Corticothalamic Connectivity in Aneurysmal Subarachnoid Hemorrhage: Relationship with Disordered Consciousness and Clinical Outcomes.

BACKGROUND: We present an exploratory analysis of the occurrence of early corticothalamic connectivity disruption after aneurysmal subarachnoid hemorrhage (SAH) and its correlation with clinical outcomes. METHODS: We conducted a retrospective study of patients with acute SAH who underwent continuous electroencephalography (EEG) for impairment of consciousness. Only patients undergoing endovascular aneurysm treatment were included. Continuous EEG tracings were reviewed to obtain artifact-free segments. Power spectral analyses were performed, and segments were classified as A (only delta power), B (predominant delta and theta), C (predominant theta and beta), or D (predominant alpha and beta). Each incremental category from A to D implies greater preservation of corticothalamic connectivity. We dichotomized categories as AB for poor connectivity and CD for good connectivity. The modified Rankin Scale score at follow-up and in-hospital mortality were used as outcome measures. RESULTS: Sixty-nine patients were included, of whom 58 had good quality EEG segments for classification: 28 were AB and 30 were CD. Hunt and Hess and World Federation of Neurological Surgeons grades were higher and the initial Glasgow Coma Scale score was lower in the AB group compared with the CD group. AB classification was associated with an adjusted odds ratio of 5.71 (95% confidence interval 1.61-20.30; p < 0.01) for poor outcome (modified Rankin Scale score 4-6) at a median follow-up of 4 months (interquartile range 2-6) and an odds ratio of 5.6 (95% confidence interval 0.98-31.95; p = 0.03) for in-hospital mortality, compared with CD. CONCLUSIONS: EEG spectral-power-based classification demonstrates early corticothalamic connectivity disruption following aneurysmal SAH and may be a mechanism involved in early brain injury. Furthermore, the extent of this disruption appears to be associated with functional outcome and in-hospital mortality in patients with aneurysmal SAH and appears to be a potentially useful predictive tool that must be validated prospectively.

Automated spectrographic seizure detection using convolutional neural networks.

PURPOSE: Non-convulsive seizures are common in critically ill patients, and delays in diagnosis contribute to increased morbidity and mortality. Many intensive care units employ continuous EEG (cEEG) for seizure monitoring. Although cEEG is continuously recorded, it is often reviewed intermittently, which may delay seizure diagnosis and treatment. This may be mitigated with automated seizure detection. In this study, we develop and evaluate convolutional neural networks (CNN) to automate seizure detection on EEG spectrograms. METHODS: Adult EEGs (12 patients, 12 EEGs, 33 seizures) from New-York Presbyterian Hospital (NYP) and pediatric EEGs (22 patients, 130 EEGs, 177 seizures) from Children's Hospital Boston (CHB) were converted into spectrograms. To simulate a telemetry display, seizure and non-seizure events on spectrograms were sequentially sampled as images across a detection window (26,380 total images). Four CNN models of increasing complexity (number of layers) were trained, cross-validated, and tested on CHB and NYP spectrographic images. All CNNs were based on the VGG-net architecture, with adjustments to alleviate overfitting. RESULTS: For spectrographically visible seizures, two CNN models (containing 4 and 7 convolution layers) achieved >90% seizure detection sensitivity and specificity on the CHB test set and >90% sensitivity and 75-80% specificity on the NYP test set. The one CNN model (10 convolution layers) did not converge during training; while another CNN (2 convolution layers) performed poorly (60% sensitivity and 32% specificity) on the NYP test set. CONCLUSIONS: Seizure detection on EEG spectrograms with CNN models is feasible with sensitivity and specificity potentially suitable for clinical use.

Association of Seizure Occurrence with Aneurysm Treatment Modality in Aneurysmal Subarachnoid Hemorrhage Patients.

