Pediatric Moyamoya Revascularization Perioperative Care: A Modified Delphi Study

BACKGROUND

Surgical revascularization decreases the long-term risk of stroke in children with moyamoya arteriopathy but can be associated with an increased risk of stroke during the perioperative period. Evidence-based approaches to optimize perioperative management are limited and practice varies widely. Using a modified Delphi process, we sought to establish expert consensus on key components of the perioperative care of children with moyamoya undergoing indirect revascularization surgery and identify areas of equipoise to define future research priorities.

METHODS

Thirty neurologists, neurosurgeons, and intensivists practicing in North America with expertise in the management of pediatric moyamoya were invited to participate in a three-round, modified Delphi process consisting of a 138-item practice patterns survey, anonymous electronic evaluation of 88 consensus statements on a 5-point Likert scale, and a virtual group meeting during which statements were discussed, revised, and reassessed. Consensus was defined as ≥ 80% agreement or disagreement.

RESULTS

Thirty-nine statements regarding perioperative pediatric moyamoya care for indirect revascularization surgery reached consensus. Salient areas of consensus included the following: (1) children at a high risk for stroke and those with sickle cell disease should be preadmitted prior to indirect revascularization; (2) intravenous isotonic fluids should be administered in all patients for at least 4 h before and 24 h after surgery; (3) aspirin should not be discontinued in the immediate preoperative and postoperative periods; (4) arterial lines for blood pressure monitoring should be continued for at least 24 h after surgery and until active interventions to achieve blood pressure goals are not needed; (5) postoperative care should include hourly vital signs for at least 24 h, hourly neurologic assessments for at least 12 h, adequate pain control, maintaining normoxia and normothermia, and avoiding hypotension; and (6) intravenous fluid bolus administration should be considered the first-line intervention for new focal neurologic deficits following indirect revascularization surgery.

CONCLUSIONS

In the absence of data supporting specific care practices before and after indirect revascularization surgery in children with moyamoya, this Delphi process defined areas of consensus among neurosurgeons, neurologists, and intensivists with moyamoya expertise. Research priorities identified include determining the role of continuous electroencephalography in postoperative moyamoya care, optimal perioperative blood pressure and hemoglobin targets, and the role of supplemental oxygen for treatment of suspected postoperative ischemia.

Abstract 65: Cerebrovascular Reactivity Imaging Is Well Tolerated In Children With And Without Sickle Cell Disease

Background: Cerebrovascular reactivity (CVR) refers to a stimulus-induced change in blood flow, reflecting hemodynamic reserve. Carbon dioxide (CO 2 ) stimulus increases cerebral blood flow (CBF), and exogenous noninvasive CO 2 with concurrent MRI is an established CVR method in adults. However, children with and without sickle cell disease (SCD) have increased resting CBF compared to adults, and few studies have examined CVR in children. Our objective was to determine tolerability of CO 2 CVR MRI in children with and without SCD.

Methods: Children undergoing CO 2 CVR MRI had post-scan assessments of their experience and comfort during CVR testing. This included a preference rating ranking 8 experiences, including CVR MRI, in order from most to least enjoyable, a 6-point comfort scale, and a yes/maybe/no question asking if they would do CVR MRI again. The 8 preference items were grouped into quartiles for analysis. The Wilcoxon test was used to compare comfort rating based on SCD status and quartile preference rankings. Spearman’s correlation was used to compare peak end-tidal CO 2 with comfort rating.

Results: Nine children with SCD and 22 controls ages 8-21 years old completed CVR MRI, with a median peak end-tidal CO 2 of 50 mmHg [IQR 47, 51]. No child had a serious adverse event. On a 6-point scale, the median comfort rating was 4 (slightly uncomfortable) [IQR 2,4]. There was no difference in comfort based on SCD status (p=0.70) or peak end-tidal CO 2 (p=0.65). Across all participants, CVR fell in the 3rd preference quartile. CVR rated similarly to visiting the dentist (p=0.95), but was better than throwing up (p=0.03). Despite ranking the scan in the 3 rd quartile, the majority reported willingness to repeat CVR MRI, and only one participant (control) said they would not.

Conclusion: Noninvasive CO 2 CVR imaging is safe and tolerable in children with and without SCD. After completing the session, most children would be willing to undergo similar testing again.

