CT is the most common imaging modality in traumatic brain injury (TBI). However, its conventional use requires expert clinical interpretation and does not provide detailed quantitative outputs, which may have prognostic importance. We aimed to use deep learning to reliably and efficiently quantify and detect different lesion types. Patients were recruited between Dec 9, 2014, and Dec 17, 2017, in 60 centres across Europe. We trained and validated an initial convolutional neural network (CNN) on expert manual segmentations (dataset 1). This CNN was used to automatically segment a new dataset of scans, which we then corrected manually (dataset 2). From this dataset, we used a subset of scans to train a final CNN for multiclass, voxel-wise segmentation of lesion types. The performance of this CNN was evaluated on a test subset. Performance was measured for lesion volume quantification, lesion progression, and lesion detection and lesion volume classification. For lesion detection, external validation was done on an independent set of 500 patients from India. 98 scans from one centre were included in dataset 1. Dataset 2 comprised 839 scans from 38 centres: 184 scans were used in the training subset and 655 in the test subset. Compared with manual reference, CNN-derived lesion volumes showed a mean difference of 0·86 mL (95% CI -5·23 to 6·94) for intraparenchymal haemorrhage, 1·83 mL (-12·01 to 15·66) for extra-axial haemorrhage, 2·09 mL (-9·38 to 13·56) for perilesional oedema, and 0·07 mL (-1·00 to 1·13) for intraventricular haemorrhage. We show the ability of a CNN to separately segment, quantify, and detect multiclass haemorrhagic lesions and perilesional oedema. These volumetric lesion estimates allow clinically relevant quantification of lesion burden and progression, with potential applications for personalised treatment strategies and clinical research in TBI. European Union 7th Framework Programme, Hannelore Kohl Stiftung, OneMind, NeuroTrauma Sciences, Integra Neurosciences, European Research Council Horizon 2020.

An increasing number of elderly patients are being affected by traumatic brain injury (TBI) and a significant proportion are on pre-hospital antithrombotic therapy for cardio- or cerebrovascular indications. We have quantified the impact of antiplatelet/anticoagulant (APAC) agents on radiological lesion progression in acute TBI, using a novel, semi-automated approach to volumetric lesion measurement, and explored the impact of use on clinical outcomes in the Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) study. We used a 1:1 propensity-matched cohort design, matching controls to APAC users based on demographics, baseline clinical status, pre-injury comorbidities, and injury severity. Subjects were selected from a pool of patients enrolled in CENTER-TBI with computed tomography (CT) scan at admission and repeated within 7 days of injury. We calculated absolute changes in volume of intraparenchymal, extra-axial, intraventricular, and total intracranial hemorrhage (ICH) between scans, and compared volume of hemorrhagic progression, proportion of patients with significant degree of progression (>25% of initial volume), proportion with new ICH on follow-up CT, as well as clinical course and outcomes. A total of 316 patients were included (158 APAC users; 158 controls). The mean volume of progression was significantly higher in the APAC group for extra-axial (3.1 vs. 1.3 mL, p = 0.01), but not intraparenchymal (3.8 vs. 4.6 mL, p = 0.65), intraventricular (0.2 vs. 0.0 mL, p = 0.79), or total intracranial hemorrhage (ICH; 7.0 vs. 6.0 mL, p = 0.08). More patients had significant hemorrhage growth (54.1 vs. 37.0%, p = 0.003) and delayed ICH (4 of 18 vs. none; p = 0.04) in the APAC group compared with controls, but this was not associated with differences in length of stay (LOS), rates of neurosurgical intervention, mortality or Glasgow Outcome Scale Extended (GOS-E) score at 6 months. Pre-injury use of antithrombotic agents was associated with greater expansion of extra-axial lesions, higher rates of significant hemorrhagic progression, and higher risk of delayed traumatic ICH, but this was not associated with worse clinical course or functional outcomes.

