Predicting Outcome after Traumatic Brain Injury: Development and Validation of a Prognostic Score Based on Admission Characteristics

The Australian Traumatic Brain Injury Initiative: Systematic Review and Consensus Process to Determine the Predictive Value of Demographic, Injury Event, and Social Characteristics on Outcomes for People With Moderate-Severe Traumatic Brain Injury

The objective of the Australian Traumatic Brain Injury (AUS-TBI) Initiative is to develop a data dictionary to inform data collection and facilitate prediction of outcomes of people who experience moderate-severe TBI in Australia. The aim of this systematic review was to summarize the evidence of the association between demographic, injury event, and social characteristics with outcomes, in people with moderate-severe TBI, to identify potentially predictive indicators. Standardized searches were implemented across bibliographic databases to March 31, 2022. English-language reports, excluding case series, which evaluated the association between demographic, injury event, and social characteristics, and any clinical outcome in at least 10 patients with moderate-severe TBI were included. Abstracts and full text records were independently screened by at least two reviewers in Covidence. A pre-defined algorithm was used to assign a judgement of predictive value to each observed association. The review findings were discussed with an expert panel to determine the feasibility of incorporation of routine measurement into standard care. The search strategy retrieved 16,685 records; 867 full-length records were screened, and 111 studies included. Twenty-two predictors of 32 different outcomes were identified; 7 were classified as high-level (age, sex, ethnicity, employment, insurance, education, and living situation at the time of injury). After discussion with an expert consensus group, 15 were recommended for inclusion in the data dictionary. This review identified numerous predictors capable of enabling early identification of those at risk for poor outcomes and improved personalization of care through inclusion in routine data collection.

Health-related quality of life measures in patients undergoing decompressive craniectomy for severe traumatic brain injury: a 6-year follow-up analysis

PURPOSE

We aimed to assess the long-term neurological outcomes and the functionality and QoL in patients undergoing decompressive craniectomy for severe traumatic brain injury, respectively.

MATERIALS AND METHODS

Among the 120 patients who underwent decompressive craniectomy for severe TBI between 2002 and 2007, 101 were included based on the inclusion criteria. Long-term follow-up results (minimum 3 years) were available for 22 patients. The outcomes were assessed using the Glasgow Outcome Scale (GOS) and the functionality and HRQoL were assessed using the Short Form-36 (SF-36) (v2) and Quality of Life After Brain Injury (QoLIBRI) questionnaires.

RESULTS

Among the patients with severe TBI, 62 (61.4%) died and 39 (38.6%) were discharged to either home or a physical therapy facility. Eleven of the thirty-nine patients could not be reached and were excluded from the final analysis. The mean GOS of the remaining 28 patients was 4.14 ± 0.8 after 6.46 ± 1.64 years of follow-up. The HRQoL was assessed in 22 of the 28 patients. The HRQoL scores were lower in patients with TBI than in healthy controls. Furthermore, there was a significant difference in the HRQoL scores in patients with improved GOS scores than in those with unimproved GOS scores.

CONCLUSIONS

Health-related outcome scores could help clinicians understand the requirements of survivors of severe TBI to create a realistic rehabilitation target for them. QoLIBRI served as a good way of communication in these subjects.

Outcomes After the Surgical Evacuation of Traumatic Acute Subdural Hematomas: The tASDH Risk Score

BACKGROUND

Acute subdural hematoma (ASDH) is a common pathology following traumatic brain injury (TBI). There is sparse data on the prediction of clinical outcomes following traumatic ASDH (tASDH) evacuation. We investigated prognosticators of outcome following evacuation of tASDHs, with subset analysis in a cohort of octogenarians. We developed a scoring system for stratifying the risk of in-hospital mortality for patients undergoing tASDH evacuation.

