Predicting outcome in traumatic brain injury: development of a novel computerized tomography classification system (Helsinki computerized tomography score)

Moderate Traumatic Brain Injury in Adult Population: The Latin American Brain Injury Consortium Consensus for Definition and Categorization

Moderate traumatic brain injury (TBI) is a diagnosis that describes diverse patients with heterogeneity of primary injuries. Defined by a Glasgow Coma Scale between 9 and 12, this category includes patients who may neurologically worsen and require increasing intensive care resources and/or emergency neurosurgery. Despite the unique characteristics of these patients, there have not been specific guidelines published before this effort to support decision-making in these patients. A Delphi consensus group from the Latin American Brain Injury Consortium was established to generate recommendations related to the definition and categorization of moderate TBI. Before an in-person meeting, a systematic review of the literature was performed identifying evidence relevant to planned topics. Blinded voting assessed support for each recommendation. A priori the threshold for consensus was set at 80% agreement. Nine PICOT questions were generated by the panel, including definition, categorization, grouping, and diagnosis of moderate TBI. Here, we report the results of our work including relevant consensus statements and discussion for each question. Moderate TBI is an entity for which there is little published evidence available supporting definition, diagnosis, and management. Recommendations based on experts' opinion were informed by available evidence and aim to refine the definition and categorization of moderate TBI. Further studies evaluating the impact of these recommendations will be required.

Assessing outcomes in traumatic brain injury: Helsinki score versus Glasgow coma scale

BACKGROUND

The precision of assessment and prognosis in traumatic brain injury (TBI) is paramount for effective triage and informed therapeutic strategies. While the Glasgow Coma Scale (GCS) remains the cornerstone for TBI evaluation, it overlooks critical primary imaging findings. The Helsinki Score (HS), a novel tool designed to incorporate radiological data, offers a promising approach to predicting TBI outcomes. This study aims to evaluate the prognostic efficacy of HS in comparison to GCS across a substantial TBI patient cohort.

METHODS

This retrospective study encompassed TBI patients treated at our institution between 2008 and 2019, specifically those with an admission GCS of 14 or lower. We assessed both the initial GCS and the HS derived from primary CT scans. Key outcome metrics included the Glasgow Outcome Scale (GOS) and mortality rates at hospital discharge and at 6 and 12-month intervals post-discharge. Predictive performances of GCS and HS were analyzed through Receiver Operating Characteristic (ROC) curves and Kendall tau-b correlation coefficients against each outcome.

RESULTS

The study included 544 patients, with an average age of 62.2 ± 21.5 years, median initial GCS of 14, and a median HS of 3. The mortality rate at discharge stood at 8.6%, with a median GOS of 4. Both GCS and HS demonstrated significant correlations with mortality and GOS outcomes (p < 0.05). Notably, HS showed a markedly superior correlation with mortality (τb = 0.36) compared to GCS (τb = -0.11) and with GOS outcomes (τb = -0.40 for HS vs. τb = 0.33 for GCS). ROC analyses affirmed HS's enhanced predictive accuracy over GCS for both mortality (AUC of 0.79 for HS vs. 0.62 for GCS) and overall outcomes (AUC of 0.77 for HS vs. 0.71 for GCS).

CONCLUSION

The findings validate the HS in a large German cohort and suggest that radiological assessments alone, as exemplified by HS, can surpass the traditional GCS in predicting TBI outcomes. However, the HS, despite its efficacy, lacks the integration of clinical evaluation, a vital component in TBI management. This underscores the necessity for a holistic approach that amalgamates both radiological and clinical insights for a more comprehensive and accurate prognostication in TBI care.

Characteristics of Post-traumatic Shunt-dependent Hydrocephalus After Decompressive Craniectomy: Are Computed Tomography Scoring Systems Predictors?

OBJECTIVE

Traumatic brain injury (TBI) significantly contributes to morbidity rates. While computed tomography (CT) scoring systems have been recognized as predictive factors for TBI outcomes, their association with shunt dependency in patients undergoing decompressive craniectomy (DC) has not been investigated. This study aimed to evaluate the predictive utility of CT scoring systems concerning shunt-dependent hydrocephalus in patients post-DC for TBI.

METHODS

In this retrospective study, we enrolled 162 patients who underwent DC and survived more than 7 days following TBI. The pre-DC CT scans were evaluated using the Marshall, Rotterdam, and Helsinki CT scoring systems. The primary event of interest was shunt-dependent hydrocephalus during the follow-up period, with unfavorable outcomes denoted by a Glasgow Outcome Scale score ranging from 1 to 3.

RESULTS

Analysis of the CT scans showed that the Rotterdam scores had a mean of 4.81 ± 0.91 for the group with shunt-dependent hydrocephalus and 4.41 ± 1.24 for the non-shunt-dependent hydrocephalus group (P = 0.033). However, multivariate logistic regression revealed no significant correlation between the Rotterdam CT score and shunt-dependent hydrocephalus, showing an odds ratio of 1.09 and a 95% confidence interval of 0.71 to 1.67 (P = 0.684). Notably, the Kaplan-Meier outcome curves highlighted a pronounced difference between groups based on shunt dependency (log-rank test: P = 0.012).

CONCLUSIONS

The CT scoring systems proved insufficient for predicting shunt-dependent hydrocephalus following DC for TBI. However, our observations underscore a significant correlation between post-traumatic shunt dependency after DC and an increased incidence of unfavorable outcomes during long-term follow-up.

Early Intravenous Beta-Blockade with Esmolol in Adults with Severe Traumatic Brain Injury: A Phase 2a Intervention Design Study

BACKGROUND

Targeted beta-blockade after severe traumatic brain injury may reduce secondary brain injury by attenuating the sympathoadrenal response. The potential role and optimal dosage for esmolol, a selective, short-acting, titratable beta-1 beta-blocker, as a safe, putative early therapy after major traumatic brain injury has not been assessed.

METHODS

We conducted a single-center, open-label dose-finding study using an adaptive model-based design. Adults (18 years or older) with severe traumatic brain injury and intracranial pressure monitoring received esmolol within 24 h of injury to reduce their heart rate by 15% from baseline of the preceding 4 h while ensuring cerebral perfusion pressure was maintained above 60 mm Hg. In cohorts of three, the starting dosage and dosage increments were escalated according to a prespecified plan in the absence of dose-limiting toxicity. Dose-limiting toxicity was defined as failure to maintain cerebral perfusion pressure, triggering cessation of esmolol infusion. The primary outcome was the maximum tolerated dosage schedule of esmolol, defined as that associated with less than 10% probability of dose-limiting toxicity. Secondary outcomes include 6-month mortality and 6-month extended Glasgow Outcome Scale score.

RESULTS

Sixteen patients (6 [37.5%] female patients; mean age 36 years [standard deviation 13 years]) with a median Glasgow Coma Scale score of 6.5 (interquartile range 5-7) received esmolol. The optimal starting dosage of esmolol was 10 μg/kg/min, with increments every 30 min of 5 μg/kg/min, as it was the highest dosage with less than 10% estimated probability of dose-limiting toxicity (7%). All-cause mortality was 12.5% at 6 months (corresponding to a standardized mortality ratio of 0.63). One dose-limiting toxicity event and no serious adverse hemodynamic effects were seen.

CONCLUSIONS

Esmolol administration, titrated to a heart rate reduction of 15%, is feasible within 24 h of severe traumatic brain injury. The probability of dose-limiting toxicity requiring withdrawal of esmolol when using the optimized schedule is low. Trial registrationI SRCTN, ISRCTN11038397, registered retrospectively January 7, 2021 ( https://www.isrctn.com/ISRCTN11038397 ).

Prediction Value of Initial Serum Levels of SERPINA3 in Intracranial Pressure and Long-Term Neurological Outcomes in Traumatic Brain Injury

Traumatic brain injury (TBI) is a severe neurological condition characterized by inflammation in the central nervous system. SERPINA3 has garnered attention as a potential biomarker for assessing this inflammation. Our study aimed to explore the predictive value of postoperative serum SERPINA3 levels in identifying the risk of cerebral edema and its prognostic implications in TBI. This study is a prospective observational study, including 37 patients with TBI who finally met our criteria. The Glasgow Outcome Scale (GOS), Levels of Cognitive Functioning (LCF), Disability Rating Scale (DRS), and Early Rehabilitation Barthel Index (ERBI) scores at six months after trauma were defined as the main study endpoint. We further calculated the ventricle-to-intracranial-volume ratio (VBR) at 6 months from CT scans. The study included patients with Glasgow Coma Scale (GCS) scores ranging from 3 to 8, who were subsequently categorized into two groups: the critical TBI group (GCS 3-5 points) and the severe TBI group (GCS 6-8 points). Within the critical TBI group, SERPINA3 levels were notably lower. However, among patients with elevated SERPINA3 levels, both the peak intracranial pressure (ICP) and average mannitol consumption were significantly reduced compared with those of patients with lower SERPINA3 levels. In terms of the 6-month outcomes measured via the GOS, LCF, DRS, and ERBI, lower levels of SERPINA3 were indicative of poorer prognosis. Furthermore, we found a negative correlation between serum SERPINA3 levels and the VBR. The receiver operating characteristic (ROC) curve and decision curve analysis (DCA) demonstrated the predictive performance of SERPINA3. In conclusion, incorporating the novel biomarker SERPINA3 alongside traditional assessment tools offers neurosurgeons an effective and easily accessible means, which is readily accessible early on, to predict the risk of intracranial pressure elevation and long-term prognosis in TBI patients.

Traumatic Brain Injury and Traumatic Spinal Cord Injury

OBJECTIVE

This article reviews the mechanisms of primary traumatic injury to the brain and spinal cord, with an emphasis on grading severity, identifying surgical indications, anticipating complications, and managing secondary injury.

