Admission Levels of Interleukin 10 and Amyloid β 1-40 Improve the Outcome Prediction Performance of the Helsinki Computed Tomography Score in Traumatic Brain Injury

Enhancing hospital course and outcome prediction in patients with traumatic brain injury: A machine learning study

PURPOSE

We aimed to use machine learning (ML) algorithms with clinical, lab, and imaging data as input to predict various outcomes in traumatic brain injury (TBI) patients.

METHODS

In this retrospective study, blood samples were analyzed for glial fibrillary acidic protein (GFAP) and ubiquitin C-terminal hydrolase L1 (UCH-L1). The non-contrast head CTs were reviewed by two neuroradiologists for TBI common data elements (CDE). Three outcomes were designed to predict: discharged or admitted for further management (prediction 1), deceased or not deceased (prediction 2), and admission only, prolonged stay, or neurosurgery performed (prediction 3). Five ML models were trained. SHapley Additive exPlanations (SHAP) analyses were used to assess the relative significance of variables.

RESULTS

Four hundred forty patients were used to predict predictions 1 and 2, while 271 patients were used in prediction 3. Due to Prediction 3's hospitalization requirement, deceased and discharged patients could not be utilized. The Random Forest model achieved an average accuracy of 1.00 for prediction 1 and an accuracy of 0.99 for prediction 2. The Random Forest model achieved a mean accuracy of 0.93 for prediction 3. Key features were extracranial injury, hemorrhage, UCH-L1 for prediction 1; The Glasgow Coma Scale, age, GFAP for prediction 2; and GFAP, subdural hemorrhage volume, and pneumocephalus for prediction 3, per SHAP analysis.

CONCLUSION

Combining clinical and laboratory parameters with non-contrast CT CDEs allowed our ML models to accurately predict the designed outcomes of TBI patients. GFAP and UCH-L1 were among the significant predictor variables, demonstrating the importance of these biomarkers.

Blood biomarkers for traumatic brain injury: A narrative review of current evidence

Introduction

A blood-based biomarker (BBBM) test could help to better stratify patients with traumatic brain injury (TBI), reduce unnecessary imaging, to detect and treat secondary insults, predict outcomes, and monitor treatment effects and quality of care.

Research question

What evidence is available for clinical applications of BBBMs in TBI and how to advance this field?

Material and methods

This narrative review discusses the potential clinical applications of core BBBMs in TBI. A literature search in PubMed, Scopus, and ISI Web of Knowledge focused on articles in English with the words "traumatic brain injury" together with the words "blood biomarkers", "diagnostics", "outcome prediction", "extracranial injury" and "assay method" alone-, or in combination.

Results

Glial fibrillary acidic protein (GFAP) combined with Ubiquitin C-terminal hydrolase-L1(UCH-L1) has received FDA clearance to aid computed tomography (CT)-detection of brain lesions in mild (m) TBI. Application of S100B led to reduction of head CT scans. GFAP may also predict magnetic resonance imaging (MRI) abnormalities in CT-negative cases of TBI. Further, UCH-L1, S100B, Neurofilament light (NF-L), and total tau showed value for predicting mortality or unfavourable outcome. Nevertheless, biomarkers have less role in outcome prediction in mTBI. S100B could serve as a tool in the multimodality monitoring of patients in the neurointensive care unit.

Discussion and conclusion

Largescale systematic studies are required to explore the kinetics of BBBMs and their use in multiple clinical groups. Assay development/cross validation should advance the generalizability of those results which implicated GFAP, S100B and NF-L as most promising biomarkers in the diagnostics of TBI.

Inflammation biomarkers IL‑6 and IL‑10 may improve the diagnostic and prognostic accuracy of currently authorized traumatic brain injury tools

