Traumatic Brain Injury Prognostic Models Research Articles

Machine learning models to predict traumatic brain injury outcomes in Tanzania: Using delays to emergency care as predictors.

Constraints to emergency department resources may prevent the timely provision of care following a patient's arrival to the hospital. In-hospital delays may adversely affect health outcomes, particularly among trauma patients who require prompt management. Prognostic models can help optimize resource allocation thereby reducing in-hospital delays and improving trauma outcomes. The objective of this study was to investigate the predictive value of delays to emergency care in machine learning based traumatic brain injury (TBI) prognostic models. Our data source was a TBI registry from Kilimanjaro Christian Medical Centre Emergency Department in Moshi, Tanzania. We created twelve unique variables representing delays to emergency care and included them in eight different machine learning based TBI prognostic models that predict in-hospital outcome. Model performance was compared using the area under the receiver operating characteristic curve (AUC). Inclusion of our twelve time to care variables improved predictability in each of our eight prognostic models. Our Bayesian generalized linear model produced the largest AUC, with a value of 89.5 (95% CI: 88.8, 90.3). Time to care variables were among the most important predictors of in-hospital outcome in our best three performing models. In low-resource settings where delays to care are highly prevalent and contribute to high mortality rates, incorporation of care delays into prediction models that support clinical decision making may benefit both emergency medicine physicians and trauma patients by improving prognostication performance.

Performance of CRASH and IMPACT Prognostic Models for Traumatic Brain Injury at 12 and 24 Months Post-Injury.

The Corticoid Randomization after Significant Head Injury (CRASH) and International Mission for Prognosis and Analysis of Clinical Trials (IMPACT) prognostic models are the most reported prognostic models for traumatic brain injury (TBI) in the scientific literature. However, these models were developed and validated to predict 6-month unfavorable outcome and mortality, and growing evidence supports continuous improvements in functional outcome after severe TBI up to 2 years post-injury. The purpose of this study was to evaluate CRASH and IMPACT model performance beyond 6 months post-injury to include 12 and 24 months post-injury. Discriminative validity remained consistent over time and comparable to earlier recovery time points (area under the curve = 0.77-0.83). Both models had poor fit for unfavorable outcomes, explaining less than one quarter of the variation in outcomes for severe TBI patients. The CRASH model had significant values for the Hosmer-Lemeshow test at 12 and 24 months, indicating poor model fit past the previous validation point. There is concern in the scientific literature that TBI prognostic models are being used by neurotrauma clinicians to support clinical decision making despite the goal of the models' development being to support research study design. The results of this study indicate that the CRASH and IMPACT models should not be used in routine clinical practice because of poor model fit that worsens over time and the large, unexplained variance in outcomes.

Development and Verification of Prognostic Prediction Models for Patients with Brain Trauma Based on Coagulation Function Indexes.

Objective To assess the effect of adding coagulation indices to the currently existing prognostic prediction models of traumatic brain injury (TBI) in the prediction of outcome. Methods A total of 210 TBI patients from 2017 to 2019 and 131 TBI patients in 2020 were selected for development and internal verification of the new model. The primary outcomes include death at 14 days and Glasgow Outcome Score (GOS) at 6 months. The performance of each model is evaluated by means of discrimination (area under the curve (AUC)), calibration (Hosmer-Lemeshow (H-L) goodness-of-fit test), and precision (Brier score). Results The IMPACT Core model showed better prediction ability than the CRASH Basic model. Adding one coagulation index at a time to the IMPACT Core model, the new combined models IMPACT Core+FIB and IMPACT Core+APTT are optimal for the 6-month unfavorable outcome and 6-month mortality, respectively (AUC, 0.830 and 0.878). The new models were built based on the regression coefficients of the models. Internal verification indicated that for the prediction of 6-month unfavorable outcome and 6-month mortality, both the IMPACT Core+FIB model and the IMPACT Core+APTT model show better discrimination (AUC, 0.823 vs. 0.818 and 0.853 vs. 0.837), better calibration (HL, p = 0.114 and p = 0.317) and higher precision (Brier score, 0.148 vs. 0.141 and 0.147 vs. 0.164), respectively, than the original models. Conclusion Our research shows that the combination of the traumatic brain injury prognostic models and coagulation indices can improve the 6-month outcome prediction of patients with TBI.