Abstract TMP85: Hyperacute Pediatric NIH Stroke Scale Does Not Predict 3-month Outcome In Children

Background: The assumption of severity of initial stroke symptoms predicting long-term outcomes is foundational to deciding whether to give hyperacute therapies. The relationship is well-established in adult stroke, however, children often have additional symptoms at presentation, such as altered mental status, which may influence the pediatric NIH stroke scale (PedNIHSS). Prior preliminary work found that PedNIHSS obtained at a median of 28 hours from symptom onset was associated with long-term outcomes. We therefore sought to establish whether the initial hyperacute PedNIHSS, obtained within a possible hyperacute treatment window, correlates with long-term outcomes.

Hypothesis: We hypothesized that hyperacute PedNIHSS would correlate with pediatric stroke outcome measure (PSOM) at 3 months.

Methods: Multicenter data was prospectively collected and retrospectively reviewed. The hyperacute PedNIHSS was defined as PedNIHSS obtained within 24 hours of symptom onset and time last seen well (TLSW). Data were analyzed using Spearman’s correlation, quantile regression, and ordinal logistic regression.

Results: Among a total of 79 children, mean age was 12.7 years and 51% were female. PedNIHSS was obtained 0 to 23.6 hours, median 2.0 hours from symptom onset and/or TLSW. Median PedNIHSS score was 13 [IQR 9-17]. 93% received hyperacute therapies. Three children died within first 3 months. Median PSOM was 1 [IQR 0.5-2.5] at 3 months. Hyperacute NIHSS and PSOM at 3 months were both available in a total of 57 patients. There was no correlation between hyperacute NIHSS and PSOM at 3 months (r=0.27, n=57) ( Figure ).

Conclusion: Hyperacute PedNIHSS did not correlate with outcomes in our data. Further work is needed to understand factors influencing initial PedNIHSS and acute management that may confound the relationship between initial assessment and long-term outcome in children with acute ischemic stroke.

Pediatric Organ Dysfunction Information Update Mandate (PODIUM) Contemporary Organ Dysfunction Criteria: Executive Summary

Prior criteria for organ dysfunction in critically ill children were based mainly on expert opinion. We convened the Pediatric Organ Dysfunction Information Update Mandate (PODIUM) expert panel to summarize data characterizing single and multiple organ dysfunction and to derive contemporary criteria for pediatric organ dysfunction. The panel was composed of 88 members representing 47 institutions and 7 countries. We conducted systematic reviews of the literature to derive evidence-based criteria for single organ dysfunction for neurologic, cardiovascular, respiratory, gastrointestinal, acute liver, renal, hematologic, coagulation, endocrine, endothelial, and immune system dysfunction. We searched PubMed and Embase from January 1992 to January 2020. Study identification was accomplished using a combination of medical subject headings terms and keywords related to concepts of pediatric organ dysfunction. Electronic searches were performed by medical librarians. Studies were eligible for inclusion if the authors reported original data collected in critically ill children; evaluated performance characteristics of scoring tools or clinical assessments for organ dysfunction; and assessed a patient-centered, clinically meaningful outcome. Data were abstracted from each included study into an electronic data extraction form. Risk of bias was assessed using the Quality in Prognosis Studies tool. Consensus was achieved for a final set of 43 criteria for pediatric organ dysfunction through iterative voting and discussion. Although the PODIUM criteria for organ dysfunction were limited by available evidence and will require validation, they provide a contemporary foundation for researchers to identify and study single and multiple organ dysfunction in critically ill children.

Acute Neurologic Dysfunction in Critically Ill Children: The PODIUM Consensus Conference

CONTEXT

Acute neurologic dysfunction is common in critically ill children and contributes to outcomes and end of life decision-making.

OBJECTIVE

To develop consensus criteria for neurologic dysfunction in critically ill children by evaluating the evidence supporting such criteria and their association with outcomes.

DATA SOURCES

Electronic searches of PubMed and Embase were conducted from January 1992 to January 2020, by using a combination of medical subject heading terms and text words to define concepts of neurologic dysfunction, pediatric critical illness, and outcomes of interest.

STUDY SELECTION

Studies were included if the researchers evaluated critically ill children with neurologic injury, evaluated the performance characteristics of assessment and scoring tools to screen for neurologic dysfunction, and assessed outcomes related to mortality, functional status, organ-specific outcomes, or other patient-centered outcomes. Studies with an adult population or premature infants (≤36 weeks' gestational age), animal studies, reviews or commentaries, case series with sample size ≤10, and studies not published in English with an inability to determine eligibility criteria were excluded.

DATA EXTRACTION

Data were abstracted from each study meeting inclusion criteria into a standard data extraction form by task force members.

DATA SYNTHESIS

The systematic review supported the following criteria for neurologic dysfunction as any 1 of the following: (1) Glasgow Coma Scale score ≤8; (2) Glasgow Coma Scale motor score ≤4; (3) Cornell Assessment of Pediatric Delirium score ≥9; or (4) electroencephalography revealing attenuation, suppression, or electrographic seizures.

