Prognosis of Six-Month Glasgow Outcome Scale in Severe Traumatic Brain Injury Using Hospital Admission Characteristics, Injury Severity Characteristics, and Physiological Monitoring during the First Day Post-Injury

Investigating the Effect of High-Dose Vitamin D3 Administration on Inflammatory Biomarkers in Patients with Moderate to Severe Traumatic Brain Injury: A Randomized Clinical Trial

Background

Traumatic brain injury (TBI) is one of the most common neurological disorders worldwide. We aimed to investigate the efficacy of high-dose vitamin D3 on inflammatory biomarkers in patients with moderate to severe TBI.

Methods

Thirty-five moderate to severe TBI patients were randomly assigned to intervention and control groups. Patients in the intervention group received a single intramuscular (IM) dose of 300,000 IU vitamin D. The primary endpoints were interleukin levels (IL-1β and IL-6), and the secondary endpoints were changes in neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR), Glasgow Coma scale (GCS), and Glasgow Outcome Scale-Extended (GOS-E) scores compared between intervention and control arms of the study. The linear Generalized Estimating Equations were used for trend analysis and evaluating the association of independent factors to each outcome.

Results

The results revealed a significant decrease in IL-1β levels (-2.71±3.02, in the intervention group: P=0.001 vs. -0.14±3.70, in the control group: P=0.876) and IL-6 (-88.05±148.45, in the intervention group: P=0.0001 vs. -35.54±175.79, in the control groupL P=0.325) 3 days after the intervention. The improvement in the GCS score (P=0.001), reduction in NLR (P=0.001) and PLR (P=0.002), and improvement in the GOS-E score (P=0.039) was found to be greater in the vitamin D3 arm of the study than the control group.

Conclusion

Administration of high-dose vitamin D3 in the acute phase of TBI could be effective in lowering the inflammatory markers and improving the level of consciousness and long-term performance outcomes.Trial Registration Number: IRCT20180522039777N2.

Prediction of Neurological Outcomes in Elderly Patients With Head Trauma Using the Geriatric Trauma Outcome Score: A Retrospective Observational Study

Introduction Head trauma in elderly people is a problem in today's aging society. Elderly people are susceptible to head trauma because of their declining physical function; this tends to be severe. Outcome prediction is important in decision-making regarding treatment strategies; however, there is no unified method for predicting neurological outcomes in elderly patients with head trauma. Methods Elderly patients with head trauma admitted to the Japan Red Cross Narita Hospital between January 2019 and August 2023 were enrolled in this single-center, retrospective observational study. A favorable neurological outcome was defined as a cerebral performance category scale of 1 or 2. Multivariate logistic regression analysis and receiver operating characteristic curve analysis were performed to investigate the association between geriatric trauma outcome scores and outcomes and to evaluate the predictive value of geriatric trauma outcome scores. The primary outcome was a favorable neurological outcome at discharge, and the secondary outcome was in-hospital mortality. Results A total of 313 elderly patients with head trauma were eligible for analysis. Multivariate logistic regression analysis revealed that the geriatric trauma outcome score was significantly associated with a favorable neurological outcome at discharge (odds ratio 0.94, P <0.0001). In the receiver operating characteristic curve analysis, the geriatric trauma outcome score had a good predictive value for favorable neurological outcomes at discharge (area under the receiver operating characteristic curve 0.83). Conclusions The geriatric trauma outcome score had good predictive value for favorable neurological outcomes at discharge in elderly patients with head trauma and has the potential to aid in decision-making regarding treatment strategies for elderly patients with head trauma.

Utility of Early Magnetic Resonance Imaging to Enhance Outcome Prediction in Critically Ill Children with Severe Traumatic Brain Injury

BACKGROUND

Many children with severe traumatic brain injury (TBI) receive magnetic resonance imaging (MRI) during hospitalization. There are insufficient data on how different patterns of injury on early MRI inform outcomes.