LATEST DEVELOPMENTS

Serum biomarkers have emerged for clinical decision making and prognosis after traumatic injury. Cortical spreading depolarization has been identified as a potentially modifiable mechanism of secondary injury after traumatic brain injury. Innovative methods to detect covert consciousness may inform prognosis and enrich future studies of coma recovery. The time-sensitive nature of spinal decompression is being elucidated.

ESSENTIAL POINTS

Proven management strategies for patients with severe neurotrauma in the intensive care unit include surgical decompression when appropriate, the optimization of perfusion, and the anticipation and treatment of complications. Despite validated models, predicting outcomes after traumatic brain injury remains challenging, requiring prognostic humility and a model of shared decision making with surrogate decision makers to establish care goals. Penetrating injuries, especially gunshot wounds, are often devastating and require public health and policy approaches that target prevention.

Development of a Multimodal Machine Learning-Based Prognostication Model for Traumatic Brain Injury Using Clinical Data and Computed Tomography Scans: A CENTER-TBI and CINTER-TBI Study

Computed tomography (CT) is an important imaging modality for guiding prognostication in patients with traumatic brain injury (TBI). However, because of the specialized expertise necessary, timely and dependable TBI prognostication based on CT imaging remains challenging. This study aimed to enhance the efficiency and reliability of TBI prognostication by employing machine learning (ML) techniques on CT images. A retrospective analysis was conducted on the Collaborative European NeuroTrauma Effectiveness Research in TBI (CENTER-TBI) data set (n = 1016). An ML-driven binary classifier was developed to predict favorable or unfavorable outcomes at 6 months post-injury. The prognostic performance was assessed using the area under the curve (AUC) over fivefold cross-validation and compared with conventional models that depend on clinical variables and CT scoring systems. An external validation was performed using the Comparative Indian Neurotrauma Effectiveness Research in Traumatic Brain Injury (CINTER-TBI) data set (n = 348). The developed model achieved superior performance without the necessity for manual CT assessments (AUC = 0.846 [95% CI: 0.843-0.849]) compared with the model based on the clinical and laboratory variables (AUC = 0.817 [95% CI: 0.814-0.820]) and established CT scoring systems requiring manual interpretations (AUC = 0.829 [95% CI: 0.826-0.832] for Marshall and 0.838 [95% CI: 0.835-0.841] for International Mission for Prognosis and Analysis of Clinical Trials in TBI [IMPACT]). The external validation demonstrated the prognostic capacity of the developed model to be significantly better (AUC = 0.859 [95% CI: 0.857-0.862]) than the model using clinical variables (AUC = 0.809 [95% CI: 0.798-0.820]). This study established an ML-based model that provides efficient and reliable TBI prognosis based on CT scans, with potential implications for earlier intervention and improved patient outcomes.

Imaging Findings in Acute Traumatic Brain Injury: a National Institute of Neurological Disorders and Stroke Common Data Element-Based Pictorial Review and Analysis of Over 4000 Admission Brain Computed Tomography Scans from the Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) Study

In 2010, the National Institute of Neurological Disorders and Stroke (NINDS) created a set of common data elements (CDEs) to help standardize the assessment and reporting of imaging findings in traumatic brain injury (TBI). However, as opposed to other standardized radiology reporting systems, a visual overview and data to support the proposed standardized lexicon are lacking. We used over 4000 admission computed tomography (CT) scans of patients with TBI from the Collaborative European NeuroTrauma Effectiveness Research in Traumatic Brain Injury (CENTER-TBI) study to develop an extensive pictorial overview of the NINDS TBI CDEs, with visual examples and background information on individual pathoanatomical lesion types, up to the level of supplemental and emerging information (e.g., location and estimated volumes). We documented the frequency of lesion occurrence, aiming to quantify the relative importance of different CDEs for characterizing TBI, and performed a critical appraisal of our experience with the intent to inform updating of the CDEs. In addition, we investigated the co-occurrence and clustering of lesion types and the distribution of six CT classification systems. The median age of the 4087 patients in our dataset was 50 years (interquartile range, 29-66; range, 0-96), including 238 patients under 18 years old (5.8%). Traumatic subarachnoid hemorrhage (45.3%), skull fractures (37.4%), contusions (31.3%), and acute subdural hematoma (28.9%) were the most frequently occurring CT findings in acute TBI. The ranking of these lesions was the same in patients with mild TBI (baseline Glasgow Coma Scale [GCS] score 13-15) compared with those with moderate-severe TBI (baseline GCS score 3-12), but the frequency of occurrence was up to three times higher in moderate-severe TBI. In most TBI patients with CT abnormalities, there was co-occurrence and clustering of different lesion types, with significant differences between mild and moderate-severe TBI patients. More specifically, lesion patterns were more complex in moderate-severe TBI patients, with more co-existing lesions and more frequent signs of mass effect. These patients also had higher and more heterogeneous CT score distributions, associated with worse predicted outcomes. The critical appraisal of the NINDS CDEs was highly positive, but revealed that full assessment can be time consuming, that some CDEs had very low frequencies, and identified a few redundancies and ambiguity in some definitions. Whilst primarily developed for research, implementation of CDE templates for use in clinical practice is advocated, but this will require development of an abbreviated version. In conclusion, with this study, we provide an educational resource for clinicians and researchers to help assess, characterize, and report the vast and complex spectrum of imaging findings in patients with TBI. Our data provides a comprehensive overview of the contemporary landscape of TBI imaging pathology in Europe, and the findings can serve as empirical evidence for updating the current NINDS radiologic CDEs to version 3.0.

Guidelines for Neuroprognostication in Critically Ill Adults with Moderate-Severe Traumatic Brain Injury

BACKGROUND

Moderate-severe traumatic brain injury (msTBI) carries high morbidity and mortality worldwide. Accurate neuroprognostication is essential in guiding clinical decisions, including patient triage and transition to comfort measures. Here we provide recommendations regarding the reliability of major clinical predictors and prediction models commonly used in msTBI neuroprognostication, guiding clinicians in counseling surrogate decision-makers.

METHODS

Using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) methodology, we conducted a systematic narrative review of the most clinically relevant predictors and prediction models cited in the literature. The review involved framing specific population/intervention/comparator/outcome/timing/setting (PICOTS) questions and employing stringent full-text screening criteria to examine the literature, focusing on four GRADE criteria: quality of evidence, desirability of outcomes, values and preferences, and resource use. Moreover, good practice recommendations addressing the key principles of neuroprognostication were drafted.

RESULTS

After screening 8125 articles, 41 met our eligibility criteria. Ten clinical variables and nine grading scales were selected. Many articles varied in defining "poor" functional outcomes. For consistency, we treated "poor" as "unfavorable". Although many clinical variables are associated with poor outcome in msTBI, only the presence of bilateral pupillary nonreactivity on admission, conditional on accurate assessment without confounding from medications or injuries, was deemed moderately reliable for counseling surrogates regarding 6-month functional outcomes or in-hospital mortality. In terms of prediction models, the Corticosteroid Randomization After Significant Head Injury (CRASH)-basic, CRASH-CT (CRASH-basic extended by computed tomography features), International Mission for Prognosis and Analysis of Clinical Trials in TBI (IMPACT)-core, IMPACT-extended, and IMPACT-lab models were recommended as moderately reliable in predicting 14-day to 6-month mortality and functional outcomes at 6 months and beyond. When using "moderately reliable" predictors or prediction models, the clinician must acknowledge "substantial" uncertainty in the prognosis.

CONCLUSIONS

These guidelines provide recommendations to clinicians on the formal reliability of individual predictors and prediction models of poor outcome when counseling surrogates of patients with msTBI and suggest broad principles of neuroprognostication.

The Australian Traumatic Brain Injury Initiative: Systematic Review of Predictive Value of Biological Markers for People With Moderate-Severe Traumatic Brain Injury

The Australian Traumatic Brain Injury Initiative (AUS-TBI) aims to co-design a data resource to predict outcomes for people with moderate-severe traumatic brain injury (TBI) across Australia. Fundamental to this resource is the data dictionary, which is an ontology of data items. Here, we report the systematic review and consensus process for inclusion of biological markers in the data dictionary. Standardized database searches were implemented from inception through April 2022. English-language studies evaluating association between a fluid, tissue, or imaging marker and any clinical outcome in at least 10 patients with moderate-severe TBI were included. Records were screened using a prioritization algorithm and saturation threshold in Research Screener. Full-length records were then screened in Covidence. A pre-defined algorithm was used to assign a judgement of predictive value to each observed association, and high-value predictors were discussed in a consensus process. Searches retrieved 106,593 records; 1,417 full-length records were screened, resulting in 546 included records. Two hundred thirty-nine individual markers were extracted, evaluated against 101 outcomes. Forty-one markers were judged to be high-value predictors of 15 outcomes. Fluid markers retained following the consensus process included ubiquitin C-terminal hydrolase L1 (UCH-L1), S100, and glial fibrillary acidic protein (GFAP). Imaging markers included computed tomography (CT) scores (e.g., Marshall scores), pathological observations (e.g., hemorrhage, midline shift), and magnetic resonance imaging (MRI) classification (e.g., diffuse axonal injury). Clinical context and time of sampling of potential predictive indicators are important considerations for utility. This systematic review and consensus process has identified fluid and imaging biomarkers with high predictive value of clinical and long-term outcomes following moderate-severe TBI.

Impact of Automated Prognostication on Traumatic Brain Injury Care: A Focus Group Study

BACKGROUND

Prognosticating outcomes for traumatic brain injury (TBI) patients is challenging due to the required specialized skills and variability among clinicians. Recent attempts to standardize TBI prognosis have leveraged machine learning (ML) methodologies. This study evaluates the necessity and influence of ML-assisted TBI prognostication through healthcare professionals' perspectives via focus group discussions.

