Quantitative Electroencephalographic Changes Associated With Brain Tissue Hypoxia After Pediatric Traumatic Brain Injury: A Retrospective Exploratory Analysis.

PURPOSE: Brain tissue hypoxia is associated with poor outcomes after pediatric traumatic brain injury. Although invasive brain oxygenation (PbtO 2 ) monitoring is available, noninvasive methods assessing correlates to brain tissue hypoxia are needed. We investigated EEG characteristics associated with brain tissue hypoxia. METHODS: We performed a retrospective analysis of 19 pediatric traumatic brain injury patients undergoing multimodality neuromonitoring that included PbtO 2 and quantitative electroencephalography(QEEG). Quantitative electroencephalography characteristics were analyzed over electrodes adjacent to PbtO 2 monitoring and over the entire scalp, and included power in alpha and beta frequencies and the alpha-delta power ratio. To investigate relationships of PbtO 2 to quantitative electroencephalography features using time series data, we fit linear mixed effects models with a random intercept for each subject and one fixed effect, and an auto-regressive order of 1 to model between-subject variation and correlation for within-subject observations. Least squares (LS) means were used to investigate for fixed effects of quantitative electroencephalography features to changes in PbtO 2 across thresholds of 10, 15, 20, and 25 mm Hg. RESULTS: Within the region of PbtO 2 monitoring, changes in PbtO 2 < 10 mm Hg were associated with reductions of alpha-delta power ratio (LS mean difference -0.01, 95% confidence interval (CI) [-0.02, -0.00], p = 0.0362). Changes in PbtO 2 < 25 mm Hg were associated with increases in alpha power (LS mean difference 0.04, 95% CI [0.01, 0.07], p = 0.0222). CONCLUSIONS: Alpha-delta power ratio changes are observed across a PbtO 2 threshold of 10 mm Hg within regions of PbtO 2 monitoring, which may reflect an EEG signature of brain tissue hypoxia after pediatric traumatic brain injury.

Early Clinical Variables Associated With Refractory Convulsive Status Epilepticus in Children.

BACKGROUND AND OBJECTIVES: The objective of this study was to determine patient-specific factors known proximate to the presentation to emergency care associated with the development of refractory convulsive status epilepticus (RSE) in children. METHODS: An observational case-control study was conducted comparing pediatric patients (1 month-21 years) with convulsive SE whose seizures stopped after benzodiazepine (BZD) and a single second-line antiseizure medication (ASM) (responsive established status epilepticus [rESE]) with patients requiring more than a BZD and a single second-line ASM to stop their seizures (RSE). These subpopulations were obtained from the pediatric Status Epilepticus Research Group study cohort. We explored clinical variables that could be acquired early after presentation to emergency medical services with univariate analysis of the raw data. Variables with p < 0.1 were retained for univariable and multivariable regression analyses. Multivariable logistic regression models were fit to age-matched and sex-matched data to obtain variables associated with RSE. RESULTS: We compared data from a total of 595 episodes of pediatric SE. Univariate analysis demonstrated no differences in time to the first BZD (RSE 16 minutes [IQR 5-45]; rESE 18 minutes [IQR 6-44], p = 0.068). Time to second-line ASM was shorter in patients with RSE (RSE 65 minutes; rESE 70 minutes; p = 0.021). Both univariable and multivariable regression analyses revealed a family history of seizures (OR 0.37; 95% CI 0.20-0.70, p = 0.0022) or a prescription for rectal diazepam (OR 0.21; 95% CI 0.078-0.53, p = 0.0012) was associated with decreased odds of RSE. DISCUSSION: Time to initial BZD or second-line ASM was not associated with progression to RSE in our cohort of patients with rESE. A family history of seizures and a prescription for rectal diazepam were associated with a decreased likelihood of progression to RSE. Early attainment of these variables may help care for pediatric rESE in a more patient-tailored manner. CLASSIFICATION OF EVIDENCE: This study provides Class II evidence that patient and clinical factors may predict RSE in children with convulsive seizures.