METHODS

Children (3-17 years) admitted in 2010-2021 for severe TBI (Glasgow Coma Scale [GCS] score < 9) were identified using our site's trauma registry. We used multivariable modeling to determine whether the hemorrhagic diffuse axonal injury (DAI) grade and the number of regions with restricted diffusion (subcortical white matter, corpus callosum, deep gray matter, and brainstem) on MRI obtained within 7 days of injury were independently associated with time to follow commands and with Functional Independence Measure for Children (WeeFIM) scores at the time of discharge from inpatient rehabilitation. We controlled for the clinical variables age, preadmission cardiopulmonary resuscitation, pupil reactivity, motor GCS score, and fever (> 38 °C) in the first 12 h.

RESULTS

Of 260 patients, 136 (52%) underwent MRI within 7 days of injury at a median of 3 days (interquartile range [IQR] 2-4). Patients with early MRI were a median age of 11 years (IQR 7-14), 8 (6%) patients received cardiopulmonary resuscitation, 19 (14%) patients had bilateral unreactive pupils, the median motor GCS score was 1 (IQR 1-4), and 82 (60%) patients had fever. Grade 3 DAI was present in 46 (34%) patients, and restricted diffusion was noted in the corpus callosum in 75 (55%) patients, deep gray matter in 29 (21%) patients, subcortical white matter in 23 (17%) patients, and the brainstem in 20 (15%) patients. After controlling for clinical variables, an increased number of regions with restricted diffusion, but not hemorrhagic DAI grade, was independently associated with longer time to follow commands (hazard ratio 0.68, 95% confidence interval 0.53-0.89) and worse WeeFIM scores (estimate β - 4.67, 95% confidence interval - 8.33 to - 1.01).

CONCLUSIONS

Regional restricted diffusion on early MRI is independently associated with short-term outcomes in children with severe TBI. Multicenter cohort studies are needed to validate these findings and elucidate the association of early MRI features with long-term outcomes in children with severe TBI.

Machine learning-based prediction of clinical outcomes after traumatic brain injury: Hidden information of early physiological time series

AIMS

To assess the predictive value of early-stage physiological time-series (PTS) data and non-interrogative electronic health record (EHR) signals, collected within 24 h of ICU admission, for traumatic brain injury (TBI) patient outcomes.

METHODS

Using data from TBI patients in the multi-center eICU database, we focused on in-hospital mortality, neurological status based on the Glasgow Coma Score (mGCS) motor subscore at discharge, and prolonged ICU stay (PLOS). Three machine learning (ML) models were developed, utilizing EHR features, PTS signals collected 24 h after ICU admission, and their combination. External validation was performed using the MIMIC III dataset, and interpretability was enhanced using the Shapley Additive Explanations (SHAP) algorithm.

RESULTS

The analysis included 1085 TBI patients. Compared to individual models and existing scoring systems, the combination of EHR and PTS features demonstrated comparable or even superior performance in predicting in-hospital mortality (AUROC = 0.878), neurological outcomes (AUROC = 0.877), and PLOS (AUROC = 0.835). The model's performance was validated in the MIMIC III dataset, and SHAP algorithms identified six key intervention points for EHR features related to prognostic outcomes. Moreover, the EHR results (All AUROC >0.8) were translated into online tools for clinical use.

CONCLUSION

Our study highlights the importance of early-stage PTS signals in predicting TBI patient outcomes. The integration of interpretable algorithms and simplified prediction tools can support treatment decision-making, contributing to the development of accurate prediction models and timely clinical intervention.

A Meta-analysis of Predicting Disorders of Consciousness After Traumatic Brain Injury by Machine Learning Models

Objective

This study pursued a meta-analysis to evaluate the predictive accuracy of machine learning (ML) models in determining disorders of consciousness (DOC) among patients with traumatic brain injury (TBI).