CONCLUSIONS

We present consensus criteria for neurologic dysfunction in critically ill children.

Neurocritical Care of the Pregnant Patient

Purpose of review

To summarize recent changes in management and emerging therapies for pregnant neurocritical care patients.

Recent findings

Diagnostic and treatment options for managing neurologic emergencies in pregnant patients have expanded with both greater understanding of the effects of imaging modalities and medications on pregnancy and application of standard treatments for non-pregnant patients to pregnant populations. Specifically, this includes cerebrovascular diseases (pregnancy-associated ischemic stroke, pregnancy-associated intracerebral hemorrhage, cerebral venous sinus thrombosis), post-maternal cardiac arrest care, seizures and status epilepticus, myasthenia gravis, and fetal somatic support in maternal death by neurologic criteria.

Summary

With the exception of direct abdominal computed tomography (CT), most imaging studies are reasonably safe in pregnancy. When emergent imaging is needed to prevent maternal morbidity or mortality, any CT sequence with or without contrast is appropriate to pursue. Though new safety data on antiplatelets, antihypertensives, thrombolytics, and antiepileptic drugs have increased options for disease management in pregnancy, unfractionated and low-molecular weight heparin remain the safest options for anticoagulation. Early studies on hypothermia, ketamine, and immunomodulating therapies in pregnancy are promising. In myasthenia gravis, new data on adjunct devices may allow more patients to undergo safe vaginal delivery, avoiding cesarean section and the associated risk of crisis. When difficult decisions regarding preterm delivery arise, recent outcome studies can help inform discussion. Lastly, when the feared complication of maternal death by neurologic criteria occurs, fetal somatic support may help to save at least one life.

IMMU-07. “STROKE MIMICS” ARE NOT BENIGN IN IMMUNOCOMPROMISED CHILDREN

Objective

To determine the clinical variances between strokes and stroke mimics in a pediatric immunocompromised population that consists of children with central nervous system (CNS) and non-CNS malignancies and a history of solid organ transplantation.

Methods

We performed a retrospective cohort analysis of stroke alert activations in patients with high-grade gliomas, low-grade gliomas, atypical teratoid rhabdoid tumors, rare CNS tumors, B-cell acute lymphoblastic leukemia, T-cell acute lymphoblastic leukemia, osteosarcoma, and solid organ transplants at St. Louis Children’s Hospital between February 2013 and September 2019. We categorized final diagnoses as strokes or stroke mimics. We classified diagnoses as a neurologic emergency if the diagnosis necessitated changes in management.

Results

Out of 217 stroke alerts, 31 alerts occurred for 28 patients meeting inclusion criteria. All final diagnoses constituted neurologic emergencies, including: stroke (39%), chemotherapy-related neurotoxicity (29%), tumor progression (19%), and seizures/posterior reversible encephalopathy syndrome (13%). Patients meeting inclusion criteria with strokes and stroke mimics presented similarly, with the exception of altered mental status, which was more prevalent in patients with strokes than stroke mimics (p = 0.03). One child received hyperacute thrombectomy for stroke. Only 58% of children with stroke mimics had complete resolution of their presenting neurologic symptoms. Children with strokes and stroke mimics had similar mortality incidences of 33% and 37%, respectively.

Conclusions

Although all acute neurologic changes in immunocompromised children are not strokes, stroke mimics in this population are neither benign nor self-limited and carry long-term neurologic morbidity and mortality. This study highlights the utility of an acute stroke evaluation infrastructure and the need for acute and long-term neurology involvement in the care of these patients.

Abstract P589: Pediatric Thrombectomy: Design And Workflow Lessons From Two Experienced Centers

Introduction: While clinical trials have demonstrated the remarkable efficacy of endovascular thrombectomy (EVT) for treating adult patients suffering from acute ischemic stroke (AIS), benefits reaped from advances in adult stroke care have unfortunately not occurred in parallel with pediatric stroke care. Randomized trials of EVT in childhood stroke are unlikely given the low incidence of stroke in children compared to adults, and despite promising outcomes in small case reports and series, EVT in children remains an off-label procedure lacking established consensus guidelines. Along with a clear need to collect prospective pediatric EVT outcome data, there is a need to enhance pediatric stroke care infrastructure to provide high-quality care to children experiencing stroke.

Methods: In this work, we review two successful pediatric thrombectomy programs, examining key workflow design features that are likely to be important for other programs that aspire to implement pediatric EVT capability.