Machine Learning for Predicting Discharge Disposition After Traumatic Brain Injury.

Current traumatic brain injury (TBI) prognostic calculators are commonly used to predict the mortality and Glasgow Outcome Scale, but these outcomes are most relevant for severe TBI. Because mild and moderate TBI rarely reaches severe outcomes, there is a need for novel prognostic endpoints. To generate machine learning (ML) models with a strong predictive capacity for trichotomized discharge disposition, an outcome not previously used in TBI prognostic models. The outcome can serve as a proxy for patients' functional status, even in mild and moderate patients with TBI. Using a large data set (n = 5292) of patients with TBI from a quaternary care center and 84 predictors, including vitals, demographics, mechanism of injury, initial Glasgow Coma Scale, and comorbidities, we trained 6 different ML algorithms using a nested-stratified-cross-validation protocol. After optimizing hyperparameters and performing model selection, isotonic regression was applied to calibrate models. When maximizing the microaveraged area under the receiver operating characteristic curve during hyperparameter optimization, a random forest model exhibited top performance. A random forest model was also selected when maximizing the microaveraged area under the precision-recall curve. For both models, the weighted average area under the receiver operating characteristic curves was 0.84 (95% CI 0.81-0.87) and the weighted average area under the precision-recall curves was 0.85 (95% CI 0.82-0.88). Our group presents high-performing ML models to predict trichotomized discharge disposition. These models can assist in optimization of patient triage and treatment, especially in cases of mild and moderate TBI.

Intracranial Densitometry-Augmented Machine Learning Enhances the Prognostic Value of Brain CT in Pediatric Patients With Traumatic Brain Injury: A Retrospective Pilot Study.

Background: The inter- and intrarater variability of conventional computed tomography (CT) classification systems for evaluating the extent of ischemic-edematous insult following traumatic brain injury (TBI) may hinder the robustness of TBI prognostic models.Objective: This study aimed to employ fully automated quantitative densitometric CT parameters and a cutting-edge machine learning algorithm to construct a robust prognostic model for pediatric TBI.Methods: Fifty-eight pediatric patients with TBI who underwent brain CT were retrospectively analyzed. Intracranial densitometric information was derived from the supratentorial region as a distribution representing the proportion of Hounsfield units. Furthermore, a machine learning-based prognostic model based on gradient boosting (i.e., CatBoost) was constructed with leave-one-out cross-validation. At discharge, the outcome was assessed dichotomously with the Glasgow Outcome Scale (favorability: 1–3 vs. 4–5). In-hospital mortality, length of stay (>1 week), and need for surgery were further evaluated as alternative TBI outcome measures.Results: Densitometric parameters indicating reduced brain density due to subtle global ischemic changes were significantly different among the TBI outcome groups, except for need for surgery. The skewed intracranial densitometry of the unfavorable outcome became more distinguishable in the follow-up CT within 48 h. The prognostic model augmented by intracranial densitometric information achieved adequate AUCs for various outcome measures [favorability = 0.83 (95% CI: 0.72–0.94), in-hospital mortality = 0.91 (95% CI: 0.82–1.00), length of stay = 0.83 (95% CI: 0.72–0.94), and need for surgery = 0.71 (95% CI: 0.56–0.86)], and this model showed enhanced performance compared to the conventional CRASH-CT model.Conclusion: Densitometric parameters indicative of global ischemic changes during the acute phase of TBI are predictive of a worse outcome in pediatric patients. The robustness and predictive capacity of conventional TBI prognostic models might be significantly enhanced by incorporating densitometric parameters and machine learning techniques.

Corticosteroid Randomization after Significant Head Injury and International Mission for Prognosis and Clinical Trials in Traumatic Brain Injury Models Compared with a Machine Learning-Based Predictive Model from Tanzania.