Methods

A comprehensive literature search was conducted to identify ML applications in the establishment of a predictive model of DOC after TBI as of August 6, 2023. Two independent reviewers assessed publication eligibility based on predefined criteria. The predictive accuracy was measured using areas under the receiver operating characteristic curves (AUCs). Subsequently, a random-effects model was employed to estimate the overall effect size, and statistical heterogeneity was determined based on I2 statistic. Additionally, funnel plot asymmetry was employed to examine publication bias. Finally, subgroup analyses were performed based on age, ML type, and relevant clinical outcomes.

Results

Final analyses incorporated a total of 46 studies. Both the overall and subgroup analyses exhibited considerable statistical heterogeneity. Machine learning predictions for DOC in TBI yielded an overall pooled AUC of 0.83 (95% CI: 0.82-0.84). Subgroup analysis based on age revealed that the ML model in pediatric patients yielded an overall combined AUC of 0.88 (95% CI: 0.80-0.95); among the model subgroups, logistic regression was the most frequently employed, with an overall pooled AUC of 0.85 (95% CI: 0.83-0.87). In the clinical outcome subgroup analysis, the overall pooled AUC for distinguishing between consciousness recovery and consciousness disorders was 0.84 (95% CI: 0.82-0.85).

Conclusion

The findings of this meta-analysis demonstrated outstanding accuracy of ML models in predicting DOC among patients with brain injuries, which presented substantial research value and potential of ML application in this domain.

The Australian Traumatic Brain Injury Initiative: Statement of Working Principles and Rapid Review of Methods to Define Data Dictionaries for Neurological Conditions

The Australian Traumatic Brain Injury Initiative (AUS-TBI) aims to develop a health informatics approach to collect data predictive of outcomes for persons with moderate-severe TBI across Australia. Central to this approach is a data dictionary; however, no systematic reviews of methods to define and develop data dictionaries exist to-date. This rapid systematic review aimed to identify and characterize methods for designing data dictionaries to collect outcomes or variables in persons with neurological conditions. Database searches were conducted from inception through October 2021. Records were screened in two stages against set criteria to identify methods to define data dictionaries for neurological conditions (International Classification of Diseases, 11th Revision: 08, 22, and 23). Standardized data were extracted. Processes were checked at each stage by independent review of a random 25% of records. Consensus was reached through discussion where necessary. Thirty-nine initiatives were identified across 29 neurological conditions. No single established or recommended method for defining a data dictionary was identified. Nine initiatives conducted systematic reviews to collate information before implementing a consensus process. Thirty-seven initiatives consulted with end-users. Methods of consultation were "roundtable" discussion (n = 30); with facilitation (n = 16); that was iterative (n = 27); and frequently conducted in-person (n = 27). Researcher stakeholders were involved in all initiatives and clinicians in 25. Importantly, only six initiatives involved persons with lived experience of TBI and four involved carers. Methods for defining data dictionaries were variable and reporting is sparse. Our findings are instructive for AUS-TBI and can be used to further development of methods for defining data dictionaries.

The Australian Traumatic Brain Injury Initiative: Single Data Dictionary to Predict Outcome for People With Moderate-Severe Traumatic Brain Injury

In this series of eight articles, the Australian Traumatic Brain Injury Initiative (AUS-TBI) consortium describes the Australian approach used to select the common data elements collected acutely that have been shown to predict outcome following moderate-severe traumatic brain injury (TBI) across the lifespan. This article presents the unified single data dictionary, together with additional measures chosen to facilitate comparative effectiveness research and data linkage. Consultations with the AUS-TBI Lived Experience Expert Group provided insights on the merits and considerations regarding data elements for some of the study areas, as well as more general principles to guide the collection of data and the selection of meaningful measures. These are presented as a series of guiding principles and themes. The AUS-TBI Aboriginal and Torres Strait Islander Advisory Group identified a number of key points and considerations for the project approach specific to Aboriginal and Torres Strait Islander peoples, including key issues of data sovereignty and community involvement. These are outlined in the form of principles to guide selection of appropriate methodologies, data management, and governance. Implementation of the AUS-TBI approach aims to maximize ongoing data collection and linkage, to facilitate personalization of care and improved outcomes for people who experience moderate-severe TBI.