METHODS

Two virtual focus groups included ten key TBI care stakeholders (one neurosurgeon, two emergency clinicians, one internist, two radiologists, one registered nurse, two researchers in ML and healthcare and one patient representative). They answered six open-ended questions about their perceptions and potential ML use in TBI prognostication. Transcribed focus group discussions were thematically analyzed using qualitative data analysis software.

RESULTS

The study captured diverse perceptions and interests in TBI prognostication across clinical specialties. Notably, certain clinicians who currently do not prognosticate expressed an interest in doing so independently provided they had access to ML support. Concerns included ML's accuracy and the need for proficient ML researchers in clinical settings. The consensus suggested using ML as a secondary consultation tool and promoting collaboration with internal or external research resources. Participants believed ML prognostication could enhance disposition planning and standardize care regardless of clinician expertise or injury severity. There was no evidence of perceived bias or interference during the discussions.

CONCLUSION

Our findings revealed an overall positive attitude toward ML-based prognostication. Despite raising multiple concerns, the focus group discussions were particularly valuable in underscoring the potential of ML in democratizing and standardizing TBI prognosis practices.

Nrf2 functions as a pyroptosis-related mediator in traumatic brain injury and is correlated with cytokines and disease severity: a bioinformatics analysis and retrospective clinical study

Background

Traumatic brain injury (TBI) is a serious hazard to human health. Evidence has accumulated that pyroptosis plays an important role in brain trauma. The aim of this study is to screen potential key molecules between TBI and pyroptosis, and further explore their relationships with disease severity and cytokines.

Methods

To acquire differentially expressed genes (DEGs) before and after brain injury, the GSE89866 dataset was downloaded from the Gene Expression Omnibus (GEO) database. Meanwhile, pyroptosis-related genes were obtained from the GeneCards database, and the intersected genes were identified as differentially expressed pyroptosis-related genes (DEPGs). Moreover, the hub genes were screened via four algorithms (namely Maximum Clique Centrality, Edge Percolated Component, BottleNeck and EcCentricity) in Cytoscape software. Blood levels of Nrf2 were measured by ELISA using a commercially available kit. Finally, we further investigated the correlation between Nrf2 levels and medical indicators in TBI such as clinical characteristics, inflammatory cytokines, and severity.

Results

Altogether, we found 1,795 DEGs in GSE89866 and 98 pyroptosis-related genes in the GeneCards database. Subsequently, four hub genes were obtained, and NFE2L2 was adopted for further clinical study. By using Kruskal-Wallis test and Spearman correlation test, we found that the serum Nrf2 levels in severe TBI patients were negatively correlated with GCS scores. On the contrary, there was a positive correlation between serum Nrf2 levels and pupil parameters, Helsinki CT scores, IL-1 β and IL-18.

Conclusions

In summary, bioinformatic analyses showed NFE2L2 plays a significant role in the pathology of TBI. The clinical research indicated the increase in serum Nrf2 levels was closely related to the severity of trauma and cytokines. We speculate that serum Nrf2 may serve as a promising biochemical marker for the assessment of TBI in clinical practice.

Consensus-Based Development of a Global Registry for Traumatic Brain Injury: Establishment, Protocol, and Implementation

BACKGROUND AND OBJECTIVES

Global disparity exists in the demographics, pathology, management, and outcomes of surgically treated traumatic brain injury (TBI). However, the factors underlying these differences, including intervention effectiveness, remain unclear. Establishing a more accurate global picture of the burden of TBI represents a challenging task requiring systematic and ongoing data collection of patients with TBI across all management modalities. The objective of this study was to establish a global registry that would enable local service benchmarking against a global standard, identification of unmet need in TBI management, and its evidence-based prioritization in policymaking.

METHODS

The registry was developed in an iterative consensus-based manner by a panel of neurotrauma professionals. Proposed registry objectives, structure, and data points were established in 2 international multidisciplinary neurotrauma meetings, after which a survey consisting of the same data points was circulated within the global neurotrauma community. The survey results were disseminated in a final meeting to reach a consensus on the most pertinent registry variables.

RESULTS

A total of 156 professionals from 53 countries, including both high-income countries and low- and middle-income countries, responded to the survey. The final consensus-based registry includes patients with TBI who required neurosurgical admission, a neurosurgical procedure, or a critical care admission. The data set comprised clinically pertinent information on demographics, injury characteristics, imaging, treatments, and short-term outcomes. Based on the consensus, the Global Epidemiology and Outcomes following Traumatic Brain Injury (GEO-TBI) registry was established.

CONCLUSION

The GEO-TBI registry will enable high-quality data collection, clinical auditing, and research activity, and it is supported by the World Federation of Neurosurgical Societies and the National Institute of Health Research Global Health Program. The GEO-TBI registry ( https://geotbi.org ) is now open for participant site recruitment. Any center involved in TBI management is welcome to join the collaboration to access the registry.

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.

Outlier Analysis for Acute Blood Biomarkers of Moderate and Severe Traumatic Brain Injury

Blood biomarkers have been studied to improve the clinical assessment and prognostication of patients with moderate-severe traumatic brain injury (mo/sTBI). To assess their clinical usability, one needs to know of potential factors that might cause outlier values and affect clinical decision making. In a prospective study, we recruited patients with mo/sTBI (n = 85) and measured the blood levels of eight protein brain pathophysiology biomarkers, including glial fibrillary acidic protein (GFAP), S100 calcium-binding protein B (S100B), neurofilament light (Nf-L), heart-type fatty acid-binding protein (H-FABP), interleukin-10 (IL-10), total tau (T-tau), amyloid β40 (Aβ40) and amyloid β42 (Aβ42), within 24 h of admission. Similar analyses were conducted for controls (n = 40) with an acute orthopedic injury without any head trauma. The patients with TBI were divided into subgroups of normal versus abnormal (n = 9/76) head computed tomography (CT) and favorable (Glasgow Outcome Scale Extended [GOSE] 5-8) versus unfavorable (GOSE <5) (n = 38/42, 5 missing) outcome. Outliers were sought individually from all subgroups from and the whole TBI patient population. Biomarker levels outside Q1 - 1.5 interquartile range (IQR) or Q3 + 1.5 IQR were considered as outliers. The medical records of each outlier patient were reviewed in a team meeting to determine possible reasons for outlier values. A total of 29 patients (34%) combined from all subgroups and 12 patients (30%) among the controls showed outlier values for one or more of the eight biomarkers. Nine patients with TBI and five control patients had outlier values in more than one biomarker (up to 4). All outlier values were > Q3 + 1.5 IQR. A logical explanation was found for almost all cases, except the amyloid proteins. Explanations for outlier values included extremely severe injury, especially for GFAP and S100B. In the case of H-FABP and IL-10, the explanation was extracranial injuries (thoracic injuries for H-FABP and multi-trauma for IL-10), in some cases these also were associated with abnormally high S100B. Timing of sampling and demographic factors such as age and pre-existing neurological conditions (especially for T-tau), explained some of the abnormally high values especially for Nf-L. Similar explanations also emerged in controls, where the outlier values were caused especially by pre-existing neurological diseases. To utilize blood-based biomarkers in clinical assessment of mo/sTBI, very severe or fatal TBIs, various extracranial injuries, timing of sampling, and demographic factors such as age and pre-existing systemic or neurological conditions must be taken into consideration. Very high levels seem to be often associated with poor prognosis and mortality (GFAP and S100B).

Evaluation of Morlet Wavelet Analysis for Artifact Detection in Low-Frequency Commercial Near-Infrared Spectroscopy Systems

Regional cerebral oxygen saturation (rSO2), a method of cerebral tissue oxygenation measurement, is recorded using non-invasive near-infrared Spectroscopy (NIRS) devices. A major limitation is that recorded signals often contain artifacts. Manually removing these artifacts is both resource and time consuming. The objective was to evaluate the applicability of using wavelet analysis as an automated method for simple signal loss artifact clearance of rSO2 signals obtained from commercially available devices. A retrospective observational study using existing populations (healthy control (HC), elective spinal surgery patients (SP), and traumatic brain injury patients (TBI)) was conducted. Arterial blood pressure (ABP) and rSO2 data were collected in all patients. Wavelet analysis was determined to be successful in removing simple signal loss artifacts using wavelet coefficients and coherence to detect signal loss artifacts in rSO2 signals. The removal success rates in HC, SP, and TBI populations were 100%, 99.8%, and 99.7%, respectively (though it had limited precision in determining the exact point in time). Thus, wavelet analysis may prove to be useful in a layered approach NIRS signal artifact tool utilizing higher-frequency data; however, future work is needed.

"COAGULATION": a mnemonic device for treating coagulation disorders following traumatic brain injury-a narrative-based method in the intensive care unit

Introduction

Coagulopathy associated with isolated traumatic brain injury (C-iTBI) is a frequent complication associated with poor outcomes, primarily due to its role in the development or progression of haemorrhagic brain lesions. The independent risk factors for its onset are age, severity of traumatic brain injury (TBI), volume of fluids administered during resuscitation, and pre-injury use of antithrombotic drugs. Although the pathophysiology of C-iTBI has not been fully elucidated, two distinct stages have been identified: an initial hypocoagulable phase that begins within the first 24 h, dominated by platelet dysfunction and hyperfibrinolysis, followed by a hypercoagulable state that generally starts 72 h after the trauma. The aim of this study was to design an acronym as a mnemonic device to provide clinicians with an auxiliary tool in the treatment of this complication.

Methods

A narrative analysis was performed in which intensive care physicians were asked to list the key factors related to C-iTBI. The initial sample was comprised of 33 respondents. Respondents who were not physicians, not currently working in or with experience in coagulopathy were excluded. Interviews were conducted for a month until the sample was saturated. Each participant was asked a single question: Can you identify a factor associated with coagulopathy in patients with TBI? Factors identified by respondents were then submitted to a quality check based on published studies and proven evidence. Because all the factors identified had strong support in the literature, none was eliminated. An acronym was then developed to create the mnemonic device.