A Survey of Pediatric Neurocritical Care Fellowship Training in North America.

BACKGROUND: Pediatric neurocritical care (PNCC) has emerged as a field to care for children at the intersection of critical illness and neurological dysfunction. PNCC fellowship programs evolved over the past decade to train physicians to fill this clinical need. We aimed to characterize PNCC fellowship training infrastructure and curriculum in the United States and Canada. METHODS: Web-based survey of PNCC fellowship program leaders during November 2019 to January 2020. RESULTS: There were 14 self-identified PNCC fellowship programs. The programs were supported by Child Neurology and/or Pediatric Critical Care Medicine divisions at tertiary/quaternary care institutions. Most programs accepted trainees who were board-eligible or board-certified in child neurology or pediatric critical care medicine. Clinical training consisted mostly of rotations providing PNCC consultation (n = 13) or as a provider on the pediatric intensive care unit-based neurointensive care team (n = 2). PNCC-specific didactics were delivered at most institutions (n = 13). All institutions provided training in electroencephalography use in the intensive care unit and declaration of death by neurological criteria (n = 14). Scholarly activity was supported by most programs, including protected time for research (n = 10). CONCLUSIONS: We characterized PNCC fellowship training in the United States and Canada, which in this continuously evolving field, lays the foundation for exploring standardization of training going forward.

Development of Expert-Level Classification of Seizures and Rhythmic and Periodic Patterns During EEG Interpretation.

BACKGROUND AND OBJECTIVES: Seizures (SZs) and other SZ-like patterns of brain activity can harm the brain and contribute to in-hospital death, particularly when prolonged. However, experts qualified to interpret EEG data are scarce. Prior attempts to automate this task have been limited by small or inadequately labeled samples and have not convincingly demonstrated generalizable expert-level performance. There exists a critical unmet need for an automated method to classify SZs and other SZ-like events with expert-level reliability. This study was conducted to develop and validate a computer algorithm that matches the reliability and accuracy of experts in identifying SZs and SZ-like events, known as "ictal-interictal-injury continuum" (IIIC) patterns on EEG, including SZs, lateralized and generalized periodic discharges (LPD, GPD), and lateralized and generalized rhythmic delta activity (LRDA, GRDA), and in differentiating these patterns from non-IIIC patterns. METHODS: We used 6,095 scalp EEGs from 2,711 patients with and without IIIC events to train a deep neural network, SPaRCNet, to perform IIIC event classification. Independent training and test data sets were generated from 50,697 EEG segments, independently annotated by 20 fellowship-trained neurophysiologists. We assessed whether SPaRCNet performs at or above the sensitivity, specificity, precision, and calibration of fellowship-trained neurophysiologists for identifying IIIC events. Statistical performance was assessed by the calibration index and by the percentage of experts whose operating points were below the model's receiver operating characteristic curves (ROCs) and precision recall curves (PRCs) for the 6 pattern classes. RESULTS: SPaRCNet matches or exceeds most experts in classifying IIIC events based on both calibration and discrimination metrics. For SZ, LPD, GPD, LRDA, GRDA, and "other" classes, SPaRCNet exceeds the following percentages of 20 experts-ROC: 45%, 20%, 50%, 75%, 55%, and 40%; PRC: 50%, 35%, 50%, 90%, 70%, and 45%; and calibration: 95%, 100%, 95%, 100%, 100%, and 80%, respectively. DISCUSSION: SPaRCNet is the first algorithm to match expert performance in detecting SZs and other SZ-like events in a representative sample of EEGs. With further development, SPaRCNet may thus be a valuable tool for an expedited review of EEGs. CLASSIFICATION OF EVIDENCE: This study provides Class II evidence that among patients with epilepsy or critical illness undergoing EEG monitoring, SPaRCNet can differentiate (IIIC) patterns from non-IIIC events and expert neurophysiologists.

Interrater Reliability of Expert Electroencephalographers Identifying Seizures and Rhythmic and Periodic Patterns in EEGs.