Changes in neurologic status after traumatic brain injury in the Resuscitation Outcomes Consortium Hypertonic Saline trial

Objectives

Traumatic brain injury (TBI) is an important public health problem resulting in significant death and disability. Emergency medical services (EMS) personnel often provide initial treatment for TBI, but only limited data describe the long-term course and outcomes of this care. We sought to characterize changes in neurologic status among adults with TBI patients enrolled in the Resuscitation Outcomes Consortium Hypertonic Saline (ROC-HS) trial.

Methods

We used data from the TBI cohort of the ROC-HS trial. The trial included adults with TBI, with Glasgow Coma Scale (GCS) ≤8, and excluded those with shock (systolic blood pressure [SBP] ≤70 or SBP 71-90 with a heart rate [HR] ≥108). The primary outcome was Glasgow Outcome Scale-Extended (GOS-E; 1 = dead, 8 = no disability) determined at (a) hospital discharge and (b) 6-month follow-up. We assessed changes in GOS-E between hospital discharge and 6-month follow-up using descriptive statistics and Sankey graphs.

Results

Among 1279 TBI included in the analysis, GOS-E categories at hospital discharge were as follows: favorable (GOS-E 5-8) 220 (17.2%), unfavorable (GOS-E 2-4) 664 (51.9%), dead (GOS-E 1) 321 (25.1%), and missing 74 (5.8%). GOS-E categories at 6-month follow-up were as follows: favorable 459 (35.9%), unfavorable 279 (21.8%), dead 346 (27.1%), and missing 195 (15.2%). Among initial TBI survivors with complete GOS-E, >96% followed one of three neurologic recovery patterns: (1) favorable to favorable (20.0%), (2) unfavorable to favorable (40.3%), and (3) unfavorable to unfavorable (36.0%). Few patients deteriorated from favorable to unfavorable neurologic status, and there were few additional deaths.

Conclusions

Among TBI receiving initial prehospital care in the ROC-HS trial, changes in 6-month neurologic status followed distinct patterns. Among TBI with unfavorable neurologic status at hospital discharge, almost half improved to favorable neurologic status at 6 months. Among those with favorable neurologic status at discharge, very few worsened or died at 6 months. These findings have important implications for TBI clinical care, research, and trial design.

Low serum calcium is a novel predictor of unfavorable prognosis after traumatic brain injury

Background

Accurate and convenient serological markers for prognosis after traumatic brain injury (TBI) are still lacking. We aimed to explore the predictive value of serum calcium for prognosing outcomes within 6 months after TBI.

Methods

In this multicenter retrospective study, 1255 and 719 patients were included in development and validation cohorts, respectively, and their 6-month prognoses were recorded. Serum calcium was measured through routine blood tests within 24 h of hospital admission. Two multivariate predictive models with or without serum calcium for prognosis were developed. Receiver operating characteristics and calibration curves were applied to estimate their performance.

Results

The patients with lower serum calcium levels had a higher frequency of unfavorable 6-month prognosis than those without. Lower serum calcium level at admission was associated with an unfavorable 6-month prognosis in a wide spectrum of patients with TBI. Lower serum calcium level and our prognostic model including calcium performed well in predicting the 6-month unfavorable outcome. The calcium nomogram maintained excellent performance in discrimination and calibration in the external validation cohort.

Conclusions

Lower serum calcium level upon admission is an independent risk factor for an unfavorable 6-month prognosis after TBI. Integrating serum calcium into a multivariate predictive model improves the performance for predicting 6-month unfavorable outcomes.

Mortality of Surgically Treated Neurotrauma in Elderly Patients and the Development of a Prediction Score: Geriatric Neurotrauma Mortality Score

BACKGROUND

As the population worldwide is aging, the need for surgery in elderly patients with neurotrauma is increasing. The aim of this study was to compare the outcome of elderly patients undergoing surgery for neurotrauma with younger patients and to identify the risk factors for mortality.