Results and conclusion

Eleven factors were identified: cerebral computed tomography, oral anticoagulant & antiplatelet use, arterial blood pressure (Hypotension), goal-directed haemostatic therapy, use fluids cautiously, low calcium levels, anaemia-transfusion, temperature, international normalised ratio (INR), oral antithrombotic reversal, normal acid-base status, forming the acronym "Coagulation." This acronym is a simple mnemonic device, easy to apply for anyone facing the challenge of treating patients of moderate or severe TBI on a daily basis.

Effect of antiplatelet and anticoagulant medication use on injury severity and mortality in patients with traumatic brain injury treated in the intensive care unit

BACKGROUND

Antiplatelet and anticoagulant medication are increasingly common and can increase the risks of morbidity and mortality in traumatic brain injury (TBI) patients. Our study aimed to quantify the association of antiplatelet or anticoagulant use in intensive care unit (ICU)-treated TBI patients with 1-year mortality and head CT findings.

METHOD

We conducted a retrospective, multicenter observational study using the Finnish Intensive Care Consortium database. We included adult TBI patients admitted to four university hospital ICUs during 2003-2013. The patients were followed up until the end of 2016. The national drug reimbursement database provided information on prescribed medication for our study. We used multivariable logistic regression models to assess the association between TBI severity, prescribed antiplatelet and anticoagulant medication, and their association with 1-year mortality.

RESULTS

Of 3031 patients, 128 (4%) had antiplatelet and 342 (11%) anticoagulant medication before their TBI. Clopidogrel (2%) and warfarin (9%) were the most common antiplatelets and anticoagulants. Three patients had direct oral anticoagulant (DOAC) medication. The median age was higher among antiplatelet/anticoagulant users than in non-users (70 years vs. 52 years, p < 0.001), and their head CT findings were more severe (median Helsinki CT score 3 vs. 2, p < 0.05). In multivariable analysis, antiplatelets (OR 1.62, 95% CI 1.02-2.58) and anticoagulants (OR 1.43, 95% CI 1.06-1.94) were independently associated with higher odds of 1-year mortality. In a sensitivity analysis including only patients over 70, antiplatelets (OR 2.28, 95% CI 1.16-4.22) and anticoagulants (1.50, 95% CI 0.97-2.32) were associated with an increased risk of 1-year mortality.

CONCLUSIONS

Both antiplatelet and anticoagulant use before TBI were risk factors in our study for 1-year mortality. Antiplatelet and anticoagulation medication users had a higher radiological intracranial injury burden than non-users defined by the Helsinki CT score. Further investigation on the effect of DOACs on mortality should be done in ICU-treated TBI patients.

Computed tomography findings as early predictors of long-term language impairment in patients with traumatic brain injury

This study aims to assess the relationship between computed tomography (CT) findings, during the acute phase of hospitalization, and long-term language impairment in people with traumatic brain injury (TBI). Another aim was to assess the receptive and expressive abilities of subjects with TBI based on the location of the injury. This is a retrospective observational study including 49 participants with TBI due to war injuries. The Arabic Diagnostic Aphasia Battery (A-DAB-1) was administered to the participants and the Helsinki CT score was computed to quantify brain damage. The results showed that the Helsinki CT score was negatively correlated with the total score of the A-DAB-1 (r = -0.544, p-value < 0.0001). Simple linear regression supported such findings and reflected an inversely proportional relationship between both variables (p-value < 0.0001). When compared with subjects having right hemisphere damage, subjects with left hemisphere and bilateral brain damage performed more poorly on language tasks respectively as follows: A-DAB-1 overall score (92.08-66.08-70.28, p-value = 0.021), Content of descriptive speech (9.57-6.69-7.22, p-value = 0.034), Verbal fluency (6.57-3.54-3.89, p-value = 0.002), Auditory comprehension (9.71-7.54-7.78, p-value = 0.039), Complex auditory commands (9.71-7.65-7.56, p-value = 0.043), Repetition (9.75-7.08-7.61, p-value = 0.036), Naming (9.93-7.15-8.11, p-value = 0.046). Following TBI, CT findings on admission can significantly predict long-term language abilities, with left side lesions inducing poorer outcomes.

Machine learning-based models to predict the need for neurosurgical intervention after moderate traumatic brain injury

Background and Aims

Traumatic brain injury (TBI) is a widespread global health issue with significant economic consequences. However, no existing model exists to predict the need for neurosurgical intervention in moderate TBI patients with positive initial computed tomography scans. This study determines the efficacy of machine learning (ML)-based models in predicting the need for neurosurgical intervention.

Methods

This is a retrospective study of patients admitted to the neuro-intensive care unit of Emtiaz Hospital, Shiraz, Iran, between January 2018 and December 2020. The most clinically important variables from patients that met our inclusion and exclusion criteria were collected and used as predictors. We developed models using multilayer perceptron, random forest, support vector machines (SVM), and logistic regression. To evaluate the models, their F1-score, sensitivity, specificity, and accuracy were assessed using a fourfold cross-validation method.

Results

Based on predictive models, SVM showed the highest performance in predicting the need for neurosurgical intervention, with an F1-score of 0.83, an area under curve of 0.93, sensitivity of 0.82, specificity of 0.84, a positive predictive value of 0.83, and a negative predictive value of 0.83.

Conclusion

The use of ML-based models as decision-making tools can be effective in predicting with high accuracy whether neurosurgery will be necessary after moderate TBIs. These models may ultimately be used as decision-support tools to evaluate early intervention in TBI patients.

Comparison of the predictive value of the Helsinki, Rotterdam, and Stockholm CT scores in predicting 6-month outcomes in patients with blunt traumatic brain injuries

PURPOSE

Despite advances in modern medicine, traumatic brain injuries (TBIs) are still a major medical problem. Early diagnosis of TBI is crucial for clinical decision-making and prognosis. This study aims to compare the predictive value of Helsinki, Rotterdam, and Stockholm CT scores in predicting the 6-month outcomes in blunt TBI patients.

METHODS

This cohort study was conducted on blunt TBI patients of 15 years or older. All of them were admitted to the surgical emergency department of Shahid Beheshti Hospital in Kashan, Iran from 2020 to 2021 and had abnormal trauma-related findings on brain CT images. The patients' demographic data such as age, gender, history of comorbid conditions, mechanism of trauma, Glasgow coma scale, CT images, length of hospital stay, and surgical procedures were recorded. The Helsinki, Rotterdam, and Stockholm CT scores were simultaneously determined according to the existing guidelines. The included patients' 6-month outcome was determined using the Glasgow outcome scale extended. M Data were analyzed by SPSS software version 16.0. Sensitivity, specificity, negative/positive predictive value and the area under the receiver operating characteristic curve were calculated for each test. The Kappa agreement coefficient and Kuder Richardson-20 were used to compare the scoring systems.

RESULTS

Altogether 171 TBI patients met the inclusion and exclusion criteria, with the mean age of (44.9 ± 20.2) years. Most patients were male (80.7%), had traffic related injuries (83.1%) and mild TBIs (64.3%). Patients with lower Glasgow coma scale had higher Helsinki, Rotterdam, and Stockholm CT scores and lower Glasgow outcome scale extended scores. Among all the scoring systems, the Helsinki and Stockholm scores showed the highest agreement in predicting patients' outcomes (kappa = 0.657, p < 0.001). The Rotterdam scoring system had the highest sensitivity (90.1%) in predicting death of TBI patients, whereas the Helsinki scoring system had the highest sensitivity (89.8%) in predicting the 6-month outcome in TBI patients.

CONCLUSION

The Rotterdam scoring system was superior in predicting death in TBI patients, whereas the Helsinki scoring system was more sensitive in predicting the 6-month outcome.

Validation of a deep learning model for traumatic brain injury detection and NIRIS grading on non-contrast CT: a multi-reader study with promising results and opportunities for improvement

PURPOSE

This study aimed to assess and externally validate the performance of a deep learning (DL) model for the interpretation of non-contrast computed tomography (NCCT) scans of patients with suspicion of traumatic brain injury (TBI).

METHODS

This retrospective and multi-reader study included patients with TBI suspicion who were transported to the emergency department and underwent NCCT scans. Eight reviewers, with varying levels of training and experience (two neuroradiology attendings, two neuroradiology fellows, two neuroradiology residents, one neurosurgery attending, and one neurosurgery resident), independently evaluated NCCT head scans. The same scans were evaluated using the version 5.0 of the DL model icobrain tbi. The establishment of the ground truth involved a thorough assessment of all accessible clinical and laboratory data, as well as follow-up imaging studies, including NCCT and magnetic resonance imaging, as a consensus amongst the study reviewers. The outcomes of interest included neuroimaging radiological interpretation system (NIRIS) scores, the presence of midline shift, mass effect, hemorrhagic lesions, hydrocephalus, and severe hydrocephalus, as well as measurements of midline shift and volumes of hemorrhagic lesions. Comparisons using weighted Cohen's kappa coefficient were made. The McNemar test was used to compare the diagnostic performance. Bland-Altman plots were used to compare measurements.

RESULTS

One hundred patients were included, with the DL model successfully categorizing 77 scans. The median age for the total group was 48, with the omitted group having a median age of 44.5 and the included group having a median age of 48. The DL model demonstrated moderate agreement with the ground truth, trainees, and attendings. With the DL model's assistance, trainees' agreement with the ground truth improved. The DL model showed high specificity (0.88) and positive predictive value (0.96) in classifying NIRIS scores as 0-2 or 3-4. Trainees and attendings had the highest accuracy (0.95). The DL model's performance in classifying various TBI CT imaging common data elements was comparable to that of trainees and attendings. The average difference for the DL model in quantifying the volume of hemorrhagic lesions was 6.0 mL with a wide 95% confidence interval (CI) of - 68.32 to 80.22, and for midline shift, the average difference was 1.4 mm with a 95% CI of - 3.4 to 6.2.