Hospitals in low- and middle-income countries (LMICs) could benefit from decision support technologies to reduce time to triage, diagnosis, and surgery for patients with traumatic brain injury (TBI). Corticosteroid Randomization after Significant Head Injury (CRASH) and International Mission for Prognosis and Clinical Trials in Traumatic Brain Injury (IMPACT) are robust examples of TBI prognostic models, although they have yet to be validated in Sub-Saharan Africa (SSA). Moreover, machine learning and improved data quality in LMICs provide an opportunity to develop context-specific, and potentially more accurate, prognostic models. We aim to externally validate CRASH and IMPACT on our TBI registry and compare their performances to that of the locally derived model (from the Kilimanjaro Christian Medical Center [KCMC]). We developed a machine learning-based prognostic model from a TBI registry collected at a regional referral hospital in Moshi, Tanzania. We also used the core CRASH and IMPACT online risk calculators to generate risk scores for each patient. We compared the discrimination (area under the curve [AUC]) and calibration before and after Platt scaling (Brier, Hosmer-Lemeshow Test, and calibration plots) for CRASH, IMPACT, and the KCMC model. The outcome of interest was unfavorable in-hospital outcome defined as a Glasgow Outcome Scale score of 1-3. There were 2972 patients included in the TBI registry, of whom 11% had an unfavorable outcome. The AUCs for the KCMC model, CRASH, and IMPACT were 0.919, 0.876, and 0.821, respectively. Prior to Platt scaling, CRASH was the best calibrated model (χ2 = 68.1) followed by IMPACT (χ2 = 380.9) and KCMC (χ2 = 1025.6). We provide the first SSA validation of the core CRASH and IMPACT models. The KCMC model had better discrimination than either of these. CRASH had the best calibration, although all model predictions could be successfully calibrated. The top performing models, KCMC and CRASH, were both developed using LMIC data, suggesting that locally derived models may outperform imported ones from different contexts of care. Further work is needed to externally validate the KCMC model.

An Attitude Survey and Assessment of the Feasibility, Acceptability, and Usability of a Traumatic Brain Injury Decision Support Tool in Uganda

Traumatic brain injury (TBI) prognostic models are potential solutions to severe human and technical shortages. Although numerous TBI prognostic models have been developed, none are widely used in clinical practice, largely because of a lack of feasibility research to inform implementation. We previously developed a prognostic model and Web-based application for in-hospital TBI care in low-resource settings. In this study, we tested the feasibility, acceptability, and usability of the application with potential end-users. We performed our feasibility assessment with providers involved in TBI care at both a regional and national referral hospital in Uganda. We collected qualitative and quantitative data on decision support needs, application ease of use, and implementation design. We completed 25 questionnaires on potential uses of the app and 11 semistructured feasibility interviews. Top-cited uses were informing the decision to operate, informing the decision to send the patient to intensive care, and counseling patients and relatives. Participants affirmed the potential of the application to support difficult triage situations, particularly in the setting of limited access to diagnostics and interventions, but were hesitant to use this technology with end-of-life decisions. Although all participants were satisfied with the application and agreed that it was easy to use, several expressed a need for this technology to be accessible by smartphone and offline. We elucidated several potential uses for our app and important contextual factors that will support future implementation. This investigation helps address an unmet need to determine the feasibility of TBI clinical decision support systems in low-resource settings.

Prognostic significance of abnormal hematological parameters in severe traumatic brain injury requiring decompressive craniectomy