Discussion: While pediatric EVT workflows will vary between centers, we identify several key elements of programmatic success shared between the two reviewed stroke programs that may serve as foundational design considerations for centers aiming to develop their own pediatric EVT programs. These elements include a formalized protocol and workflow, integration with an adult EVT workflow, simplification and automation of workflow steps, pediatric adaptations of stroke imaging, advocacy of pediatric stroke care, and collaboration between providers, among others. These essential features transcend any single hospital environment and may provide an important foundation for other pediatric centers that aim to enhance the care of children with stroke.

Conclusion: EVT shows promise in reducing stroke-associated morbidity in children. To maximize the efficacy of this intervention, workflow optimizations discussed here should be implemented by centers seeking to develop local pediatric EVT capability.

Bulk volume susceptibility difference between deoxyhemoglobin and oxyhemoglobin for HbA and HbS: A comparative study

PURPOSE

Sickle cell anemia is a blood disorder that alters the morphology and the oxygen affinity of the red blood cells. Cerebral oxygen extraction fraction measurements using quantitative BOLD contrast have been used for assessing inadequate oxygen delivery and the subsequent risk of ischemic stroke in sickle cell anemia. The BOLD signal in MRI studies relies on Δ χ do , the bulk volume susceptibility difference between fully oxygenated and fully deoxygenated blood. Several studies have measured Δ χ do for normal hemoglobin A (HbA). However, it is not known whether the value is different for sickle hemoglobin. In this study, Δ χ do was measured for both HbA and sickle hemoglobin.

METHODS

Six sickle cell anemia patients and 6 controls were recruited. Various blood oxygenation levels were achieved through in vivo manipulations to keep the blood close to its natural state. To account for the differences in oxygen affinity, Hill's equations were used to translate partial pressure of oxygen to oxygen saturation for HbA, sickle hemoglobin, and fetal hemoglobin (HbF) separately. The pH and PCO₂ corrections were performed. Temperature and magnetic field drift were controlled for. A multivariate generalized linear mixed model with random participant effect was used.

RESULTS

Assuming that Δ χ do is similar for HbA and HbF and that Δ χ metHb is 5/4 of Δ χ do for HbA, it was found that the Δ χ do values for HbA and sickle hemoglobin were not statistically significantly different from each other.

CONCLUSION

The same Δ χ do value can be used for both types of hemoglobin in quantitative BOLD analysis.

Abstract TMP112: Increased Large Vessel Diameter is Associated With Increased Cerebral Metabolic Stress in Children With Sickle Cell Anemia

Background: Children with sickle cell anemia (SCA) invoke compensatory mechanisms to meet cerebral oxygen demand in the setting of low oxygen carrying capacity. Whereas cerebral blood flow (CBF) is largely regulated by cerebral arterioles as one compensatory mechanism, we sought to determine if the cross-sectional area of the large in-flow arteries was also increased.

Methods: Children and young adults with SCA underwent 3 Tesla magnetic resonance angiography (MRA) and imaging (MRI) with pseudocontinuous arterial spin labeling (pCASL) and asymmetric spin echo sequence to measure CBF and oxygen extraction fraction (OEF) respectively. Using the double oblique method, we measured artery diameters and calculated the area (A) of the vessel as A=πr 2 . The total inflow area (A T-inflow ) was the summation of the areas of distal portions of both internal carotid arteries and basilar artery. CBF, OEF, arterial oxygen content [(CaO 2 ) = 1.36*Hemoglobin*%saturation on pulse oximetry] and A T-inflow were compared between groups with nonparametric statistics. Spearman’s correlation determined relationship of A T-inflow with other variables.

Results: Data was available for 29 subjects with SCA (median age 15 years (range 9-26), 45% male) and 17 age-matched controls (15y (range 9-18), 47% male) underwent scanning. Children with SCA had larger A T-inflow (36 [29, 40] mm 2 ) compared to age-matched controls (29 [26, 31] mm 2 , p=0.01). CBF and OEF were higher (p=0.001, p=0.002), and CaO 2 was lower (p<0.001) in SCA compared to controls. A T-inflow positively correlated with CBF (p=0.002) and OEF (p<0.001), and negatively correlated with CaO 2 (p=0.02; Figure ).

Conclusion: Increased cross sectional area of large cerebral arteries may provide an additional compensatory mechanism in SCA to augment CBF and aid in meeting cerebral oxygen demand. Further work is needed to determine if MRA measurements provide a non-invasive precursor to increased cerebral metabolic stress and stroke risk.