METHODS

A retrospective chart review was performed to identify all patients who underwent tASDH evacuation. Baseline clinical and demographic data including age, traumatic brain injury mechanism, admission Glasgow Coma Scale (GCS), and Rotterdam computed tomography Scale (RCS) were collected. In-hospital outcomes such as mortality and discharge disposition were collected. A scoring system (tASDH Score) which incorporates RCS (1-2 points), admissions GCS (0-1 points), and age (0-1 point) was created to predict the risk of in-hospital mortality following tASDH evacuation.

RESULTS

Being an octogenarian (OR = 6.91 [2.20-21.71], P = 0.0009), having a GCS of 9-12 (OR = 1.58 [1.32-4.12], P = 0.027) or 3-8 (OR = 2.07 [1.41-10.38], P = 0.018), and having an RCS of 4-6 (OR = 3.49 [1.45-8.44], P = 0.0055) were independently predictive of in-hospital mortality. The in-hospital mortality rate was lower for those with a tASDH score of 1 (10%), compared to those with a score of 2 (12%), 3 (42%), and 4 (100%).

CONCLUSIONS

Octogenarians with an RCS of 4-6 and an admission GCS <13 have a high risk of mortality following tASDH evacuation. Knowledge of which patients are unlikely to survive ASDH evacuation may help guide neurosurgeons in prognostication and goals of care discussions.

Prediction of therapeutic intensity level from automatic multiclass segmentation of traumatic brain injury lesions on CT-scans

The prediction of the therapeutic intensity level (TIL) for severe traumatic brain injury (TBI) patients at the early phase of intensive care unit (ICU) remains challenging. Computed tomography images are still manually quantified and then underexploited. In this study, we develop an artificial intelligence-based tool to segment brain lesions on admission CT-scan and predict TIL within the first week in the ICU. A cohort of 29 head injured patients (87 CT-scans; Dataset1) was used to localize (using a structural atlas), segment (manually or automatically with or without transfer learning) 4 or 7 types of lesions and use these metrics to train classifiers, evaluated with AUC on a nested cross-validation, to predict requirements for TIL sum of 11 points or more during the 8 first days in ICU. The validation of the performances of both segmentation and classification tasks was done with Dice and accuracy scores on a sub-dataset of Dataset1 (internal validation) and an external dataset of 12 TBI patients (12 CT-scans; Dataset2). Automatic 4-class segmentation (without transfer learning) was not able to correctly predict the apparition of a day of extreme TIL (AUC = 60 ± 23%). In contrast, manual quantification of volumes of 7 lesions and their spatial location provided a significantly better prediction power (AUC = 89 ± 17%). Transfer learning significantly improved the automatic 4-class segmentation (DICE scores 0.63 vs 0.34) and trained more efficiently a 7-class convolutional neural network (DICE = 0.64). Both validations showed that segmentations based on transfer learning were able to predict extreme TIL with better or equivalent accuracy (83%) as those made with manual segmentations. Our automatic characterization (volume, type and spatial location) of initial brain lesions observed on CT-scan, publicly available on a dedicated computing platform, could predict requirements for high TIL during the first 8 days after severe TBI. Transfer learning strategies may improve the accuracy of CNN-based segmentation models.Trial registrations Radiomic-TBI cohort; NCT04058379, first posted: 15 august 2019; Radioxy-TC cohort; Health Data Hub index F20220207212747, first posted: 7 February 2022.

Clinical Grading Scales and Neuroprognostication in Acute Brain Injury

Prediction of neurological clinical outcome after acute brain injury is critical because it helps guide discussions with patients and families and informs treatment plans and allocation of resources. Numerous clinical grading scales have been published that aim to support prognostication after acute brain injury. However, the development and validation of clinical scales lack a standardized approach. This in turn makes it difficult for clinicians to rely on prognostic grading scales and to integrate them into clinical practice. In this review, we discuss quality measures of score development and validation and summarize available scales to prognosticate outcomes after acute brain injury. These include scales developed for patients with coma, cardiac arrest, ischemic stroke, nontraumatic intracerebral hemorrhage, subarachnoid hemorrhage, and traumatic brain injury; for each scale, we discuss available validation studies.