Hematological abnormalities after severe traumatic brain injury (TBI) are common, and are associated with a poor outcome. Whether these abnormalities offer additional prognostic significance over and beyond validated TBI prognostic models is uncertain. This retrospective cohort study compared the ability of admission hematological abnormalities to that of the IMPACT (International Mission for Prognosis and Analysis of Clinical Trials) prognostic model to predict 18-month neurological outcome of 388 patients who required a decompressive craniectomy after severe TBI, between 2004 and 2016, in Western Australia. Area under the receiver operating characteristic (AUROC) curve was used to assess predictors' ability to discriminate between patients with and without an unfavorable outcome of death, vegetative state, or severe disability. Of the 388 patients included in the study, 151 (38.9%) had an unfavorable outcome at 18 months after decompressive craniectomy for severe TBI. Abnormalities in admission hemoglobin (AUROC 0.594, p = 0.002), plasma glucose (AUROC 0.592, p = 0.002), fibrinogen (AUROC 0.563, p = 0.036), international normalized ratio (INR; AUROC 0.645, p = 0.001), activated partial thromboplastin time (AUROC 0.564, p = 0.033), and disseminated intravascular coagulation score (AUROC 0.623, p = 0.001) were all associated with a higher risk of unfavorable outcome at 18 months after severe TBI. As a marker of inflammation, neutrophil to lymphocyte ratio was not significantly associated with the risk of unfavorable outcome (AUROC 0.500, p = 0.998). However, none of these parameters, in addition to the platelet count, were significantly associated with an unfavorable outcome after adjusting for the IMPACT predicted risk (odds ratio [OR] per 10% increment in risk 2.473, 95% confidence interval [CI] 2.061-2.967; p = 0.001). After excluding 8 patients (2.1%) who were treated with warfarin prior to the injury, there was a suggestion that INR was associated with some additional prognostic significance (OR 3.183, 95% CI 0.856-11.833; p = 0.084) after adjusting for the IMPACT predicted risk. In isolation, INR was the best hematological prognostic parameter in severe TBI requiring decompressive craniectomy, especially when patients treated with warfarin were excluded. However, the prognostic significance of admission hematological abnormalities was mostly captured by the IMPACT prognostic model, such that they did not offer any additional prognostic information beyond the IMPACT predicted risk. These results suggest that new prognostic factors for TBI should be evaluated in conjunction with predicted risks of a comprehensive prognostic model that has been validated, such as the IMPACT prognostic model.

Changes in the gray and white matter of patients with ischemic-edematous insults after traumatic brain injury.

Gray matter (GM) and white matter (WM) are vulnerable to ischemic-edematous insults after traumatic brain injury (TBI). The extent of secondary insult after brain injury is quantifiable using quantitative CT analysis. One conventional quantitative CT measure, the gray-white matter ratio (GWR), and a more recently proposed densitometric analysis are used to assess the extent of these insults. However, the prognostic capacity of the GWR in patients with TBI has not yet been validated. This study aims to test the prognostic value of the GWR and evaluate the alternative parameters derived from the densitometric analysis acquired during the acute phase of TBI. In addition, the prognostic ability of the conventional TBI prognostic models (i.e., IMPACT [International Mission for Prognosis and Analysis of Clinical Trials in TBI] and CRASH [Corticosteroid Randomisation After Significant Head Injury] models) were compared to that of the quantitative CT measures. Three hundred patients with TBI of varying ages (92 pediatric, 94 adult, and 114 geriatric patients) and admitted between 2008 and 2013 were included in this retrospective cohort study. The normality of the density of the deep GM and whole WM was evaluated as the proportion of CT pixels with Hounsfield unit values of 31-35 for GM and 26-30 for WM on CT images of the entire supratentorial brain. The outcome was evaluated using the Glasgow Outcome Scale (GOS) at discharge (GOS score ≤ 3, n = 100). Lower proportions of normal densities in the deep GM and whole WM indicated worse outcomes. The proportion of normal WM exhibited a significant prognostic capacity (area under the curve [AUC] = 0.844). The association between the outcome and the normality of the WM density was significant in adult (AUC = 0.792), pediatric (AUC = 0.814), and geriatric (AUC = 0.885) patients. In pediatric patients, the normality of the overall density and the density of the GM were indicative of the outcome (AUC = 0.751). The average GWR was not associated with the outcome (AUC = 0.511). IMPACT and CRASH models showed adequate and reliable performance in the pediatric and geriatric groups but not in the adult group. The highest overall predictive performance was achieved by the densitometry-augmented IMPACT model (AUC = 0.881). Both deep GM and WM are susceptible to ischemic-edematous insults during the early phase of TBI. The extent of the secondary injury was better evaluated by analyzing the normality of the deep GM and WM rather than by calculating the GWR.