BACKGROUND: Data on new-onset seizures after treatment of aneurysmal subarachnoid hemorrhage (aSAH) patients are limited and variable. We examined the association between new-onset seizures after aSAH and aneurysm treatment modality, as well their relationship with initial clinical severity of aSAH and outcomes. METHODS: This is a retrospective cohort study of all aSAH patients admitted to our institution over a 6-year period. 'Seizures' were defined as any observed clinical seizure or electrographic seizure on continuous electroencephalogram (cEEG) recordings, as determined by the reviewing neurophysiologist. Subgroup analyses were performed in low-grade (Hunt-Hess 1-3) and high-grade (Hunt-Hess 4-5) patients. Outcomes measures were Glasgow Coma Score (GCS) at intensive care unit (ICU) discharge and modified Rankin Scale (mRS) at outpatient follow-up. RESULTS: There were 282 patients with aSAH; 203 (72.0%) suffered low-grade and 79 (28%) high-grade aSAH. Patients were treated with endovascular coiling (N = 194, 68.8%) or surgical clipping (N = 66, 23.4%). Eighteen (6.4%) patients had seizures, of whom 10 (5.5%) had aneurysm coiling and 7 (10.6%) underwent clipping (p = 0.15). In low-grade patients, seizures occurred less frequently (p = 0.016) and were more common after surgical clipping (p = 0.0089). Seizures correlated with lower GCS upon ICU discharge (p < 0.001), in clipped (p = 0.011) and coiled (p < 0.001) patients and in low-grade aSAH (p < 0.001). Seizures correlated with higher mRS on follow-up (p < 0.001), in clipped (p = 0.032) and coiled (p = 0.004) patients and in low-grade aSAH (p = 0.003). CONCLUSIONS: New-onset seizures after aSAH occurred infrequently, and their incidence after aneurysm clipping versus coiling was not significantly different. However, in low-grade patients, new seizures were more frequently associated with clipping than coiling. Additionally, non-convulsive seizures did not occur in low-grade patients treated with coiling. These findings may explain, in part, previous work suggesting better outcomes in coiled patients and encourage physicians to have a lower threshold for cEEG utilization in low-grade patients suspected to have acute seizures after surgical clipping.

Age-specific cerebral perfusion pressure thresholds and survival in children and adolescents with severe traumatic brain injury*.

OBJECTIVES: Evidence-based traumatic brain injury guidelines support cerebral perfusion pressure thresholds for adults at a class 2 level, but evidence is lacking in younger patients. The purpose of this study is to identify the impact of age-specific cerebral perfusion pressure thresholds on short-term survival among patients with severe traumatic brain injury. DESIGN: Institutional review board-approved, prospective, observational cohort study. SETTING: Level I or II trauma centers in New York State. PATIENTS: Data on all patients with a postresuscitation Glasgow Coma Score less than 9 were added in the Brain Trauma Foundation prospective New York State TBI-trac database. MEASUREMENTS AND MAIN RESULTS: We calculated the survival rates and relative risks of mortality for patients with severe traumatic brain injury based on predefined age-specific cerebral perfusion pressure thresholds. A higher threshold and a lower threshold were defined for each age group: 60 and 50 mm Hg for 12 years old or older, 50 and 35 mm Hg for 6-11 years, and 40 and 30 mm Hg for 0-5 years. Patients were stratified into age groups of 0-11, 12-17, and 18 years old or older. Three exclusive groups of CPP-L (events below low cerebral perfusion pressure threshold), CPP-B (events between high and low cerebral perfusion pressure thresholds), and CPP-H (events above high cerebral perfusion pressure threshold) were defined. As an internal control, we evaluated the associations between cerebral perfusion pressure events and events of hypotension and elevated intracranial pressure. Survival was significantly higher in 0-11 and 18 years old or older age groups for patients with CPP-H events compared with those with CPP-L events. There was a significant decrease in survival with prolonged exposure to CPP-B events for the 0-11 and 18 years old and older age groups when compared with the patients with CPP-H events (p = 0.0001 and p = 0.042, respectively). There was also a significant decrease in survival with prolonged exposure to CPP-L events in all age groups compared with the patients with CPP-H events (p< 0.0001 for 0- to 11-yr olds, p = 0.0240 for 12- to 17-yr olds, and p < 0.0001 for 18-yr old and older age groups). The 12- to 17-year olds had a significantly higher likelihood of survival compared with adults with prolonged exposure to CPP-L events (< 50 mm Hg). CPP-L events were significantly related to systemic hypotension for the 12- to 17-year-old group (p = 0.004) and the 18-year-old and older group (p < 0.0001). CPP-B events were significantly related to systemic hypotension in the 0- to 11-year-old group (p = 0.014). CPP-B and CPP-L events were significantly related to elevated intracranial pressure in all age groups. CONCLUSIONS: Our data provide new evidence that cerebral perfusion pressure targets should be age specific. Furthermore, cerebral perfusion pressure goals above 50 or 60 mm Hg in adults, above 50 mm Hg in 6- to 17-year olds, and above 40 mm Hg in 0- to 5-year olds seem to be appropriate targets for treatment-based studies. Systemic hypotension had an inconsistent relationship to events of low cerebral perfusion pressure, whereas elevated intracranial pressure was significantly related to all low cerebral perfusion pressure events across all age groups. This may impart a clinically important difference in care, highlighting the necessity of controlling intracranial pressure at all times, while targeting systolic blood pressure in specific instances.