Improving Neurotrauma by Depolarization Inhibition With Combination Therapy: A Phase 2 Randomized Feasibility Trial

BACKGROUND AND OBJECTIVES

Spreading depolarizations (SDs) are a pathological mechanism that mediates lesion development in cerebral gray matter. They occur in ∼60% of patients with severe traumatic brain injury (TBI), often in recurring and progressive patterns from days 0 to 10 after injury, and are associated with worse outcomes. However, there are no protocols or trials suggesting how SD monitoring might be incorporated into clinical management. The objective of this protocol is to determine the feasibility and efficacy of implementing a treatment protocol for intensive care of patients with severe TBI that is guided by electrocorticographic monitoring of SDs.

METHODS

Patients who undergo surgery for severe TBI with placement of a subdural electrode strip will be eligible for enrollment. Those who exhibit SDs on electrocorticography during intensive care will be randomized 1:1 to either (1) standard care that is blinded to the further course of SDs or (2) a tiered intervention protocol based on efficacy to suppress further SDs. Interventions aim to block the triggering and propagation of SDs and include adjusted targets for management of blood pressure, CO 2 , temperature, and glucose, as well as ketamine pharmacotherapy up to 4 mg/kg/ hour. Interventions will be escalated and de-escalated depending on the course of SD pathology.

EXPECTED OUTCOMES

We expect to demonstrate that electrocorticographic monitoring of SDs can be used as a real- time diagnostic in intensive care that leads to meaningful changes in patient management and a reduction in secondary injury, as compared with standard care, without increasing medical complications or adverse events.

DISCUSSION

This trial holds potential for personalization of intensive care management by tailoring therapies based on monitoring and confirmation of the targeted neuronal mechanism of SD. Results are expected to validate the concept of this approach, inform refinement of the treatment protocol, and lead to larger-scale trials.

Early Measures of TBI Severity Poorly Predict Later Individual Impairment in a Rat Fluid Percussion Model

BACKGROUND

Multiple measures of injury severity are suggested as common data elements in preclinical traumatic brain injury (TBI) research. The robustness of these measures in characterizing injury severity is unclear. In particular, it is not known how reliably they predict individual outcomes after experimental TBI.

METHODS

We assessed several commonly used measures of initial injury severity for their ability to predict chronic cognitive outcomes in a rat lateral fluid percussion (LFPI) model of TBI. At the time of injury, we assessed reflex righting time, neurologic severity scores, and 24 h weight loss. Sixty days after LFPI, we evaluated working memory using a spontaneous alternation T-maze task.

RESULTS

We found that righting time and weight loss had no correlation to chronic T-maze performance, while neurologic severity score correlated weakly.

DISCUSSION

Taken together, our results indicate that commonly used early measures of injury severity do not robustly predict longer-term outcomes. This finding parallels the uncertainty in predicting individual outcomes in TBI clinical populations.

Using metabolomics to predict severe traumatic brain injury outcome (GOSE) at 3 and 12 months

BACKGROUND

Prognostication is very important to clinicians and families during the early management of severe traumatic brain injury (sTBI), however, there are no gold standard biomarkers to determine prognosis in sTBI. As has been demonstrated in several diseases, early measurement of serum metabolomic profiles can be used as sensitive and specific biomarkers to predict outcomes.

METHODS

We prospectively enrolled 59 adults with sTBI (Glasgow coma scale, GCS ≤ 8) in a multicenter Canadian TBI (CanTBI) study. Serum samples were drawn for metabolomic profiling on the 1st and 4th days following injury. The Glasgow outcome scale extended (GOSE) was collected at 3- and 12-months post-injury. Targeted direct infusion liquid chromatography-tandem mass spectrometry (DI/LC-MS/MS) and untargeted proton nuclear magnetic resonance spectroscopy (¹H-NMR) were used to profile serum metabolites. Multivariate analysis was used to determine the association between serum metabolomics and GOSE, dichotomized into favorable (GOSE 5-8) and unfavorable (GOSE 1-4), outcomes.