BACKGROUND AND OBJECTIVES: The validity of brain monitoring using electroencephalography (EEG), particularly to guide care in patients with acute or critical illness, requires that experts can reliably identify seizures and other potentially harmful rhythmic and periodic brain activity, collectively referred to as "ictal-interictal-injury continuum" (IIIC). Previous interrater reliability (IRR) studies are limited by small samples and selection bias. This study was conducted to assess the reliability of experts in identifying IIIC. METHODS: This prospective analysis included 30 experts with subspecialty clinical neurophysiology training from 18 institutions. Experts independently scored varying numbers of ten-second EEG segments as "seizure (SZ)," "lateralized periodic discharges (LPDs)," "generalized periodic discharges (GPDs)," "lateralized rhythmic delta activity (LRDA)," "generalized rhythmic delta activity (GRDA)," or "other." EEGs were performed for clinical indications at Massachusetts General Hospital between 2006 and 2020. Primary outcome measures were pairwise IRR (average percent agreement [PA] between pairs of experts) and majority IRR (average PA with group consensus) for each class and beyond chance agreement (κ). Secondary outcomes were calibration of expert scoring to group consensus, and latent trait analysis to investigate contributions of bias and noise to scoring variability. RESULTS: Among 2,711 EEGs, 49% were from women, and the median (IQR) age was 55 (41) years. In total, experts scored 50,697 EEG segments; the median [range] number scored by each expert was 6,287.5 [1,002, 45,267]. Overall pairwise IRR was moderate (PA 52%, κ 42%), and majority IRR was substantial (PA 65%, κ 61%). Noise-bias analysis demonstrated that a single underlying receiver operating curve can account for most variation in experts' false-positive vs true-positive characteristics (median [range] of variance explained ([Formula: see text]): 95 [93, 98]%) and for most variation in experts' precision vs sensitivity characteristics ([Formula: see text]: 75 [59, 89]%). Thus, variation between experts is mostly attributable not to differences in expertise but rather to variation in decision thresholds. DISCUSSION: Our results provide precise estimates of expert reliability from a large and diverse sample and a parsimonious theory to explain the origin of disagreements between experts. The results also establish a standard for how well an automated IIIC classifier must perform to match experts. CLASSIFICATION OF EVIDENCE: This study provides Class II evidence that an independent expert review reliably identifies ictal-interictal injury continuum patterns on EEG compared with expert consensus.

Association of Cerebral and Systemic Physiology With Quantitative Electroencephalographic Characteristics of Early Posttraumatic Seizures.

PURPOSE: Early posttraumatic seizures (EPTS) occur after pediatric traumatic brain injury and have been associated with unfavorable outcomes. We aimed to characterize the relationship among quantitative EEG characteristics of early posttraumatic seizures, cerebral and somatic physiologic measures. METHODS: Differences in baseline physiologic, neuroimaging, and demographic characteristics between those with and without early posttraumatic seizures were investigated using Mann-Whitney U test or Fisher exact test. Multivariable dynamic structural equations modeling was used to investigate time series associations between ictal quantitative EEG characteristics with intracranial pressure, arterial blood pressure, heart rate (HR), and cerebral regional oximetry. Quantitative EEG characteristics included amplitude, total power, spectral edge frequency, peak value frequency, complexity, and periodicity. RESULTS: Among 72 children, 146 seizures were identified from 19 patients. Early posttraumatic seizures were associated with younger age (P = 0.0034), increased HR (P = 0.0018), and increased Glasgow Outcome Scale-Extended scores (P = 0.0377). Group dynamic structural equations modeling analysis of the first seizure for patients demonstrated that intracranial pressure is negatively associated with spectral edge frequency (standardized regression coefficient -0.12, 99% credible interval [-0.21 to -0.04]), and HR is positively associated with peak value frequency (standardized regression coefficient 0.16, [0.00-0.31]). Among nine patients with seizures arising over the frontal lobe regions, HR was positively associated with peak value frequency (standardized regression coefficient 0.26 [0.02-0.50]) and complexity (standardized regression coefficient 0.14 [0.03-0.26]). Variation in strength and direction of associations was observed between subjects for relationships that were significant during group analysis. CONCLUSIONS: Quantitative EEG characteristics of pediatric early posttraumatic seizures are associated with variable changes in cerebral and systemic physiology, with spectral edge frequency negatively associated with intracranial pressure and peak value frequency positively associated with HR.