Continuous brain tissue oxygenation monitoring in the management of pediatric stroke.

BACKGROUND: Direct invasive monitoring of brain tissue oxygenation (PbtO(2)) has been routinely utilized to predict cerebral ischemia and to prevent secondary injury in patients with traumatic brain injury (TBI) and vasospasm secondary to subarachnoid hemorrhage (SAH). The safety and utility of these devices in the pediatric population have been examined in a few small studies. No studies, however, have examined the use of PbtO(2) monitoring in stroke patients. METHODS: Retrospective chart review of the first two consecutive, critically ill pediatric patients in the pediatric intensive care unit requiring brain tissue oxygen monitoring for newly diagnosed cerebral ischemia. ICP, CPP, PbtO(2), SaO(2), BP, and RR were all continually monitored during their care and were retrospectively collected and reviewed. RESULTS: We present two pediatric stroke patients managed in a critical care setting with PbtO(2) monitoring in addition to ICP, MAP, CPP, and SaO(2). Both patients had multiple events of low brain tissue oxygen (PbtO(2) <20 torr), independent of abnormal values in other monitoring parameters, which required physician intervention. No new ischemic damage occurred after PbtO(2) monitoring began in either patient. CONCLUSIONS: There is currently inadequate data to support the application of PbtO(2) monitoring in children with stroke to prevent progressive ischemia and to improve outcome. However, the positive results for these two patients support the need for further study in this area.

Understanding the role of tumor stem cells in glioblastoma multiforme: a review article.

It has been hypothesized that cancer stem cells (CSC) may account for the pathogenesis underlying various tumors, including GBM. Markers of these CSCs can be potentially used as therapeutic targets. In this review, we discuss the most recent information regarding CSCs, their molecular biology and their potential role in GBM.

Novel gene and gene model detection using a whole genome open reading frame analysis in proteomics.

BACKGROUND: Defining the location of genes and the precise nature of gene products remains a fundamental challenge in genome annotation. Interrogating tandem mass spectrometry data using genomic sequence provides an unbiased method to identify novel translation products. A six-frame translation of the entire human genome was used as the query database to search for novel blood proteins in the data from the Human Proteome Organization Plasma Proteome Project. Because this target database is orders of magnitude larger than the databases traditionally employed in tandem mass spectra analysis, careful attention to significance testing is required. Confidence of identification is assessed using our previously described Poisson statistic, which estimates the significance of multi-peptide identifications incorporating the length of the matching sequence, number of spectra searched and size of the target sequence database. RESULTS: Applying a false discovery rate threshold of 0.05, we identified 282 significant open reading frames, each containing two or more peptide matches. There were 627 novel peptides associated with these open reading frames that mapped to a unique genomic coordinate placed within the start/stop points of previously annotated genes. These peptides matched 1,110 distinct tandem MS spectra. Peptides fell into four categories based upon where their genomic coordinates placed them relative to annotated exons within the parent gene. CONCLUSION: This work provides evidence for novel alternative splice variants in many previously annotated genes. These findings suggest that annotation of the genome is not yet complete and that proteomics has the potential to further add to our understanding of gene structures.