METHODS

We retrospectively analyzed consecutive patients undergoing craniotomy or craniectomy for neurotrauma at our institution from 2012 to 2019. Patients were divided into two groups (≥70 years or <70 years) and compared. The primary outcome was the 30-day mortality rate. Potential risk factors for 30-day mortality were assessed in a uni- and multivariate regression model for both age groups, forming the basis of a 30-day mortality prediction score.

RESULTS

We included 163 consecutive patients (average age 57.98 ± 19.87 years); 54 patients were ≥70 years. Patients ≥70 years showed a significantly better median preoperative Glasgow Coma Scale (GCS) score compared with young patients (P < 0.001), and fewer pupil asymmetry (P = 0.001), despite having a higher Marshall score (P = 0.07) at admission. Multivariate regression analysis identified low pre- and postoperative GCS scores and the lack of prompt postoperative prophylactic low-molecular-weight heparin treatment as risk factors for 30-day mortality. Our score showed moderate accuracy in predicting 30-day mortality with an area under the curve of 0.76.

CONCLUSIONS

Elderly patients after neurotrauma present with a better GCS at admission despite having more severe radiographic injuries. Mortality and favorable outcome rates are comparable between the age groups.

Development and Validation of Two Prediction Models for 72-Hour Mortality in High-Risk Trauma Patients Using a Benchmark Dataset: A Comparative Study of Logistic Regression and Neural Networks Models

Background Many studies have been conducted to develop scoring systems for trauma patients, with the majority using logistic regression (LR) models. Neural networks (NN), which is a machine learning algorithm, has a potential to increase the performance of these models. Objectives The aim of this study was to develop and validate two separate prediction models for 72-hour mortality of high-risk trauma patients using LR and NN and to compare the performances of these models in detail. We also aimed to share the SPSS calculators for our models. Materials and methods This is a retrospective, single-center study conducted using a benchmark dataset where the patients were retrospectively gathered from a level 1 trauma center. Patients older than 18 years of age, who had multiple injuries, and were treated at the University Hospital Zurich between January 1, 1996, and January 1, 2013, were included. Patients with a condition that may have an impact on the musculoskeletal system, with Injury Severity Score<16, and with missing outcome data were excluded. Results A total of 3,075 patients were included in the analysis. The area under the curve values of the LR and NN models for predicting 72-hour mortality in patients with high-risk trauma in the hold-out cohort were 0.859 (95% CI=0.836 to 0.883) and 0.856 (95% CI=0.831 to 0.880), respectively. There was no statistically significant difference in the performance of the models (p = 0.554, DeLong's test). Conclusion Both of the models showed good discrimination. Our study suggests that the NN and LR models we developed hold promise as screening tools for predicting 72-hour mortality in high-risk trauma patients. These models were made available to clinicians as clinical prediction tools via SPSS calculators. However, further external validation studies in diverse populations are necessary to substantiate their clinical utility. Moreover, in subsequent studies, it would be beneficial to derive NN models with substantial events per predictor variable to attain more robust and greater predictive accuracy. If the dataset is relatively limited, using LR seems to be a viable alternative.

Prognosis prediction in traumatic brain injury patients using machine learning algorithms

Predicting treatment outcomes in traumatic brain injury (TBI) patients is challenging worldwide. The present study aimed to achieve the most accurate machine learning (ML) algorithms to predict the outcomes of TBI treatment by evaluating demographic features, laboratory data, imaging indices, and clinical features. We used data from 3347 patients admitted to a tertiary trauma centre in Iran from 2016 to 2021. After the exclusion of incomplete data, 1653 patients remained. We used ML algorithms such as random forest (RF) and decision tree (DT) with ten-fold cross-validation to develop the best prediction model. Our findings reveal that among different variables included in this study, the motor component of the Glasgow coma scale, the condition of pupils, and the condition of cisterns were the most reliable features for predicting in-hospital mortality, while the patients' age takes the place of cisterns condition when considering the long-term survival of TBI patients. Also, we found that the RF algorithm is the best model to predict the short-term mortality of TBI patients. However, the generalized linear model (GLM) algorithm showed the best performance (with an accuracy rate of 82.03 ± 2.34) in predicting the long-term survival of patients. Our results showed that using appropriate markers and with further development, ML has the potential to predict TBI patients' survival in the short- and long-term.