BACKGROUND: Blood biomarkers are of increasing importance in the diagnosis and assessment of traumatic brain injury (TBI). However, the relationship between them and lesions seen on imaging remains unclear. OBJECTIVE: To perform a systematic review of the relationship between blood biomarkers and intracranial lesion types, intracranial lesion injury patterns, volume/number of intracranial lesions, and imaging classification systems. METHODS: We searched Medical Literature Analysis and Retrieval System Online, Excerpta Medica dataBASE, and Cumulative Index to Nursing and Allied Health Literature from inception to May 2021, and the references of included studies were also screened. Heterogeneity in study design, biomarker types, imaging modalities, and analyses inhibited quantitative analysis, with a qualitative synthesis presented. RESULTS: Fifty-nine papers were included assessing one or more biomarker to imaging comparisons per paper: 30 assessed imaging classifications or injury patterns, 28 assessed lesion type, and 11 assessed lesion volume or number. Biomarker concentrations were associated with the burden of brain injury, as assessed by increasing intracranial lesion volume, increasing numbers of traumatic intracranial lesions, and positive correlations with imaging classification scores. There were inconsistent findings associating different biomarkers with specific imaging phenotypes including diffuse axonal injury, cerebral edema, and intracranial hemorrhage. CONCLUSION: Blood-based biomarker concentrations after TBI are consistently demonstrated to correlate burden of intracranial disease. The relation with specific injury types is unclear suggesting a lack of diagnostic specificity and/or is the result of the complex and heterogeneous nature of TBI.

Failure of cerebral autoregulation has been linked to unfavourable outcome after traumatic brain injury. Preliminary evidence from a retrospective single-center analysis suggest that autoregulatory dysfunction may be associated with traumatic lesion expansion, particularly for pericontusional edema. The goal of this study was to explore these associations using prospective multi-center data from the Collaborative European Neurotrauma Effectiveness Research in TBI and to further assess the relationship between autoregulatory failure, lesion progression and patient outcome. 88 subjects from the CENTER-TBI High Resolution ICU cohort were included. All patients had an admission computed tomography (CT) scan and early repeat scan available and high-frequency neurophysiological recordings covering the between-scan interval. Using a novel, semi-automated approach at lesion segmentation, we calculated absolute changes in volume of contusion core, pericontusional edema and extra-axial hemorrhage between the imaging studies. We then evaluated associations between cerebrovascular reactivity metrics and radiological lesion progression using mixed regression. Analyses were adjusted for baseline covariates and non-neurophysiological factors associated with lesion growth using multivariate methods. Impairment in cerebrovascular reactivity was significantly associated with progression of pericontusional edema, and to a lesser degree, intraparenchymal hemorrhage. In contrast, there were no significant associations with extra-axial hemorrhage. The strongest relationships were observed between RAC-based metrics and edema formation. PAx showed weaker, but consistent associations with contusion growth. Cerebrovascular reactivity metrics remained strongly associated with lesion progression after taking into account contributions from non-neurophysiological factors and mean CPP. Total hemorrhagic core and edema volumes on repeat CT were significantly larger in patients who were deceased at 6 months, and the amount of edema was greater in patients with an unfavourable outcome (GOSE 1-4). Our study suggests associations between autoregulatory failure, traumatic edema progression and poor outcome. This is in keeping with findings from a single-center retrospective analysis, providing multi-center prospective data to support those results.

Background: Failure of cerebral autoregulation and progression of intracranial lesion have both been shown to contribute to poor outcome in patients with acute traumatic brain injury (TBI), but the interplay between the two phenomena has not been investigated. Preliminary evidence leads us to hypothesize that brain tissue adjacent to primary injury foci may be more vulnerable to large fluctuations in blood flow in the absence of intact autoregulatory mechanisms. The goal of this study was therefore to assess the influence of cerebrovascular reactivity measures on radiological lesion expansion in a cohort of patients with acute TBI. Methods: We conducted a retrospective cohort analysis on 50 TBI patients who had undergone high-frequency multimodal intracranial monitoring and for which at least two brain computed tomography (CT) scans had been performed in the acute phase of injury. We first performed univariate analyses on the full cohort to identify non-neurophysiological factors (i.e., initial lesion volume, timing of scan, coagulopathy) associated with traumatic lesion growth in this population. In a subset analysis of 23 patients who had intracranial recording data covering the period between the initial and repeat CT scan, we then correlated changes in serial volumetric lesion measurements with cerebrovascular reactivity metrics derived from the pressure reactivity index (PRx), pulse amplitude index (PAx), and RAC (correlation coefficient between the pulse amplitude of intracranial pressure and cerebral perfusion pressure). Using multivariate methods, these results were subsequently adjusted for the non-neurophysiological confounders identified in the univariate analyses. Results: We observed significant positive linear associations between the degree of cerebrovascular reactivity impairment and progression of pericontusional edema. The strongest correlations were observed between edema progression and the following indices of cerebrovascular reactivity between sequential scans: % time PRx > 0.25 (r = 0.69, p = 0.002) and % time PAx > 0.25 (r = 0.64, p = 0.006). These associations remained significant after adjusting for initial lesion volume and mean cerebral perfusion pressure. In contrast, progression of the hemorrhagic core and extra-axial hemorrhage volume did not appear to be strongly influenced by autoregulatory status. Conclusions: Our preliminary findings suggest a possible link between autoregulatory failure and traumatic edema progression, which warrants re-evaluation in larger-scale prospective studies.