CONCLUSION

While the DL model outperformed trainees in some aspects, attendings' assessments remained superior in most instances. Using the DL model as an assistive tool benefited trainees, improving their NIRIS score agreement with the ground truth. Although the DL model showed high potential in classifying some TBI CT imaging common data elements, further refinement and optimization are necessary to enhance its clinical utility.

Maximizing the Clinical Value of Blood-Based Biomarkers for Mild Traumatic Brain Injury

Mild traumatic brain injury (TBI) and concussion can have serious consequences that develop over time with unpredictable levels of recovery. Millions of concussions occur yearly, and a substantial number result in lingering symptoms, loss of productivity, and lower quality of life. The diagnosis may not be made for multiple reasons, including due to patient hesitancy to undergo neuroimaging and inability of imaging to detect minimal damage. Biomarkers could fill this gap, but the time needed to send blood to a laboratory for analysis made this impractical until point-of-care measurement became available. A handheld blood test is now on the market for diagnosis of concussion based on the specific blood biomarkers glial fibrillary acidic protein (GFAP) and ubiquitin carboxyl terminal hydrolase L1 (UCH-L1). This paper discusses rapid blood biomarker assessment for mild TBI and its implications in improving prediction of TBI course, avoiding repeated head trauma, and its potential role in assessing new therapeutic options. Although we focus on the Abbott i-STAT TBI plasma test because it is the first to be FDA-cleared, our discussion applies to any comparable test systems that may become available in the future. The difficulties in changing emergency department protocols to include new technology are addressed.

Prediction of Mortality and Morbidity After Severe Traumatic Brain Injury: A Comparison Between Rotterdam and Richmond Computed Tomography Scan Scoring System

OBJECTIVE

Accurate prediction of the morbidity and mortality outcomes of traumatic brain injury patients is still challenging. In the present study, we aimed to compare the predictive value of the Richmond and Rotterdam scoring systems as two novel computed tomography-based predictive models.

METHODS

We retrospectively analyzed 1400 subjects who suffered from severe traumatic brain injury and were admitted to Emtiaz Hospital, a tertiary referral trauma center in Shiraz, south of Iran, from January 2018 to December 2019. We evaluated the 1-month results; considering two primary factors: mortality and morbidity. The patients' condition was the basis for this assessment. We conducted a logistic regression analysis to determine the association between scoring systems and outcomes. To determine the optimal threshold value, we utilized the receiver operating characteristic curve model.

RESULTS

The mean age of participants was 36.61 ± 17.58 years, respectively. Concerning predicting the mortality rate, the area under the curve (AUC) for the Rotterdam score was relatively low 0.64 (95% confidence interval: 0.60, 0.67), while the Richmond score had a higher AUC 0.74 (0.71-0.77), which demonstrated the superiority of this scoring system. Moreover, the Richmond score was more accurate for predicting 1-month morbidity with AUC: 0.71 (0.69, 0.74) versus 0.62 (0.59, 0.65).

CONCLUSIONS

The Richmond scoring system demonstrated more accurate predictions for the present outcomes. The simplicity and predictive value of the Richmond score make this system an ideal option for use in emergency settings and centers with high patient loads.

Prognostic Neuroimaging Biomarkers in Acute Vascular Brain Injury and Traumatic Brain Injury

Prognostic imaging biomarkers after acute brain injury inform treatment decisions, track the progression of intracranial injury, and can be used in shared decision-making processes with families. Herein, key established biomarkers and prognostic scoring systems are surveyed in the literature, and their applications in clinical practice and clinical trials are discussed. Biomarkers in acute ischemic stroke include computed tomography (CT) hypodensity scoring, diffusion-weighted lesion volume, and core infarct size on perfusion imaging. Intracerebral hemorrhage biomarkers include hemorrhage volume, expansion, and location. Aneurysmal subarachnoid biomarkers include hemorrhage grading, presence of diffusion-restricting lesions, and acute hydrocephalus. Traumatic brain injury CT scoring systems, contusion expansion, and diffuse axonal injury grading are reviewed. Emerging biomarkers including white matter disease scoring, diffusion tensor imaging, and the automated calculation of scoring systems and volumetrics are discussed.

Comparing Predictive Utility of Head Computed Tomography Scan-Based Scoring Systems for Traumatic Brain Injury: A Retrospective Study

This study compared the predictive utility of Marshall, Rotterdam, Stockholm, Helsinki, and NeuroImaging Radiological Interpretation System (NIRIS) scorings based on early non-contrast brain computed tomography (CT) scans in patients with traumatic brain injury (TBI). The area under a receiver operating characteristic curve (AUROC) was used to determine the predictive utility of scoring systems. Subgroup analyses were performed among patients with head AIS scores > 1. A total of 996 patients were included, of whom 786 (78.9%) were males. In-hospital mortality, ICU admission, neurosurgical intervention, and prolonged total hospital length of stay (THLOS) were recorded for 27 (2.7%), 207 (20.8%), 82 (8.2%), and 205 (20.6%) patients, respectively. For predicting in-hospital mortality, all scoring systems had AUROC point estimates above 0.9 and 0.75 among all included patients and patients with head AIS > 1, respectively, without any significant differences. The Marshall and NIRIS scoring systems had higher AUROCs for predicting ICU admission and neurosurgery than the other scoring systems. For predicting THLOS ≥ seven days, although the NIRIS and Marshall scoring systems seemed to have higher AUROC point estimates when all patients were analyzed, five scoring systems performed roughly the same in the head AIS > 1 subgroup.

Label-efficient deep semantic segmentation of intracranial hemorrhages in CT-scans

Intracranial hemorrhage (ICH) is a common finding in traumatic brain injury (TBI) and computed tomography (CT) is considered the gold standard for diagnosis. Automated detection of ICH provides clinical value in diagnostics and in the ability to feed robust quantification measures into future prediction models. Several studies have explored ICH detection and segmentation but the research process is somewhat hindered due to a lack of open large and labeled datasets, making validation and comparison almost impossible. The complexity of the task is further challenged by the heterogeneity of ICH patterns, requiring a large number of labeled data to train robust and reliable models. Consequently, due to the labeling cost, there is a need for label-efficient algorithms that can exploit easily available unlabeled or weakly-labeled data. Our aims for this study were to evaluate whether transfer learning can improve ICH segmentation performance and to compare a variety of transfer learning approaches that harness unlabeled and weakly-labeled data. Three self-supervised and three weakly-supervised transfer learning approaches were explored. To be used in our comparisons, we also manually labeled a dataset of 51 CT scans. We demonstrate that transfer learning improves ICH segmentation performance on both datasets. Unlike most studies on ICH segmentation our work relies exclusively on publicly available datasets, allowing for easy comparison of performances in future studies. To further promote comparison between studies, we also present a new public dataset of ICH-labeled CT scans, Seq-CQ500.

Prediction of In-Hospital Outcomes in Patients with Traumatic Brain Injury Using Computed Tomographic Scoring Systems: A Comparison Between Marshall, Rotterdam, and Neuroimaging Radiological Interpretation Systems

OBJECTIVE

This study aimed to compare the prognostic value of Marshall, Rotterdam, and Neuroimaging Radiological Interpretation Systems (NIRIS) in predicting the in-hospital outcomes of patients with traumatic brain injury.

METHODS

We identified 250 patients with traumatic brain injury in a retrospective single-center cohort from 2019 to 2020. Computed tomography (CT) scans were reviewed by two radiologists and scored according to three CT scoring systems. One-month outcomes were evaluated, including hospitalization, intensive care unit admission, neurosurgical procedure, and mortality. Logistic regression analysis was performed to identify scoring systems and outcome relationships. The best cutoff value was calculated using the receiver operating characteristic curve model.

RESULTS

Eighteen patients (7.2%) died in the 1-month follow-up. The mean age and Glasgow Coma Scale of survivors differed significantly from nonsurvivors. Subarachnoid hemorrhage and compressed/absent cisterns were dead patients' most frequent CT findings. All three scoring systems had good discrimination power in mortality prediction (area under the receiver operating characteristic curve of the Marshall, Rotterdam, and NIRIS was 0.78, 0.86, and 0.84, respectively). Regarding outcome, three systems directly correlated with unfavorable outcome prediction.

CONCLUSIONS

The Marshall, Rotterdam, and NIRIS are good predictive models for mortality and outcome prediction, with slight superiority of the Rotterdam in mortality prediction and the Marshall in intensive care unit admission and neurosurgical procedures.

Trajectories of interleukin 10 and heart fatty acid-binding protein levels in traumatic brain injury patients with or without extracranial injuries

Background

Interleukin 10 (IL-10) and heart fatty acid-binding protein (H-FABP) have gained interest as diagnostic biomarkers of traumatic brain injury (TBI), but factors affecting their blood levels in patients with moderate-to-severe TBI are largely unknown.

Objective

To investigate the trajectories of IL-10 and H-FABP between TBI patients with and without extracranial injuries (ECI); to investigate if there is a correlation between the levels of IL-10 and H-FABP with the levels of inflammation/infection markers C-reactive protein (CRP) and leukocytes; and to investigate if there is a correlation between the admission level of H-FABP with admission levels of cardiac injury markers, troponin (TnT), creatine kinase (CK), and creatine kinase MB isoenzyme mass (CK-MBm).

Materials and methods

The admission levels of IL-10, H-FABP, CRP, and leukocytes were measured within 24 h post-TBI and on days 1, 2, 3, and 7 after TBI. The admission levels of TnT, CK, and CK-MBm were measured within 24 h post-TBI.