AUS-TBI: The Australian Health Informatics Approach to Predict Outcomes and Monitor Intervention Efficacy after Moderate-to-Severe Traumatic Brain Injury

Predicting and optimizing outcomes after traumatic brain injury (TBI) remains a major challenge because of the breadth of injury characteristics and complexity of brain responses. AUS-TBI is a new Australian Government–funded initiative that aims to improve personalized care and treatment for children and adults who have sustained a TBI. The AUS-TBI team aims to address a number of key knowledge gaps, by designing an approach to bring together data describing psychosocial modulators, social determinants, clinical parameters, imaging data, biomarker profiles, and rehabilitation outcomes in order to assess the influence that they have on long-term outcome. Data management systems will be designed to track a broad range of suitable potential indicators and outcomes, which will be organized to facilitate secure data collection, linkage, storage, curation, management, and analysis. It is believed that these objectives are achievable because of our consortium of highly committed national and international leaders, expert committees, and partner organizations in TBI and health informatics. It is anticipated that the resulting large-scale data resource will facilitate personalization, prediction, and improvement of outcomes post-TBI.

Salivary S100 calcium-binding protein beta (S100B) and neurofilament light (NfL) after acute exposure to repeated head impacts in collegiate water polo players

Blood-based biomarkers of brain injury may be useful for monitoring brain health in athletes at risk for concussions. Two putative biomarkers of sport-related concussion, neurofilament light (NfL), an axonal structural protein, and S100 calcium-binding protein beta (S100B), an astrocyte-derived protein, were measured in saliva, a biofluid which can be sampled in an athletic setting without the risks and burdens associated with blood sampled by venipuncture. Samples were collected from men’s and women’s collegiate water polo players (n = 65) before and after a competitive tournament. Head impacts were measured using sensors previously evaluated for use in water polo, and video recordings were independently reviewed for the purpose of validating impacts recorded by the sensors. Athletes sustained a total of 107 head impacts, all of which were asymptomatic (i.e., no athlete was diagnosed with a concussion or more serious). Post-tournament salivary NfL was directly associated with head impact frequency (RR = 1.151, p = 0.025) and cumulative head impact magnitude (RR = 1.008, p = 0.014), while controlling for baseline salivary NfL. Change in S100B was not associated with head impact exposure (RR < 1.001, p > 0.483). These patterns suggest that repeated head impacts may cause axonal injury, even in asymptomatic athletes.

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.

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

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

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

A machine learning model for predicting favorable outcome in severe traumatic brain injury patients after 6 months

Traumatic brain injury (TBI), which occurs commonly worldwide, is among the more costly of health and socioeconomic problems. Accurate prediction of favorable outcome in severe TBI patients could assist with optimizing treatment procedures, predicting clinical outcomes, and result in substantial economic savings. In this study, we examined the capability of a machine learning-based model in predicting "favorable" or "unfavorable" outcome after 6 months in severe TBI patients using only parameters measured on admission. Three models were developed using logistic regression, random forest, and support vector machines trained on parameters recorded from 2,381 severe TBI patients admitted to the neuro-intensive care unit of Rajaee (Emtiaz) Hospital (Shiraz, Iran) between 2015 and 2017. Model performance was evaluated using three indices: sensitivity, specificity, accuracy, and area under the curve (AUC). Ten-fold cross-validation method was used to estimate these indices. Overall, the developed models showed excellent performance with AUC >0.81, sensitivity and specificity of > 0.78. The top-three factors important in predicting 6-month post-trauma survival status in TBI patients are "GCS motor response," "pupillary reactivity," and "age." Machine learning techniques might be used to predict the 6-month outcome in TBI patients using only the parameters measured on admission when the machine learning is trained using a large data set.