RESULTS

Serum metabolic profiles on days 1 and 4 post-injury were highly predictive (Q² > 0.4-0.5) and highly accurate (AUC > 0.99) to predict GOSE outcome at 3- and 12-months post-injury and mortality at 3 months. The metabolic profiles on day 4 were more predictive (Q² > 0.55) than those measured on day 1 post-injury. Unfavorable outcomes were associated with considerable metabolite changes from day 1 to day 4 compared to favorable outcomes. Increased lysophosphatidylcholines, acylcarnitines, energy-related metabolites (glucose, lactate), aromatic amino acids, and glutamate were associated with poor outcomes and mortality.

DISCUSSION

Metabolomic profiles were strongly associated with the prognosis of GOSE outcome at 3 and 12 months and mortality following sTBI in adults. The metabolic phenotypes on day 4 post-injury were more predictive and significant for predicting the sTBI outcome compared to the day 1 sample. This may reflect the larger contribution of secondary brain injury (day 4) to sTBI outcome. Patients with unfavorable outcomes demonstrated more metabolite changes from day 1 to day 4 post-injury. These findings highlighted increased concentration of neurobiomarkers such as N-acetylaspartate (NAA) and tyrosine, decreased concentrations of ketone bodies, and decreased urea cycle metabolites on day 4 presenting potential metabolites to predict the outcome. The current findings strongly support the use of serum metabolomics, that are shown to be better than clinical data, in determining prognosis in adults with sTBI in the early days post-injury. Our findings, however, require validation in a larger cohort of adults with sTBI to be used for clinical practice.

Dynamic predictors of in-hospital and 3-year mortality after traumatic brain injury: A retrospective cohort study

BACKGROUND

Mortality risks after Traumatic Brain Injury (TBI) are understudied in critical illness. We sought to identify risks of mortality in critically ill patients with TBI using time-varying covariates.

METHODS

This single-center, six-year (2006-2012), retrospective cohort study measured demographics, injury characteristics, and daily data of acute TBI patients in the Intensive Care Unit (ICU). Time-varying Cox proportional hazards models assessed in-hospital and 3-year mortality.

RESULTS

Post-TBI ICU patients (n = 2664) experienced 20% in-hospital mortality (n = 529) and 27% (n = 706) 3-year mortality. Glasgow Coma Scale motor subscore (hazard ratio (HR) 0.58, p < 0.001), pupil reactivity (HR 3.17, p < 0.001), minimum glucose (HR 1.44, p < 0.001), mSOFA score (HR 1.81, p < 0.001), coma (HR 2.26, p < 0.001), and benzodiazepines (HR 1.38, p < 0.001) were associated with in-hospital mortality. At three years, public insurance (HR 1.78, p = 0.011) and discharge disposition (HR 4.48, p < 0.001) were associated with death.

CONCLUSIONS

Time-varying characteristics influenced in-hospital mortality post-TBI. Socioeconomic factors primarily affect three-year mortality.

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.

Prediction of motor function in patients with traumatic brain injury using genetic algorithms modified back propagation neural network: A data-based study

Objective

Traumatic brain injury (TBI) is one of the leading causes of death and disability worldwide. In this study, the characteristics of the patients, who were admitted to the China Rehabilitation Research Center, were elucidated in the TBI database, and a prediction model based on the Fugl-Meyer assessment scale (FMA) was established using this database.

Methods

A retrospective analysis of 463 TBI patients, who were hospitalized from June 2016 to June 2020, was performed. The data of the patients used for this study included the age and gender of the patients, course of TBI, complications, and concurrent dysfunctions, which were assessed using FMA and other measures. The information was collected at the time of admission to the hospital and 1 month after hospitalization. After 1 month, a prediction model, based on the correlation analyses and a 1-layer genetic algorithms modified back propagation (GA-BP) neural network with 175 patients, was established to predict the FMA. The correlations between the predicted and actual values of 58 patients (prediction set) were described.