Results

There was a significant difference in the concentration of H-FABP between TBI patients with and without ECI on day 0 (48.2 ± 20.5 and 12.4 ± 14.7 ng/ml, p = 0.02, respectively). There was no significant difference in the levels of IL-10 between these groups at any timepoints. There was a statistically significant positive correlation between IL-10 and CRP on days 2 (R = 0.43, p &lt; 0.01) and 7 (R = 0.46, p = 0.03) after injury, and a negative correlation between H-FABP and CRP on day 0 (R = -0.45, p = 0.01). The levels of IL-10 or H-FABP did not correlate with leukocyte counts at any timepoint. The admission levels of H-FABP correlated with CK (R = 0.70, p &lt; 0.001) and CK-MBm (R = 0.61, p &lt; 0.001), but not with TnT.

Conclusion

Inflammatory reactions during the early days after a TBI do not significantly confound the use of IL-10 and H-FABP as TBI biomarkers. Extracranial injuries and cardiac sources may influence the levels of H-FABP in patients with moderate-to-severe TBI.

Prediction of Mortality among Patients with Isolated Traumatic Brain Injury Using Machine Learning Models in Asian Countries: An International Multi-Center Cohort Study

Traumatic brain injury (TBI) is a significant healthcare concern in several countries, accounting for a major burden of morbidity, mortality, disability, and socioeconomic losses. Although conventional prognostic models for patients with TBI have been validated, their performance has been limited. Therefore, we aimed to construct machine learning (ML) models to predict the clinical outcomes in adult patients with isolated TBI in Asian countries. The Pan-Asian Trauma Outcome Study registry was used in this study, and the data were prospectively collected from January 1, 2015, to December 31, 2020. Among a total of 6540 patients (≥ 15 years) with isolated moderate and severe TBI, 3276 (50.1%) patients were randomly included with stratification by outcomes and subgrouping variables for model evaluation, and 3264 (49.9%) patients were included for model training and validation. Logistic regression was considered as a baseline, and ML models were constructed and evaluated using the area under the precision-recall curve (AUPRC) as the primary outcome metric, area under the receiver operating characteristic curve (AUROC), and precision at fixed levels of recall. The contribution of the variables to the model prediction was measured using the SHapley Additive exPlanations (SHAP) method. The ML models outperformed logistic regression in predicting the in-hospital mortality. Among the tested models, the gradient-boosted decision tree showed the best performance (AUPRC, 0.746 [0.700-0.789]; AUROC, 0.940 [0.929-0.952]). The most powerful contributors to model prediction were the Glasgow Coma Scale, O₂ saturation, transfusion, systolic and diastolic blood pressure, body temperature, and age. Our study suggests that ML techniques might perform better than conventional multi-variate models in predicting the outcomes among adult patients with isolated moderate and severe TBI.

Multi-Mechanistic Approaches to the Treatment of Traumatic Brain Injury: A Review

Traumatic brain injury (TBI) is a leading cause of death and disability worldwide. Despite extensive research efforts, the majority of trialed monotherapies to date have failed to demonstrate significant benefit. It has been suggested that this is due to the complex pathophysiology of TBI, which may possibly be addressed by a combination of therapeutic interventions. In this article, we have reviewed combinations of different pharmacologic treatments, combinations of non-pharmacologic interventions, and combined pharmacologic and non-pharmacologic interventions for TBI. Both preclinical and clinical studies have been included. While promising results have been found in animal models, clinical trials of combination therapies have not yet shown clear benefit. This may possibly be due to their application without consideration of the evolving pathophysiology of TBI. Improvements of this paradigm may come from novel interventions guided by multimodal neuromonitoring and multimodal imaging techniques, as well as the application of multi-targeted non-pharmacologic and endogenous therapies. There also needs to be a greater representation of female subjects in preclinical and clinical studies.

Novel CT-based parameters assessing relative cross-sectional area to guide surgical management and predict clinical outcomes in patients with acute subdural hematoma

INTRODUCTION

Acute subdural hematoma (aSDH) is one of the most devastating entities secondary to traumatic brain injury (TBI). Even though radiological computed tomography (CT) findings, such as hematoma thickness (HT), midline shift (MLS), and MLS/HT ratio, have an important prognostic role, they suffer from important drawbacks. We hypothesized that relative cross-sectional area (rCSA) of specific brain regions would provide valuable information about brain compression and swelling, thus being a key determining factor governing the clinical course.

METHODS

We performed an 8-year retrospective analysis of patients with moderate to severe TBI with surgically evacuated, isolated, unilateral aSDH. We investigated the influence of aSDH rCSA and ipsilateral hemisphere rCSA along the supratentorial region on the subsequent operative technique employed for aSDH evacuation and patient's clinical outcomes (early death and Glasgow Outcome Scale [GOS] at discharge and after 1-year follow-up). Different conventional radiological variables were also assessed.

RESULTS

The study included 39 patients. Lower HT, MLS, hematoma volume, and aSDH rCSA showed a significant association with decompressive craniectomy (DC) procedure. Conversely, higher ipsilateral hemisphere rCSA along the dorso-ventral axis and, specifically, ipsilateral hemisphere rCSA at the high convexity level were predictors for DC. CT segmentation analysis exhibited a modest relationship with early death, which was limited to the basal supratentorial subregion, but could not predict long-term outcome.

CONCLUSION

rCSA is an objectifiable and reliable radiologic parameter available on admission CT that might provide valuable information to optimize surgical treatment.

Accuracy of head computed tomography scoring systems in predicting outcomes for patients with moderate to severe traumatic brain injury: A ProTECT III ancillary study

BACKGROUND

Several types of head CT classification systems have been developed to prognosticate and stratify TBI patients.

OBJECTIVE

The purpose of our study was to compare the predictive value and accuracy of the different CT scoring systems, including the Marshall, Rotterdam, Stockholm, Helsinki, and NIRIS systems, to inform specific patient management actions, using the ProTECT III population of patients with moderate to severe acute traumatic brain injury (TBI).

METHODS

We used the data collected in the patients with moderate to severe (GCS score of 4-12) TBI enrolled in the ProTECT III clinical trial. ProTECT III was a NIH-funded, prospective, multicenter, randomized, double-blind, placebo-controlled clinical trial designed to determine the efficacy of early administration of IV progesterone. The CT scoring systems listed above were applied to the baseline CT scans obtained in the trial. We assessed the predictive accuracy of these scoring systems with respect to Glasgow Outcome Scale-Extended at 6 months, disability rating scale score, and mortality.

RESULTS

A total of 882 subjects were enrolled in ProTECT III. Worse scores for each head CT scoring systems were highly correlated with unfavorable outcome, disability outcome, and mortality. The NIRIS classification was more strongly correlated than the Stockholm and Rotterdam CT scores, followed by the Helsinki and Marshall CT classification. The highest correlation was observed between NIRIS and mortality (estimated odds ratios of 4.83).

CONCLUSION

All scores were highly associated with 6-month unfavorable, disability and mortality outcomes. NIRIS was also accurate in predicting TBI patients' management and disposition.

Evaluation of the prognosis of acute subdural hematoma according to the density differences between gray and white matter

Objective

Acute subdural hematoma (ASDH) is a common neurological emergency, and its appearance on head-computed tomographic (CT) imaging helps guide clinical treatment. To provide a basis for clinical decision-making, we analyzed that the density difference between the gray and white matter of the CT image is associated with the prognosis of patients with ASDH.

Methods

We analyzed the data of 194 patients who had ASDH as a result of closed traumatic brain injury (TBI) between 2018 and 2021. The patients were subdivided into surgical and non-surgical groups, and the non-surgical group was further subdivided into "diffused [hematoma]" and "non-diffused" groups. The control group's CT scans were normal. The 3D Slicer software was used to quantitatively analyze the density of gray and white matter depicted in the CT images.

Results

Imaging evaluation showed that the median difference in density between the gray and white matter on the injured side was 4.12 HU (IQR, 3.91-4.22 HU; p < 0.001) and on the non-injured side was 4.07 HU (IQR, 3.90-4.19 HU; p < 0.001), and the hematoma needs to be surgically removed. The median density difference value of the gray and white matter on the injured side was 3.74 HU (IQR, 3.53-4.01 HU; p < 0.001) and on the non-injured side was 3.71 HU (IQR, 3.69-3.73 HU; p < 0.001), and the hematoma could diffuse in a short time.

Conclusion

Quantitative analysis of the density differences in the gray and white matter of the CT images can be used to evaluate the clinical prognosis of patients with ASDH.

Automated identification and quantification of traumatic brain injury from CT scans: Are we there yet?

BACKGROUND

The purpose of this study was to conduct a systematic review for understanding the availability and limitations of artificial intelligence (AI) approaches that could automatically identify and quantify computed tomography (CT) findings in traumatic brain injury (TBI).

METHODS

Systematic review, in accordance with PRISMA 2020 and SPIRIT-AI extension guidelines, with a search of 4 databases (Medline, Embase, IEEE Xplore, and Web of Science) was performed to find AI studies that automated the clinical tasks for identifying and quantifying CT findings of TBI-related abnormalities.

RESULTS

A total of 531 unique publications were reviewed, which resulted in 66 articles that met our inclusion criteria. The following components for identification and quantification regarding TBI were covered and automated by existing AI studies: identification of TBI-related abnormalities; classification of intracranial hemorrhage types; slice-, pixel-, and voxel-level localization of hemorrhage; measurement of midline shift; and measurement of hematoma volume. Automated identification of obliterated basal cisterns was not investigated in the existing AI studies. Most of the AI algorithms were based on deep neural networks that were trained on 2- or 3-dimensional CT imaging datasets.

CONCLUSION

We identified several important TBI-related CT findings that can be automatically identified and quantified with AI. A combination of these techniques may provide useful tools to enhance reproducibility of TBI identification and quantification by supporting radiologists and clinicians in their TBI assessments and reducing subjective human factors.