Traumatic brain injury (TBI) is currently one of the leading causes of mortality and disability worldwide. At present, no reliable inflammatory or specific molecular neurobiomarker exists in any of the standard models proposed for TBI classification or prognostication. Therefore, the present study was designed to assess the value of a group of inflammatory mediators for evaluating acute TBI, in combination with clinical, laboratory and radiological indices and prognostic clinical scales. In the present single-centre, prospective observational study, 109 adult patients with TBI, 20 adult healthy controls and a pilot group of 17 paediatric patients with TBI from a Neurosurgical Department and two intensive care units of University General Hospital of Heraklion, Greece were recruited. Blood measurements using the ELISA method, of cytokines IL-6, IL-8 and IL-10, ubiquitin C-terminal hydrolase L1 (UCH-L1) and glial fibrillary acidic protein, were performed. Compared with those in healthy control individuals, elevated IL-6 and IL-10 but reduced levels of IL-8 were found on day 1 in adult patients with TBI. In terms of TBI severity classifications, higher levels of IL-6 (P=0.001) and IL-10 (P=0.009) on day 1 in the adult group were found to be associated with more severe TBI according to widely used clinical and functional scales. Moreover, elevated IL-6 and IL-10 in adults were found to be associated with more serious brain imaging findings (r_s<0.442; P<0.007). Subsequent multivariate logistic regression analysis in adults revealed that early-measured (day 1) IL-6 [odds ratio (OR)=0.987; P=0.025] and UCH-L1 (OR=0.993; P=0.032) are significant independent predictors of an unfavourable outcome. In conclusion, results from the present study suggest that inflammatory molecular biomarkers may prove to be valuable diagnostic and prognostic tools for TBI.

GFAP and S100B: What You Always Wanted to Know and Never Dared to Ask

Traumatic brain injury (TBI) is a major global health issue, with outcomes spanning from intracranial bleeding, debilitating sequelae, and invalidity with consequences for individuals, families, and healthcare systems. Early diagnosis of TBI by testing peripheral fluids such as blood or saliva has been the focus of many research efforts, leading to FDA approval for a bench-top assay for blood GFAP and UCH-L1 and a plasma point-of-care test for GFAP. The biomarker S100B has been included in clinical guidelines for mTBI (mTBI) in Europe. Despite these successes, several unresolved issues have been recognized, including the robustness of prior data, the presence of biomarkers in tissues beyond the central nervous system, and the time course of biomarkers in peripheral body fluids. In this review article, we present some of these issues and provide a viewpoint derived from an analysis of existing literature. We focus on two astrocytic proteins, S100B and GFAP, the most commonly employed biomarkers used in mTBI. We also offer recommendations that may translate into a broader acceptance of these clinical tools.

Post-traumatic Neuroinflammation: Relevance to Pediatrics

Both detrimental and beneficial effects of post-traumatic neuroinflammation have become a major research focus as they offer the potential for immediate as well as delayed targeted reparative therapies. Understanding the complex interactions of central and peripheral immunocompetent cells as well as their mediators on brain injury and recovery is complicated by the temporal, regional, and developmental differences in their response to injuries. Microglia, the brain-resident macrophages, have become central in these investigations as they serve a major surveillance function, have the ability to react swiftly to injury, recruit various cellular and chemical mediators, and monitor the reparative/degenerative processes. In this review we describe selected aspects of this burgeoning literature, describing the critical role of cytokines and chemokines, microglia, advances in neuroimaging, genetics and fractal morphology analysis, our research efforts in this area, and selected aspects of pediatric post-traumatic neuroinflammation.

DAMPs and RAGE Pathophysiology at the Acute Phase of Brain Injury: An Overview

Early or primary injury due to brain aggression, such as mechanical trauma, hemorrhage or is-chemia, triggers the release of damage-associated molecular patterns (DAMPs) in the extracellular space. Some DAMPs, such as S100B, participate in the regulation of cell growth and survival but may also trigger cellular damage as their concentration increases in the extracellular space. When DAMPs bind to pattern-recognition receptors, such as the receptor of advanced glycation end-products (RAGE), they lead to non-infectious inflammation that will contribute to necrotic cell clearance but may also worsen brain injury. In this narrative review, we describe the role and ki-netics of DAMPs and RAGE at the acute phase of brain injury. We searched the MEDLINE database for “DAMPs” or “RAGE” or “S100B” and “traumatic brain injury” or “subarachnoid hemorrhage” or “stroke”. We selected original articles reporting data on acute brain injury pathophysiology, from which we describe DAMPs release and clearance upon acute brain injury, and the implication of RAGE in the development of brain injury. We will also discuss the clinical strategies that emerge from this overview in terms of biomarkers and therapeutic perspectives