Digital signatures for early traumatic brain injury outcome prediction in the intensive care unit

Traumatic brain injury (TBI) is a leading neurological cause of death and disability across the world. Early characterization of TBI severity could provide a window for therapeutic intervention and contribute to improved outcome. We hypothesized that granular electronic health record data available in the first 24 h following admission to the intensive care unit (ICU) can be used to differentiate outcomes at discharge. Working from two ICU datasets we focused on patients with a primary admission diagnosis of TBI whose length of stay in ICU was ≥ 24 h (N = 1689 and 127). Features derived from clinical, laboratory, medication, and physiological time series data in the first 24 h after ICU admission were used to train elastic-net regularized Generalized Linear Models for the prediction of mortality and neurological function at ICU discharge. Model discrimination, determined by area under the receiver operating characteristic curve (AUC) analysis, was 0.903 and 0.874 for mortality and neurological function, respectively. Model performance was successfully validated in an external dataset (AUC 0.958 and 0.878 for mortality and neurological function, respectively). These results demonstrate that computational analysis of data routinely collected in the first 24 h after admission accurately and reliably predict discharge outcomes in ICU stratum TBI patients.

Disability Rating Scale in the First Few Weeks After a Severe Traumatic Brain Injury as a Predictor of 6-Month Functional Outcome

BACKGROUND

An early acute marker of long-term neurological outcome would be useful to help guide clinical decision making and therapeutic effectiveness after severe traumatic brain injury (TBI). We investigated the utility of the Disability Rating Scale (DRS) as early as 1 wk after TBI as a predictor of favorable 6-mo Glasgow Outcome Scale extended (GOS-E).

OBJECTIVE

To determine the predictability of a favorable 6-mo GOS-E using the DRS measured during weeks 1 to 4 of injury.

METHODS

The study is a sub analysis of patients enrolled in the Epo Severe TBI Trial (n = 200) to train and validate L1-regularized logistic regression models. DRS was collected at weeks 1 to 4 and GOS-E at 6 mo.

RESULTS

The average area under the receiver operating characteristic curve was 0.82 for the model with baseline demographic and injury severity variables and week 1 DRS and increased to 0.88 when including weekly DRS until week 4.

CONCLUSION

This study suggests that week 1 to 4 DRS may be predictors of favorable 6-mo outcome in severe TBI patients and thus useful both for clinical prognostication as well as surrogate endpoints for adaptive clinical trials.

The patient with severe traumatic brain injury: clinical decision-making: the first 60 min and beyond

PURPOSE OF REVIEW

There is an urgent need to discuss the uncertainties and paradoxes in clinical decision-making after severe traumatic brain injury (s-TBI). This could improve transparency, reduce variability of practice and enhance shared decision-making with proxies.

RECENT FINDINGS

Clinical decision-making on initiation, continuation and discontinuation of medical treatment may encompass substantial consequences as well as lead to presumed patient benefits. Such decisions, unfortunately, often lack transparency and may be controversial in nature. The very process of decision-making is frequently characterized by both a lack of objective criteria and the absence of validated prognostic models that could predict relevant outcome measures, such as long-term quality and satisfaction with life. In practice, while treatment-limiting decisions are often made in patients during the acute phase immediately after s-TBI, other such severely injured TBI patients have been managed with continued aggressive medical care, and surgical or other procedural interventions have been undertaken in the context of pursuing a more favorable patient outcome. Given this spectrum of care offered to identical patient cohorts, there is clearly a need to identify and decrease existing selectivity, and better ascertain the objective criteria helpful towards more consistent decision-making and thereby reduce the impact of subjective valuations of predicted patient outcome.

SUMMARY

Recent efforts by multiple medical groups have contributed to reduce uncertainty and to improve care and outcome along the entire chain of care. Although an unlimited endeavor for sustaining life seems unrealistic, treatment-limiting decisions should not deprive patients of a chance on achieving an outcome they would have considered acceptable.