Performance of the IMPACT and Helsinki models for predicting 6-month outcomes in a cohort of patients with traumatic brain injury undergoing cranial surgery

Background and aim

Prediction models for patients with traumatic brain injury (TBI) require generalizability and should apply to different settings. We aimed to validate the IMPACT and Helsinki prognostic models in patients with TBI who underwent cranial surgery in a Chinese center.

Methods

This validation study included 607 surgical patients with moderate to severe TBI (Glasgow Coma Scale [GCS] score ≤12) who were consecutively admitted to the Neurotrauma Center of People's Liberation Army (PLANC), China, between 2009 and 2021. The IMPACT models (core, extended and lab) and the Helsinki CT clinical model were used to estimate 6-month mortality and unfavorable outcomes. To assess performance, we studied discrimination and calibration.

Results

In the PLANC database, the observed 6-month mortality rate was 28%, and the 6-month unfavorable outcome was 52%. Significant differences in case mix existed between the PLANC cohort and the development populations for the IMPACT and, to a lesser extent, for the Helsinki models. Discrimination of the IMPACT and Helsinki models was excellent, with most AUC values ≥0.80. The highest values were found for the IMPACT lab model (AUC 0.87) and the Helsinki CT clinical model (AUC 0.86) for the prediction of unfavorable outcomes. Overestimation was found for all models, but the degree of miscalibration was lower in the Helsinki CT clinical model.

Conclusion

In our population of surgical TBI patients, the IMPACT and Helsinki CT clinical models demonstrated good performance, with excellent discrimination but suboptimal calibration. The good discrimination confirms the validity of the predictors, but the poorer calibration suggests a need to recalibrate the models to specific settings.

Impact of Intracranial Hypertension on Outcome of Severe Traumatic Brain Injury Pediatric Patients: A 15-Year Single Center Experience

Background: Intracranial hypertension (IC-HTN) is significantly associated with higher risk for an unfavorable outcome in pediatric trauma. Intracranial pressure (ICP) monitoring is widely becoming a standard of neurocritical care for children. Methods: The present study was designed to evaluate influences of IC-HTN on clinical outcomes of pediatric TBI patients. Demographic, injury severity, radiologic characteristics were used as possible predictors of IC-HTN or of functional outcome. Results: A total of 118 pediatric intensive care unit (PICU) patients with severe TBI (sTBI) were included. Among sTBI cases, patients with GCS < 5 had significantly higher risk for IC-HTN and for mortality. Moreover, there was a statistically significant positive correlation between IC-HTN and severity scoring systems. Kaplan−Meier analysis determined a significant difference for good recovery among patients who had no ICP elevations, compared to those who had at least one episode of IC-HTN (log-rank chi-square = 11.16, p = 0.001). A multivariable predictive logistic regression analysis distinguished the ICP-monitored patients at risk for developing IC-HTN. The model finally revealed that higher ISS and Helsinki CT score increased the odds for developing IC-HTN (p < 0.05). Conclusion: The present study highlights the importance of ICP-guided clinical practices, which may lead to increasing percentages of good recovery for children.

Clustering identifies endotypes of traumatic brain injury in an intensive care cohort: a CENTER-TBI study

BACKGROUND

While the Glasgow coma scale (GCS) is one of the strongest outcome predictors, the current classification of traumatic brain injury (TBI) as 'mild', 'moderate' or 'severe' based on this fails to capture enormous heterogeneity in pathophysiology and treatment response. We hypothesized that data-driven characterization of TBI could identify distinct endotypes and give mechanistic insights.

METHODS

We developed an unsupervised statistical clustering model based on a mixture of probabilistic graphs for presentation (< 24 h) demographic, clinical, physiological, laboratory and imaging data to identify subgroups of TBI patients admitted to the intensive care unit in the CENTER-TBI dataset (N = 1,728). A cluster similarity index was used for robust determination of optimal cluster number. Mutual information was used to quantify feature importance and for cluster interpretation.

RESULTS

Six stable endotypes were identified with distinct GCS and composite systemic metabolic stress profiles, distinguished by GCS, blood lactate, oxygen saturation, serum creatinine, glucose, base excess, pH, arterial partial pressure of carbon dioxide, and body temperature. Notably, a cluster with 'moderate' TBI (by traditional classification) and deranged metabolic profile, had a worse outcome than a cluster with 'severe' GCS and a normal metabolic profile. Addition of cluster labels significantly improved the prognostic precision of the IMPACT (International Mission for Prognosis and Analysis of Clinical trials in TBI) extended model, for prediction of both unfavourable outcome and mortality (both p < 0.001).

CONCLUSIONS

Six stable and clinically distinct TBI endotypes were identified by probabilistic unsupervised clustering. In addition to presenting neurology, a profile of biochemical derangement was found to be an important distinguishing feature that was both biologically plausible and associated with outcome. Our work motivates refining current TBI classifications with factors describing metabolic stress. Such data-driven clusters suggest TBI endotypes that merit investigation to identify bespoke treatment strategies to improve care. Trial registration The core study was registered with ClinicalTrials.gov, number NCT02210221 , registered on August 06, 2014, with Resource Identification Portal (RRID: SCR_015582).

A proposed novel traumatic brain injury classification system - an overview and inter-rater reliability validation on behalf of the Society of British Neurological Surgeons

INTRODUCTION

The measurement of traumatic brain injury (TBI) 'severity' has traditionally been based on the earliest Glasgow Coma Score (GCS) recorded, however, the underlying parenchymal pathology is highly heterogonous. This heterogeneity renders prediction of outcome on an individual patient level inaccurate and makes comparison between patients both in clinical practice and research difficult. The complexity of this heterogeneity has resulted in generic all encompassing 'traumatic brain injury protocols'. Early management and studies of neuro-protectants are often done irrespective of TBI type, yet it may well be that a specific treatment may be beneficial in a subset of TBI pathologies.

METHODS

A simple CT-based classification system rating the recognised types of blunt TBI (extradural, subdural, subarachnoid haemorrhage, contusions/intracerebral haematoma and diffuse axonal injury) as mild (1), moderate (2) or severe (3) is proposed. Hypoxic brain injury, a common secondary injury following TBI, is also included. Scores can be combined to reflect concomitant types of TBI and predominant location of injury is also recorded. To assess interrater reliability, 50 patient CT images were assessed by 5 independent clinicians of varying experience. Interrater reliability was calculated using overall agreement through Cronbach's alpha including confidence intervals for intra-class coefficients.

RESULTS

Interrater reliability scores showed strong agreement for same score and same injury for TBIs with blood on CT and Cronbach's alpha co-efficient (range 0.87-0.93) demonstrated excellent correlation between raters. Cronbach's alpha was not affected when individual raters were removed.

CONCLUSIONS

The proposed simple CT classification system has good inter-rater reliability and hence potentially could enable better individual prognostication and targeted treatments to be compared while also accounting for multiple intracranial injury types. Further studies are proposed and underway.

Mild-to-Moderate Traumatic Brain Injury: A Review with Focus on the Visual System

Traumatic Brain Injury (TBI) is a major global public health problem. Neurological damage from TBI may be mild, moderate, or severe and occurs both immediately at the time of impact (primary injury) and continues to evolve afterwards (secondary injury). In mild (m)TBI, common symptoms are headaches, dizziness and fatigue. Visual impairment is especially prevalent. Insomnia, attentional deficits and memory problems often occur. Neuroimaging methods for the management of TBI include computed tomography and magnetic resonance imaging. The location and the extent of injuries determine the motor and/or sensory deficits that result. Parietal lobe damage can lead to deficits in sensorimotor function, memory, and attention span. The processing of visual information may be disrupted, with consequences such as poor hand-eye coordination and balance. TBI may cause lesions in the occipital or parietal lobe that leave the TBI patient with incomplete homonymous hemianopia. Overall, TBI can interfere with everyday life by compromising the ability to work, sleep, drive, read, communicate and perform numerous activities previously taken for granted. Treatment and rehabilitation options available to TBI sufferers are inadequate and there is a pressing need for new ways to help these patients to optimize their functioning and maintain productivity and participation in life activities, family and community.

Prognostic Value of Different Computed Tomography Scoring Systems in Patients With Severe Traumatic Brain Injury Undergoing Decompressive Craniectomy

OBJECTIVE

In this study, we investigate the preoperative and postoperative computed tomography (CT) scores in severe traumatic brain injury (TBI) patients undergoing decompressive craniectomy (DC) and compare their predictive accuracy.

METHODS

Univariate and multivariate logistic regression analyses were used to determine the relationship between CT score (preoperative and postoperative) and mortality at 30 days after injury. The discriminatory power of preoperative and postoperative CT score was assessed by the area under the receiver operating characteristic curve (AUC).

RESULTS

Multivariate logistic regression analysis adjusted for the established predictors of TBI outcomes showed that preoperative Rotterdam CT score (odds ratio [OR], 3.60; 95% confidence interval [CI], 1.13-11.50; P = 0.030), postoperative Rotterdam CT score (OR, 4.17; 95% CI, 1.63-10.66; P = 0.003), preoperative Stockholm CT score (OR, 3.41; 95% CI, 1.42-8.18; P = 0.006), postoperative Stockholm CT score (OR, 4.50; 95% CI, 1.60-12.64; P = 0.004), preoperative Helsinki CT score (OR, 1.44; 95% CI, 1.03-2.02; P = 0.031), and postoperative Helsinki CT score (OR, 2.55; 95% CI, 1.32-4.95; P = 0.005) were significantly associated with mortality. The performance of the postoperative Rotterdam CT score was superior to the preoperative Rotterdam CT score (AUC, 0.82-0.97 vs 0.71-0.91). The postoperative Stockholm CT score was superior to the preoperative Stockholm CT score (AUC, 0.76-0.94 vs 0.72-0.92). The postoperative Helsinki CT score was superior to the preoperative Helsinki CT score (AUC, 0.88-0.99 vs 0.65-0.87).

