Selection of children with ultra-severe traumatic brain injury for neurosurgical intervention

Moderate and severe TBI in children and adolescents: The effects of age, sex, and injury severity on patient outcome 6 months after injury

The interaction of age, sex, and outcomes of children with head injury remains incompletely understood and these factors need rigorous evaluation in prognostic models for pediatric head injury. We leveraged our large institutional pediatric TBI population to evaluate age and sex along with a series of predictive factors used in the acute care of injury to describe the response and outcome of children and adolescents with moderate to severe injury. We hypothesized that younger age at injury and male sex would be associated with adverse outcomes and that a novel GCS-based scale incorporating pupillary response (GCS-P) would have superior performance in predicting 6-month outcome. GCS and GCS-P along with established CT scan variables associated with neurologic outcomes were retrospectively reviewed in children (age birth to 18 years) with moderate or severe head injury. GOS-E was prospectively collected 6 months after injury; 570 patients were enrolled in the study, 520 with TBI and 50 with abusive head trauma, each analyzed separately. In the TBI cohort, the median age of patients was 8 years and 42.7% had a severe head injury. Multiple predictors of outcome were identified in univariate analysis; however, based on a multivariate analysis, the GCS was identified as most reliable, outperforming GCS-P, pupil score, and other clinical and CT scan predictors. After stratifying patients for severity of injury by GCS, no age- or sex-related effects were observed in our patient population, except for a trend toward worse outcomes in the neonatal group. Patients with abusive head trauma were more likely to have severe injury on presentation, increased mortality rate, and unfavorable outcome. Additionally, there was clear evidence that secondary injuries, including hypoxia, hypotension, and hypothermia were significantly associated with lower GCS and higher mortality in both AHT and TBI populations. Our findings support the use of GCS to guide clinical decision-making and prognostication in addition to emphasizing the need to stratify head injuries for severity when undertaking outcome studies. Finally, secondary injuries are a clear predictor of poor outcome and how we record and manage these events need to be considered moving forward.

Artificial Intelligence Technologies in Neurosurgery: a Systematic Literature Review Using Topic Modeling. Part II: Research Objectives and Perspectives

The current increase in the number of publications on the use of artificial intelligence (AI) technologies in neurosurgery indicates a new trend in clinical neuroscience.

The aim of the study was to conduct a systematic literature review to highlight the main directions and trends in the use of AI in neurosurgery.