Abstract TP210: Oxygen Extraction Fraction is Increased in the Borderzone Region of Healthy Young Adults

Background: We have previously shown that the borderzone region demonstrates ischemic physiology, with low cerebral blood flow (CBF) and increased oxygen extraction fraction (OEF), in pediatric sickle cell anemia (SCA). This region is vulnerable in SCA as evidenced by increased infarct burden. Here we studied healthy adults to determine if borderzone physiology exists in the absence of apparent vascular disease.

Methods: Healthy (N=17, 33 ± 9 yr) and SCA (N=13, 28 ± 7 yr) adults were recruited from a tertiary care SCA clinic. Individuals with vascular risk factors or chronic medical or neurological diseases were excluded. SCA subjects were excluded for stroke, vasculopathy, or transfusion therapy. Brain MRIs were prospectively obtained including: T1, FLAIR, dynamic susceptibility contrast (CBF), and asymmetric spin echo (OEF). Within each subject, 3 regions of low CBF were derived by normalizing CBF to its peak CBF. For gray matter (GM), regions were: < 25 th %, 25-50 th %, and 50-75 th % of mean thalamic CBF. For white matter (WM), regions were: < 25 th %, 25-35 th %, and 35-45 th % of mean GM CBF. Mean OEF across the 3 CBF regions were compared using Friedman test to account for repeated measures (Fig*).

Results: Figure demonstrates the approximated borderzone with heatmaps (# of subjects with WM CBF < 25 th %) for healthy and SCA adults, alongside average OEF maps for each cohort. Both in healthy and SCA adults, GM and WM OEF were elevated in the region of lowest CBF. Furthermore, relative tissue volumes of CBF < 25 th % were higher in SCA than controls (GM: 16 vs 11%, p<0.001; WM: 33 vs 23% p<0.001), suggesting disease may redistribute CBF, effectively enlarging the borderzone.

Conclusion: We found ischemic physiology (low CBF / high OEF) in the borderzone of healthy adults. Additional stressors limiting cerebral oxygen delivery such as chronic anemia in SCA or carotid occlusion may lead to enlargement of the borderzone, further elevation of OEF, and a heightened vulnerability to infarction.

Pathophysiology and Management of Moderate and Severe Traumatic Brain Injury in Children

Traumatic brain injury remains a leading cause of morbidity and mortality in children. Key pathophysiologic processes of traumatic brain injury are initiated by mechanical forces at the time of trauma, followed by complex excitotoxic cascades associated with compromised cerebral autoregulation and progressive edema. Acute care focuses on avoiding secondary insults, including hypoxia, hypotension, and hyperthermia. Children with moderate or severe traumatic brain injury often require intensive monitoring and treatment of multiple parameters, including intracranial pressure, blood pressure, metabolism, and seizures, to minimize secondary brain injury. Child neurologists can play an important role in acute and long-term care. Acutely, as members of a multidisciplinary team in the intensive care unit, child neurologists monitor for early signs of neurological change, guide neuroprotective therapies, and transition patients to long-term recovery. In the longer term, neurologists are uniquely positioned to treat complications of moderate and severe traumatic brain injury, including epilepsy and cognitive and behavioral issues.

Abstract T P45: Automated CSF Segmentation to Quantify Cerebral Edema Following Large Hemispheric Ischemic Stroke

Introduction: Cerebral edema with resultant mass effect is a potentially fatal consequence of ischemic stroke, but early and sensitive biomarkers of brain tissue compression are lacking. To quantify brain mass effect, we developed a novel, automated segmentation method to delineate CSF spaces in CT images from ischemic stroke patients.

Methods: CTs from sixteen acute ischemic stroke patients (median NIHSS 16.5, median age 61.5 yrs, 14-92 hrs after stroke onset) were included after informed consent was obtained. After infarction, conventional CSF segmentation using Hounsfield unit (HU) thresholding is suboptimal due to infarct hypodensity. Utilizing manually delineated infarct and CSF spaces as training samples, we augmented conventional HU threshold segmentation with level sets, sparse regression and random forest segmentation methods. Using leave-one-out cross-validation, the combined approach was compared to HU thresholding using Dice ratios (a measure of the overlap between the segmented and the ground-truth CSF spaces).

Results: Shown is an example of a CT brain slice segmented by HU thresholding and the combined strategy: false negative (red), false positive (green), and true positive (yellow). The Dice ratios for HU thresholding and the combined approaches were 58.2±16.3% and 68.9±14.6%, respectively, demonstrating the significantly improved performance for the combined strategy (p=0.0014).

Conclusions: We have developed an advanced image segmentation strategy to delineate CSF spaces which outperforms conventional HU thresholding. An automated CSF segmentation strategy will permit quantification of cerebral edema in a large population of stroke patients, as required for genetic studies, for example.