CONCLUSIONS

In conclusion, assessing the CT score before and after DC may be more precise and efficient for predicting early mortality in severe TBI patients who undergo DC.

Prediction of Early Mortality Among Children With Moderate or Severe Traumatic Brain Injury Based on a Nomogram Integrating Radiological and Inflammation-Based Biomarkers

Inflammation-based scores have been increasingly used for prognosis prediction in neurological diseases. This study aimed to investigate the predictive value of inflammation-based scores combined with radiological characteristics in children with moderate or severe traumatic brain injury (MS-TBI). A total of 104 pediatric patients with MS-TBI were retrospectively enrolled and randomly divided into training and validation cohorts at a 7:3 ratio. Univariate and multivariate logistic regression analyses were performed to identify independent predictors of prognosis in pediatric patients with MS-TBI. A prognostic nomogram was constructed, and its predictive performance was validated in both the training and validation cohorts. Sex, admission platelet-to-lymphocyte ratio, and basal cistern status from initial CT findings were identified as independent prognostic predictors for children with MS-TBI in multivariate logistic analysis. Based on these findings, a nomogram was then developed and its concordance index values were 0.918 [95% confidence interval (CI): 0.837-0.999] in the training cohort and 0.86 (95% CI: 0.70-1.00) in the validation cohort, which significantly outperformed those of the Rotterdam, Marshall, and Helsinki CT scores. The proposed nomogram, based on routine complete blood count and initial CT scan findings, can contribute to individualized prognosis prediction and clinical decision-making in children with MS-TBI.

Rubiano A, G. Kolias A, Silva Paiva W. Evaluation of Computed Tomography Scoring Systems in the Prediction of Short-Term Mortality in Traumatic Brain Injury Patients from a Low- to Middle-Income Country

The present study aims to evaluate the accuracy of the prognostic discrimination and prediction of the short-term mortality of the Marshall computed tomography (CT) classification and Rotterdam and Helsinki CT scores in a cohort of TBI patients from a low- to middle-income country. This is a post hoc analysis of a previously conducted prospective cohort study conducted in a university-associated, tertiary-level hospital that serves a population of >12 million in Brazil. Marshall CT class, Rotterdam and Helsinki scores, and their components were evaluated in the prediction of 14-day and in-hospital mortality using Nagelkerk's pseudo-R² and area under the receiver operating characteristic curve. Multi-variate regression was performed using known outcome predictors (age, Glasgow Coma Scale, pupil response, hypoxia, hypotension, and hemoglobin values) to evaluate the increase in variance explained when adding each of the CT classification systems. Four hundred forty-seven patients were included. Mean age of the patient cohort was 40 (standard deviation, 17.83) years, and 85.5% were male. Marshall CT class was the least accurate model, showing pseudo-R² values equal to 0.122 for 14-day mortality and 0.057 for in-hospital mortality, whereas Rotterdam CT scores were 0.245 and 0.194 and Helsinki CT scores were 0.264 and 0.229. The AUC confirms the best prediction of the Rotterdam and Helsinki CT scores regarding the Marshall CT class, which presented greater discriminative ability. When associated with known outcome predictors, Marshall CT class and Rotterdam and Helsinki CT scores showed an increase in the explained variance of 2%, 13.4%, and 21.6%, respectively. In this study, Rotterdam and Helsinki scores were more accurate models in predicting short-term mortality. The study denotes a contribution to the process of external validation of the scores and may collaborate with the best risk stratification for patients with this important pathology.

Chinese Admission Warning Strategy for Predicting the Hospital Discharge Outcome in Patients with Traumatic Brain Injury

Objective: To develop and validate an admission warning strategy that incorporates the general emergency department indicators for predicting the hospital discharge outcome of patients with traumatic brain injury (TBI) in China. Methods: This admission warning strategy was developed in a primary cohort that consisted of 605 patients with TBI who were admitted within 6 h of injury. The least absolute shrinkage and selection operator and multivariable logistic regression analysis were used to develop the early warning strategy of selected indicators. Two sub-cohorts consisting of 180 and 107 patients with TBI were used for the external validation. Results: Indicators of the strategy included three categories: baseline characteristics, imaging and laboratory indicators. This strategy displayed good calibration and good discrimination. A high C-index was reached in the internal validation. The multicenter external validation cohort still showed good discrimination C-indices. Decision curve analysis (DCA) showed the actual needs of this strategy when the possibility threshold was 0.01 for the primary cohort, and at thresholds of 0.02–0.83 and 0.01–0.88 for the two sub-cohorts, respectively. In addition, this strategy exhibited a significant prognostic capacity compared to the traditional single predictors, and this optimization was also observed in two external validation cohorts. Conclusions: We developed and validated an admission warning strategy that can be quickly deployed in the emergency department. This strategy can be used as an ideal tool for predicting hospital discharge outcomes and providing objective evidence for early informed consent of the hospital discharge outcome to the family members of TBI patients.

Ideal algorithms in healthcare: Explainable, dynamic, precise, autonomous, fair, and reproducible

Established guidelines describe minimum requirements for reporting algorithms in healthcare; it is equally important to objectify the characteristics of ideal algorithms that confer maximum potential benefits to patients, clinicians, and investigators. We propose a framework for ideal algorithms, including 6 desiderata: explainable (convey the relative importance of features in determining outputs), dynamic (capture temporal changes in physiologic signals and clinical events), precise (use high-resolution, multimodal data and aptly complex architecture), autonomous (learn with minimal supervision and execute without human input), fair (evaluate and mitigate implicit bias and social inequity), and reproducible (validated externally and prospectively and shared with academic communities). We present an ideal algorithms checklist and apply it to highly cited algorithms. Strategies and tools such as the predictive, descriptive, relevant (PDR) framework, the Standard Protocol Items: Recommendations for Interventional Trials-Artificial Intelligence (SPIRIT-AI) extension, sparse regression methods, and minimizing concept drift can help healthcare algorithms achieve these objectives, toward ideal algorithms in healthcare.

External validation of the NeuroImaging Radiological Interpretation System and Helsinki computed tomography score for mortality prediction in patients with traumatic brain injury treated in the intensive care unit: a Finnish intensive care consortium study

BACKGROUND

Admission computed tomography (CT) scoring systems can be used to objectively quantify the severity of traumatic brain injury (TBI) and aid in outcome prediction. We aimed to externally validate the NeuroImaging Radiological Interpretation System (NIRIS) and the Helsinki CT score. In addition, we compared the prognostic performance of the NIRIS and the Helsinki CT score to the Marshall CT classification and to a clinical model.

METHODS

We conducted a retrospective multicenter observational study using the Finnish Intensive Care Consortium database. We included adult TBI patients admitted in four university hospital ICUs during 2003-2013. We analyzed the CT scans using the NIRIS and the Helsinki CT score and compared the results to 6-month mortality as the primary outcome. In addition, we created a clinical model (age, Glasgow Coma Scale score, Simplified Acute Physiology Score II, presence of severe comorbidity) and combined clinical and CT models to see the added predictive impact of radiological data to conventional clinical information. We measured model performance using area under curve (AUC), Nagelkerke's R² statistics, and the integrated discrimination improvement (IDI).

RESULTS

A total of 3031 patients were included in the analysis. The 6-month mortality was 710 patients (23.4%). Of the CT models, the Helsinki CT displayed best discrimination (AUC 0.73 vs. 0.70 for NIRIS) and explanatory variation (Nagelkerke's R² 0.20 vs. 0.15). The clinical model displayed an AUC of 0.86 (95% CI 0.84-0.87). All CT models increased the AUC of the clinical model by + 0.01 to 0.87 (95% CI 0.85-0.88) and the IDI by 0.01-0.03.

CONCLUSION

In patients with TBI treated in the ICU, the Helsinki CT score outperformed the NIRIS for 6-month mortality prediction. In isolation, CT models offered only moderate accuracy for outcome prediction and clinical variables outweighing the CT-based predictors in terms of predictive performance.

Extended Analysis of Axonal Injuries Detected Using Magnetic Resonance Imaging in Critically Ill Traumatic Brain Injury Patients

Studies show conflicting results regarding the prognostic significance of traumatic axonal injuries (TAI) in patients with traumatic brain injury (TBI). Therefore, we documented the presence of TAI in several brain regions, using different magnetic resonance imaging (MRI) sequences, and assessed their association to patient outcomes using machine learning. Further, we created a novel MRI-based TAI grading system with the goal of improving outcome prediction in TBI. We subsequently evaluated the performance of several TAI grading systems. We used a genetic algorithm to identify TAI that distinguish favorable from unfavorable outcomes. We assessed the discriminatory performance (area under the curve [AUC]) and goodness-of-fit (Nagelkerke pseudo-R2) of the novel Stockholm MRI grading system and the TAI grading systems of Adams and associates, Firsching and coworkers. and Abu Hamdeh and colleagues, using both univariate and multi-variate logistic regression. The dichotomized Glasgow Outcome Scale was considered the primary outcome. We examined the MRI scans of 351 critically ill patients with TBI. The TAI in several brain regions, such as the midbrain tegmentum, were strongly associated with unfavorable outcomes. The Stockholm MRI grading system exhibited the highest AUC (0.72 vs. 0.68-0.69) and Nagelkerke pseudo-R2 (0.21 vs. 0.14-0.15) values of all TAI grading systems. These differences in model performance, however, were not statistically significant (DeLong test, p > 0.05). Further, all included TAI grading systems improved outcome prediction relative to established outcome predictors of TBI, such as the Glasgow Coma Scale (likelihood-ratio test, p < 0.001). Our findings suggest that the detection of TAI using MRI is a valuable addition to prognostication in TBI.