A Virtual Community of Practice: An International Educational Series in Pediatric Neurocritical Care.

Pediatric neurocritical care (PNCC) is a rapidly growing field. Challenges posed by the COVID-19 pandemic on trainee exposure to educational opportunities involving direct patient care led to the creative solutions for virtual education supported by guiding organizations such as the Pediatric Neurocritical Care Research Group (PNCRG). Our objective is to describe the creation of an international, peer-reviewed, online PNCC educational series targeting medical trainees and faculty. More than 1600 members of departments such as pediatrics, pediatric critical care, and child neurology hailing from 75 countries across six continents have participated in this series over a 10-month period. We created an online educational channel in PNCC with over 2500 views to date and over 130 followers. This framework could serve as a roadmap for other institutions and specialties seeking to address the ongoing problems of textbook obsolescence relating to the rapid acceleration in knowledge acquisition, as well as those seeking to create new educational content that offers opportunities for an interactive, global audience. Through the creation of a virtual community of practice, we have created an international forum for pediatric healthcare providers to share and learn specialized expertise and best practices to advance global pediatric health.

Acute physiologic prediction of pediatric post-traumatic epilepsy.

OBJECTIVE: Post-traumatic epilepsy (PTE) is a known complication of traumatic brain injury (TBI). Limited physiologic biomarkers have been investigated in relation to pediatric PTE. Our aim is to identify clinical, physiologic and neuroimaging biomarkers predictive of pediatric PTE arising during the acute care phase after injury. METHODS: We performed a retrospective analysis from a prospectively collected clinical database of pediatric patients who underwent multimodality neurologic monitoring that included continuous electroencephalography and intracranial pressure (ICP) monitoring. Biomarkers included hemodynamic vital signs, model-based indices of cerebrovascular pressure reactivity (CVPR) and autonomic function (AF), electroencephalographic abnormalities, and neuroimaging abnormalities on the initial CT scan on day of imaging. Our primary outcome, PTE, was classified as the presence of unprovoked seizures 2 months post-injury or the continued need for antiseizure medications at 12-month post-injury. We utilized univariate logistic regression to identify biomarkers associated with PTE. RESULTS: 61 surviving patients were included in this study, among which 10 (16.4%) developed PTE. We identified that PTE was associated with increased ICP (odds ratio [OR] 1.25, 95% confidence interval [CI] 1.02-1.52), increased pressure reactivity indices (92.53, 2.84->999.99), increased wavelet pressure reactivity indices (121.76, 2.84->999.99), increased CT Marshall scores (1.76, 1.13-2.74), decreased HRsd (0.54, 0.33-0.87) and the presence of epileptiform discharges (8.06, 1.85-35.17), and abnormal sleep spindles (4.88, 1.18-20.00). Whereas early post-traumatic seizures within the first 7 days post-injury were associated with PTE development (7.58, 1.81-39.68), this association was significant for such seizures occurring between 24 and 168 h post-injury (21.47, 4.18-110.38), and not for seizures occurring within 24 h post-injury. Among patients experiencing early post-traumatic seizures, increased time with seizures on surface electroencephalography was associated with PTE development (7.28, 2.05-73.14). We also identified that development of PTE was associated with worsened functional outcomes identified by increased Glasgow Outcome Scale - Extended Pediatric (GOSE-PEDs) scores (3.18, 1.68-8.01). CONCLUSION: Pediatric PTE development is associated with increased ICP, impaired CVPR, low heart rate variability, worsened neuroimaging findings, and electroencephalographic abnormalities identified during intensive care. Further studies are needed to investigate strategies to mitigate pediatric PTE development.