Abstract W P32: CSF Volumetric Analysis Reliably Quantifies Cerebral Edema And Correlates With Clinical Deterioration In Large Hemispheric Infarcts

Background: Large hemispheric infarcts (LHI) may be complicated by cerebral edema. Midline shift (MLS), a standard radiographic measure, only crudely estimates extent of edema. Volumetric analysis of CSF compartments over time may provide a reliable and accurate means of quantifying severity and kinetics of edema after LHI.

Methods: We retrospectively identified stroke patients with NIHSS≥8 and baseline CT within 6 hours who developed cerebral edema (without hemorrhage) on follow-up (FU) CTs. Two raters outlined the sulci and lateral ventricles ipsilateral (IL) and contralateral (CL) to the infarct on baseline and serial FU CTs (both within 48 hours and at peak edema, 2-5 days post-stroke) and quantified CSF and infarct volumes. Changes in compartment volumes from baseline to peak edema CT were correlated with MLS and edema-related neurologic worsening (need for hemicraniectomy, osmotic therapy, or decline in GCS, associated with MLS≥5mm).

Results: Ten patients were analyzed (median NIHSS 14, time to early FU CT 30 hours, IQR 15-37 and to peak edema CT 75 hours, IQR 64-95). Inter-rater reliability for volume measures was excellent (intraclass correlation >0.97). CSF volume diminished by 37±20% (49 ml) from baseline to peak edema, over half occurring within 48 hours; net decline in CSF volume correlated with infarct volume (r=-0.63,p=0.05). Greatest reductions in CSF were seen in IL sulci and IL ventricles (Figure, top), but it was % reduction in CL sulci that correlated best with MLS (Fig, bottom), even adjusting for infarct volume (p=0.02). Decline in volumes of IL and CL sulci were greater in the 5 subjects with neurological worsening (89% vs. 56% and 40% vs. 3%, p<0.05), while infarct volume was not.

Conclusions: CSF volumetrics is a reliable tool for quantifying cerebral edema after LHI and a novel method of studying edema kinetics. Loss of sulcal volume correlates with MLS and is more strongly associated with edema-related deterioration than infarct volume alone.

Abstract T P72: Baseline Variables Have Little Influence on Early Change in Neurological Status (ΔNIHSS) After Acute Ischemic Stroke: Basis For a Genetic Study

Introduction: Neurological deficits can be highly unstable within the first 24 hours after acute ischemic stroke (AIS), with some patients showing dramatic improvement while others rapidly deteriorate. We are interested in genetic influences on early neurological recovery/deterioration. Here, we characterize NIHSS changes within the first 24 hours after stoke onset (ΔNIHSS) in a large cohort to determine baseline clinical variables that influence this outcome measure.

Methods: AIS patients presenting to two sites (Barnes-Jewish Hospital, St Louis and Vall D’Hebron Hospital Barcelona) between 2008-2013 were prospectively enrolled. Baseline NIHSS was collected within 6 hours and again at 24 hours after symptom onset. ΔNIHSS was calculated as the difference in these stroke scale scores. Demographics, baseline comorbidities and medications, as well as acute treatment variables were recorded for each subject. Stepwise multivariable regression (SAS) was used to determine variables that significantly influence ΔNIHSS.

Results: There were 954 patients enrolled (St Louis = 433, Barcelona = 521). Table 1 demonstrates the frequencies and means (SD) of the baseline variables. ΔNIHSS follows a normal distribution (figure). All baseline variables listed in table 1 were analyzed for influence on ΔNIHSS. Only baseline NIHSS (R2 = 0.0597, p<0.0001), baseline glucose (R2 = 0.0176, p=<0.0001,) and age (R2 = 0.0106, p=0.0011) independently influenced ΔNIHSS, accounting for only 8.79% of the variance.

Conclusion: Baseline variables (NIHSS, glucose and age) modestly influence early neurological recovery/deterioration. However, 91% of ΔNIHSS variability remains unexplained, suggesting that other factors such as genetics, could play an important role in early outcomes following AIS. A GWAS of ΔNIHSS is currently underway.

Abstract W P317: Early CSF Volume Changes Predict Malignant Edema in Large Hemispheric Infarction

Background: Early clinical and radiographic measures (such as NIHSS or early CT hypodensity) incompletely predict which patients with hemispheric stroke will develop malignant cerebral edema. We evaluated whether quantitative changes in volume of CSF compartments on early follow-up (FU) CT predict peak radiographic edema and clinical worsening.