There is seemingly contradictory evidence concerning relationships between day-of-injury biomarkers and outcomes after mild traumatic brain injury (mTBI). To address this issue, we examined the association between a panel of biomarkers and multidimensional TBI outcomes. Participants with mTBI (Glasgow coma scores [GCSs] 13-15) were selected from Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury, a European observational study recruiting patients with TBI with indication for brain CT and presentation within 24 hours. Exclusion criteria for this secondary analysis were age younger than 16 years, incomplete biomarker panel, death, or no recorded outcomes. Participants were separated into 2 groups, CT-negative and CT-positive. Multivariable binary logistic regression was used to assess the relation between the log biomarker level (glial fibrillary acidic protein [GFAP], neurofilament light [NfL], neuron-specific enolase [NSE], S100 calcium-binding protein B [S100B], tau, ubiquitin C-terminal hydrolase L1 [UCH-L1]) and dichotomized 6-month outcomes (functional outcomes [GOSE score <8], health-related quality of life [HRQoL; Quality of Life after Brain Injury-Overall Scale (QOLIBRI-OS) score <52, Short-Form 12-Item Survey version 2 Mental Component Summary (SF12v2 MCS) score <40, Short-Form 12-Item Survey version 2 Physical Component Summary (SF12v2 PCS) score <40], persistent postconcussion symptoms [Rivermead Post-Concussion Symptoms Questionnaire score ≥16], anxiety disorder [Generalized Anxiety Disorder-7 (GAD-7) score ≥8], depression [Patient Health Questionnaire-9 (PHQ-9) score ≥10], and post-traumatic stress disorder [PTSD Checklist for DSM-5 (PCL-5) score ≥33]). A total of 1,589 participants (865 CT-negative, 724 CT-positive) were included (77% GCS 15, median age 52 years, 66% male). Higher biomarker levels were associated with a GOSE score <8: CT-negative: S100B (odds ratio [OR] 1.78, 95% CI 1.43-2.23) and UCH-L1 (OR 1.16, 95% CI 1.01-1.33); CT-positive: GFAP (OR 1.22, 95% CI 1.11-1.36), NfL (OR 1.30, 95% CI 1.11-1.52), S100B (OR 1.51, 95% CI 1.23-1.86), tau (OR 1.36, 95% CI 1.17-1.59), and UCH-L1 (OR 1.34, 95% CI 1.17-1.53). In CT-positive participants, positive association was seen between NfL (OR 1.3, 95% CI 1.06-1.60) and UCH-L1 (OR 1.28, 95% CI 1.07-1.54) with QOLIBRI-OS; S100B (OR 1.32, 95% CI 1.02-1.70) with SF12v2 PCS; and NSE (OR 1.52, 95% CI 1.06-2.18) and UCH-L1 (OR 1.21, 95% CI 1.01-1.46) with the GAD-7. However, in CT-negative participants only, negative associations were seen between GFAP and impairment on the QOLIBRI-OS (OR 0.76, 95% CI 0.66-0.88), SF12v2 MCS (OR 0.71, 95% CI 0.61-0.82), SF12v2 PCS (OR 0.79, 95% CI 0.68-0.91), GAD-7 (OR 0.80, 0.68-0.95), PHQ-9 (OR 0.80, 95% CI 0.68-0.93), and PCL-5 (OR 0.80, 95% CI 0.66-0.97). Participants with higher biomarker levels had greater odds of impaired functional recovery. However, in CT-negative participants, higher GFAP concentrations were associated with better HRQoL and less impaired mental health. Further exploration is required of the patient phenotypes that may explain the relationships observed in this analysis.