needed.

Electroencephalography in the Acute Care Environment: A Systematic Review of Educational Initiatives for Non-Experts

McCredie Interdepartmental Division of Critical Care Medicine

Vidant Medical Center, Pharmacy, Greenville, NC, United States

To assess change in serum sodium, neurologic response, and survival to discharge in patients receiving 23.4% sodium chloride for acute treatment of intracranial hypertension.

This single center retrospective review includes patients with hemorrhagic or ischemic stroke from November 2018 to October 2019 who received one or more 30 mL doses of 23.4% sodium chloride for intracranial hypertension. The following data was collected: patient age, gender, type of stroke, dose of 23.4% sodium chloride, route of administration, length of ICU stay, serum sodium, pupillary light response, mortality, and signs of osmotic demyelination syndrome (ODS). The primary endpoint is change in serum sodium greater than 8 mEq/L. Secondary endpoints include herniation reversal, incidence of osmotic demyelination syndrome (ODS), and survival to discharge.

This study includes 81 patients (78%) with hemorrhagic stroke and 23 (22%) patients with ischemic stroke. Patients were divided into two dosing groups. The first group included 60 patients (58%) who received ≤2 vials of 23.4% sodium chloride, and the second group included 44 patients (42%) who received >2 vials. A significant change in serum sodium >8 mEq was observed between groups (>2 vials 31.8% vs. ≤2 vials 11.7%; p=0.04). Variables associated with increased sodium were larger doses of 23.4% sodium chloride, lower baseline serum sodium, and intraosseous route of administration (p=0.003, p<0.0001, and p=0.019). Survival to discharge was greater in the >2 vials group (70.5% vs. 31.7%; p<0.0001). Of the 39 patients with documented signs of brain herniation, reversal of herniation was greater in the ≤2 vials group (45% vs. 10.5%; p=0.016). No patients in either group experienced ODS.

Higher doses of 23.4% sodium chloride were associated with a significant increase in serum sodium. Patients receiving >2 vials of 23.4% sodium chloride had increased survival to discharge. 

UNIFESP-Federal University of São Paulo, São Paulo, Brazil

Despite the routine use of DTC in the diagnosis of vasospasm in patients with SAH, there are no data describing the behavior of cerebral blood flow velocities during and after treatment of patients with this complication. Objectives: Our primary objective was to evaluate cerebral blood flow velocities (BFV) kinetics in patients with SAH and DCI treated with induced hypertension (IH) and/or inotropics/vasodilators.

Methods: This was a prospective study in which all SAH patients admitted to a NeuroICU between June 2016 and June 2018 were evaluated for the development of DCI. Those with criteria for DCI who were treated with IH and/or inotropics/vasodilators were included. A TCD was performed immediately before (t0), 45 (t1) and 90 minutes (t2) after treatment onset. Patients were also evaluated clinically and neurologically at the same times points. 

Conclusions: TCD cerebral BFV reduce with DCI treatment and this is time dependent, being more substantial at 90 than at 45 minutes after treatment . Patients treated with inotropics/vasodilators had a significant decrease of MFV, which did not happen with those treated with IH.