Geriatric Traumatic Brain Injury: Epidemiology, Outcomes, Knowledge Gaps, and Future Directions.

This review of the literature on traumatic brain injury (TBI) in older adults focuses on incident TBI sustained in older adulthood ("geriatric TBI") rather than on the separate, but related, topic of older adults with a history of earlier-life TBI. We describe the epidemiology of geriatric TBI, the impact of comorbidities and pre-injury function on TBI risk and outcomes, diagnostic testing, management issues, outcomes, and critical directions for future research. The highest incidence of TBI-related emergency department visits, hospitalizations, and deaths occur in older adults. Higher morbidity and mortality rates among older versus younger individuals with TBI may contribute to an assumption of futility about aggressive management of geriatric TBI. However, many older adults with TBI respond well to aggressive management and rehabilitation, suggesting that chronological age and TBI severity alone are inadequate prognostic markers. Yet there are few geriatric-specific TBI guidelines to assist with complex management decisions, and TBI prognostic models do not perform optimally in this population. Major barriers in management of geriatric TBI include under-representation of older adults in TBI research, lack of systematic measurement of pre-injury health that may be a better predictor of outcome and response to treatment than age and TBI severity alone, and lack of geriatric-specific TBI common data elements (CDEs). This review highlights the urgent need to develop more age-inclusive TBI research protocols, geriatric TBI CDEs, geriatric TBI prognostic models, and evidence-based geriatric TBI consensus management guidelines aimed at improving short- and long-term outcomes for the large and growing geriatric TBI population.

BRAIN TOPiC Study: Assessing Variability in Traumatic Brain Injury (TBI) Outcome Prognostication – Do Self-Fulfilling Prophecies Exist in TBI, Too? (P05.202)

OBJECTIVE: In this study, we surveyed clinicians caring for patients with moderate-severe traumatic brain injury (msTBI) to assess (1) possible variability in outcome prognostication in TBI, varying by clinicians level of training and medical specialty, (2) possible biases and self-fulfilling prophecies, and (3) whether specific ICU medical complications may influence clinicians in their outcome prognostication. BACKGROUND: Patients with msTBI commonly die from withdrawal of support, likely as a consequence of an unfavorable outcome prognosis provided to the family by the treating physician. It is unknown whether prognostication may lead to self-fulfilling prophecies, and whether the presence of intensive care unit (ICU) complications may accentuate possible provider bias. DESIGN/METHODS: We conducted an anonymous electronic survey of clinicians, including faculty members (Neurology, Neurosurgery, Trauma, Anesthesia/Critical Care), neurology house staff, ICU affiliate practitioners and neuroICU nurses at a single Level I trauma center. The survey included three TBI case vignettes and their respective ICU courses. Questions were designed to assess the utilization of known TBI prognostic models, relative importance of ICU complications for outcome prognostication and aggressiveness of care recommended by the survey participant. RESULTS: The survey response rate was 72% (106 surveys returned). In all 3 cases, the majority of participants did not recommend withdrawal of care, but did predict unfavorable 6-month outcomes. 51% of participants consider medical ICU complications as very important in TBI prognostication. Age, ICU course and head CT findings are the prognostic variables considered most important to outcomes. CONCLUSIONS: We have discovered great variability in outcome predictions made by clinicians with different levels of experience in treating msTBI. Self-fulfilling prophecies may exist in msTBI outcomes. Outcome estimates should focus not only on admission variables, but also on ICU complications in order to guide clinicians in providing prognostication. Disclosure: Dr. Izzy has nothing to disclose. Dr. Compton has nothing to disclose. Dr. Carandang has nothing to disclose. Dr. Hall has nothing to disclose. Dr. Muehlschlegel has nothing to disclose.

Comparing the prognostic performance of S100B with prognostic models in traumatic brain injury

IntroductionThere are currently two prognostic tools available for predicting outcome in traumatic brain injury (TBI): prognostic models combining clinico-demographic characteristics of patients for outcome prediction and brain injury biomarkers (eg,...