Quantitative Electroencephalography After Pediatric Anterior Circulation Stroke.

OBJECTIVE: Regional differences were investigated in quantitative EEG (QEEG) characteristics and associations of QEEG to hemodynamics after pediatric acute stroke. METHODS: Quantitative EEG was analyzed, including power in delta, theta, alpha, and beta bands, alpha-delta power ratio, total power, and spectral edge frequency from 11 children with unilateral, anterior circulation strokes during the first 24 hours of continuous EEG recording. Differences between injured and uninjured hemispheres were assessed using multivariate dynamic structural equations modeling. Dynamic structural equations modeling was applied to six children with hemorrhagic stroke undergoing arterial blood pressure, heart rate, and cerebral oximetry monitoring to investigate associations between hemodynamics with QEEG adjacent to anterior circulation regions. RESULTS: All patients with acute ischemic stroke ( n = 5) had lower alpha and beta power and spectral edge frequency on injured compared with uninjured regions. This was not consistent after hemorrhagic stroke ( n = 6). All hemorrhagic stroke patients demonstrated negative association of total power with arterial blood pressure within injured regions. No consistency was observed for direction or strength of association in other QEEG measures to arterial blood pressure nor were such consistent relationships observed for any QEEG measure studied in relation to heart rate or cerebral oximetry. CONCLUSIONS: After pediatric anterior circulation acute ischemic stroke, reduced spectral edge frequency and alpha and beta power can be observed on injured as compared with noninjured regions. After pediatric anterior circulation hemorrhagic stroke, total power can be negatively associated with arterial blood pressure within injured regions. Larger studies are needed to understand conditions in which QEEG patterns manifest and relate to hemodynamics and brain penumbra.

Shift in electrocorticography electrode locations after surgical implantation in children.

Interpreting electrocorticography (ECoG) in the context of neuroimaging requires that multimodal information be integrated accurately. However, the implantation of ECoG electrodes can shift the brain impacting the spatial interpretation of electrode locations in the context of pre-implant imaging. We characterized the amount of shift in ECoG electrode locations immediately after implant in a pediatric population. Electrode-shift was quantified as the difference in the electrode locations immediately after surgery (via post-operation CT) compared to the brain surface before the operation (pre-implant T1 MRI). A total of 1140 ECoG contracts were assessed across 18 patients ranging from 3 to 19 (12.1 ± 4.8) years of age who underwent intracranial monitoring in preparation for epilepsy resection surgery. Patients had an average of 63 channels assessed with an average of 5.64 ± 3.27 mm shift from the pre-implant brain surface within 24 h of implant. This shift significantly increased with estimated intracranial volume, but not age. Shift also varied significantly depending of the lobe the contact was over; where contacts on the temporal and frontal lobe had less shift than the parietal. Furthermore, contacts on strips had significantly less shift than those on grids. The shift in the brain surface due to ECoG implantation could lead to a misinterpretation of contact location particularly in patients with larger intracranial volume and for grid contacts over the parietal lobes.

Crystal structure of phosphoglucose isomerase from Trypanosoma brucei complexed with glucose-6-phosphate at 1.6 A resolution.

Enzymes of glycolysis in Trypanosoma brucei have been identified as potential drug targets for African sleeping sickness because glycolysis is the only source of ATP for the bloodstream form of this parasite. Several inhibitors were previously reported to bind preferentially to trypanosomal phosphoglucose isomerase (PGI, the second enzyme in glycolysis) than to mammalian PGIs, which suggests that PGI might make a good target for species-specific drug design. Herein, we report recombinant expression, purification, crystallization and X-ray crystal structure determination of T. brucei PGI. One structure solved at 1.6 A resolution contains a substrate, D-glucose-6-phosphate, in an extended conformation in the active site. A second structure solved at 1.9 A resolution contains a citrate molecule in the active site. The structures are compared with the crystal structures of PGI from humans and from Leishmania mexicana. The availability of recombinant tPGI and its first high-resolution crystal structures are initial steps in considering this enzyme as a potential drug target.