Methods: We retrospectively identified patients with hemispheric infarcts, NIHSS≥8, baseline CT within 6 hours of stroke onset and FU CT within 48 hours (i.e. prior to development of maximal edema). Volumes of CSF in ipsilateral (IL) and contralateral (CL) sulci and lateral ventricles were manually outlined on both scans, as was infarct volume on FU CT. Midline shift (MLS) was measured on later CT at time of peak edema (if available). Reduction in CSF compartments from baseline and symmetry (IL:CL ratios) were correlated to peak MLS and edema-related clinical deterioration (need for hemicraniectomy, osmotic therapy, or GCS decline, with MLS>5mm).

Results: Ten patients were analyzed (median NIHSS 13, FU CT at median of 30 hours, IQR 15-37). Sulcal asymmetry (ratio of IL:CL volume) on FU CT was greater in the 4 subjects who deteriorated from malignant edema (median 0.26 vs. 0.79, see Figure), as was % reduction in IL sulcal volume from baseline (76% vs. 35%, p=0.06), while volume of early infarct was not. Hemispheric CSF asymmetry and % reduction in IL ventricular volume at early FU CT were strongly correlated with peak MLS (r=-0.95 and -0.96, both p=0.01). Linear regression found that both early infarct volume and % reduction in IL ventricle volume were strongly associated with MLS, adjusting for baseline CSF volume (Betas 0.55 and -0.48, both p=0.001).

Conclusions: In this preliminary study, early CSF asymmetry and reduction in IL CSF volumes appear to predict development of MLS and malignant edema. Further validation is needed to test whether CSF volumetrics have utility in selection of patients for early aggressive interventions.

Abstract T P370: Blood Transfusions in Pediatric Sickle Cell Disease Normalize Cerebral Blood Flow While Maintaining Constant Cerebral Oxygen Metabolism

Background: Chronic blood transfusions (Tx) reduces stroke risk in pediatric sickle cell disease (SCD). Cerebral blood flow (CBF) is elevated in SCD, likely representing a compensatory mechanism to maintain cerebral oxygen metabolism (CMRO2) in the setting of reduced arterial oxygen content (CaO2) from chronic anemia. When exhausted compensatory mechanisms are unable to meet oxygen demands, stroke ensues. We measured MR-derived CBF and oxygen extraction fraction (OEF) pre- and post-Tx, hypothesizing that Tx ‘resets’ the CBF baseline by increasing CaO2 via increased hemoglobin (Hb), while maintaining cerebral oxygen delivery and metabolism.

Methods: SCD children on chronic Tx were enrolled in a prospective, observational MRI study. MR-CBF and MR-OEF were acquired before and 2 hours after exchange Tx. MR-CBF and MR-OEF were measured using pseudocontinuous arterial spin labelling and a novel asymmetric spin echo sequence, respectively. CaO2 =1.35 x [Hb] x SaO2. CMRO2 = CaO2 x CBF x OEF.

Results: Two SCD children underwent MRI pre- and post-Tx (six more are anticipated prior to ISC). For subject #1 (18 yo F with overt stroke), mean global CBF was 128 and 98 ml/min/100g pre- and post-Tx, respectively, indicating a 24% CBF reduction. For subject #2 (6 yo F with elevated transcranial Doppler velocities), mean global CBF was 189 and 129 ml/min/100g pre- and post-Tx, respectively, a 32% CBF reduction (Fig). Both Hb and CaO2 were increased after Tx, resulting in unchanged oxygen delivery (CaO2 x CBF) post-Tx. Moreover, OEF and CMRO2 were not significantly different pre- and post-Tx, consistent with our hypothesis that CBF increases to maintain oxygen delivery.

Conclusions: Elevated CBF is likely a compensatory mechanism to maintain constant oxygen delivery in SCD children who have chronically low CaO2. In our subjects, Tx improved CaO2, allowing CBF to normalize. This reduced hemodynamic stress likely contributes to the lower stroke risk in chronically transfused SCD children.

Abstract 76: Genome Wide Association Study of Early Neurological Deterioration after Acute Ischemic Stroke Defines the Interferon-Stimulated Gene IFIT1 as a Neuroprotective Factor

Introduction: In the first hours after ischemic stroke, neurological deficits can be highly unstable. These early neurological changes are important due to their influence on long-term outcome. We performed a GWAS on ΔNIHSS 24h (NIHSS change from baseline to 24 hours after stroke onset) in acute ischemic stroke patients.

Methods: DNA from 191 stroke patients presenting within 4.5 hours of stroke onset was genotyped using an Affymetrix Exome-chip. Single variant analysis was performed using PLINK, including age, baseline NIHSS and principal component factors as covariates. European- and African-Americans were analyzed separately, with p-values combined by meta-analysis. Gene-based analysis was performed using SKAT-O, including only non-synonymous variants. Pathway analysis was performed using ALIGATOR to identify pathways with SNPs with significant associations. IFIT1 -/- and IFIT1 +/+ mice underwent one-hour MCA occlusion (tMCAO) followed by 24-hour reperfusion. Brains were removed, TTC-stained, and infarcts measured.

Results: Single variant analysis did not reveal genome-wide significant associations. One gene, IFIT1 (interferon-induced protein with tetratricopeptide repeats), passed the gene-based genome-wide multiple test correction (A). All three polymorphic variants in IFIT1 associated with neurologic deterioration with an average ΔNIHSS 24h 9.5 points lower in carriers versus non-carriers (B). Pathway analysis, including 21 interferon-related genes but excluding IFIT1 , showed a highly significant association (p=2.30х10 -3 ) with ΔNIHSS 24h. Infarct volumes in IFIT1 -/- mice were twice the size of IFIT1 +/+ mice after tMCAO (C, 110.6 mm 3 vs 52.8 mm 3 ).

Conclusion: These data suggest that IFIT1 and other interferon-related genes may function in endogenous neuroprotective responses during acute ischemic stroke.

Abstract W MP111: Prevalence and Risk Factors of Silent Cerebral Infarcts in Children with Hemoglobin SC Disease

Background: Hemoglobin (Hb) SC disease comprises 22% of all sickle hemoglobinopathies and is the second most common form of sickle cell disease. Silent cerebral infarcts (SCI) are common in Hb SS disease and are associated with male sex, higher systolic blood pressure (SBP), and lower baseline Hb concentration among children with Hb SS. However, SCI prevalence and risk factors are less well characterized in Hb SC disease. We tested the hypothesis that risk factors for SCI in Hb SC and Hb SS are similar.

Methods: Retrospective chart review of children with Hb SC seen at St. Louis Children’s Hospital (SLCH) Sickle Cell Clinic 2004-2012 who had brain magnetic resonance imaging (MRI) available. At SLCH, all children with sickle cell disease undergo screening brain MRI after their 6 th birthday; additional MRIs are obtained for concerning symptoms. Cerebral infarctions were identified as T2- or FLAIR-weighted hyperintensities identified in at least 2 planes or decreased diffusion; silent infarcts were diagnosed when the patient had no neurological symptoms that correlated with the infarct lesions. Hb concentrations and SBP were obtained from clinic well visits within a year before and after MRI. Human Studies Committee approval and waiver of consent were granted for this study.

Results: Thirteen of 96 patients with Hb SC disease had SCI on MRI. The prevalence of SCI was 13.5%. Twelve (92%) subjects had subcortical lesions; 11 (85%) had SCI in the frontal lobe; and 10 (77%) subjects had bilateral infarctions. Seven (50%) of subjects with SCI were male. The mean age at identification of SCI was 11.9 years (range, 6.2-19.3 years). Ten children with SCIs had repeat MRI (range 0.1-6.4 years) following the SCI diagnosis. No child had progression or additional SCI on repeat imaging. Sex, SBP, and Hb concentration were not significantly different between children with and without SCIs (p=0.4).

Conclusion: Our cohort of children with Hb SC had a SCI prevalence of 13.5%. The majority of children had bilateral subcortical lesions. SCIs had frontal lobe predominance. Unlike children with Hb SS, gender, elevated SBP or lower baseline Hb were not risk factors for SCIs in this cohort.

Hypothermia for pediatric refractory status epilepticus

PURPOSE

Refractory status epilepticus (RSE) is a life-threatening emergency, demonstrating, by definition, significant pharmacoresistance. We describe five cases of pediatric RSE treated with mild hypothermia.

METHODS

Retrospective chart review was performed of records of children who received hypothermia for RSE at two tertiary-care pediatric hospitals between 2009 and 2012.

KEY FINDINGS

Five children with RSE received mild hypothermia (32-35°C). Hypothermia reduced seizure burden during and after treatment in all cases. Prior to initiation of hypothermia, four children (80%) received pentobarbital infusions to treat RSE, but relapsed after pentobarbital discontinuation. No child relapsed after treatment with hypothermia. One child died after redirection of care. Remaining four children were discharged.

SIGNIFICANCE

This is the largest pediatric case series reporting treatment of RSE with mild hypothermia. Hypothermia decreased seizure burden during and after pediatric RSE and may prevent RSE relapse.