Results

Most of the TBI patients, included in this study, had severe conditions (70%). The main causes of the TBI were car accidents (56.59%), while the most common complication and dysfunctions were hydrocephalus (46.44%) and cognitive and motor dysfunction (65.23 and 63.50%), respectively. A total of 233 patients were used in the prediction model, studying the 11 prognostic factors, such as gender, course of the disease, epilepsy, and hydrocephalus. The correlation between the predicted and the actual value of 58 patients was R ² = 0.95.

Conclusion

The genetic algorithms modified back propagation neural network can predict motor function in patients with traumatic brain injury, which can be used as a reference for risk and prognosis assessment and guide clinical decision-making.

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.

An interpretable neural network for outcome prediction in traumatic brain injury

Background

Traumatic Brain Injury (TBI) is a common condition with potentially severe long-term complications, the prediction of which remains challenging. Machine learning (ML) methods have been used previously to help physicians predict long-term outcomes of TBI so that appropriate treatment plans can be adopted. However, many ML techniques are “black box”: it is difficult for humans to understand the decisions made by the model, with post-hoc explanations only identifying isolated relevant factors rather than combinations of factors. Moreover, such models often rely on many variables, some of which might not be available at the time of hospitalization.

Methods

In this study, we apply an interpretable neural network model based on tropical geometry to predict unfavorable outcomes at six months from hospitalization in TBI patients, based on information available at the time of admission.

Results

The proposed method is compared to established machine learning methods—XGBoost, Random Forest, and SVM—achieving comparable performance in terms of area under the receiver operating characteristic curve (AUC)—0.799 for the proposed method vs. 0.810 for the best black box model. Moreover, the proposed method allows for the extraction of simple, human-understandable rules that explain the model’s predictions and can be used as general guidelines by clinicians to inform treatment decisions.

Conclusions

The classification results for the proposed model are comparable with those of traditional ML methods. However, our model is interpretable, and it allows the extraction of intelligible rules. These rules can be used to determine relevant factors in assessing TBI outcomes and can be used in situations when not all necessary factors are known to inform the full model’s decision.

Electroencephalogram in the intensive care unit: a focused look at acute brain injury

Over the past decades, electroencephalography (EEG) has become a widely applied and highly sophisticated brain monitoring tool in a variety of intensive care unit (ICU) settings. The most common indication for EEG monitoring currently is the management of refractory status epilepticus. In addition, a number of studies have associated frequent seizures, including nonconvulsive status epilepticus (NCSE), with worsening secondary brain injury and with worse outcomes. With the widespread utilization of EEG (spot and continuous EEG), rhythmic and periodic patterns that do not fulfill strict seizure criteria have been identified, epidemiologically quantified, and linked to pathophysiological events across a wide spectrum of critical and acute illnesses, including acute brain injury. Increasingly, EEG is not just qualitatively described, but also quantitatively analyzed together with other modalities to generate innovative measurements with possible clinical relevance. In this review, we discuss the current knowledge and emerging applications of EEG in the ICU, including seizure detection, ischemia monitoring, detection of cortical spreading depolarizations, assessment of consciousness and prognostication. We also review some technical aspects and challenges of using EEG in the ICU including the logistics of setting up ICU EEG monitoring in resource-limited settings.

Predictive Factors of Return Home and Return to Work for Intensive Care Unit Survivors after Traumatic Brain Injury with a Follow-up Period of 2 Years

Intensive care unit (ICU) survivors after traumatic brain injury (TBI) frequently have serious disabilities with subsequent difficulty in reintegration into society. We aimed to investigate outcomes for ICU survivors after moderate to severe TBI (msTBI) and to identify predictive factors of return home (RH) and return to work (RTW). This single-center retrospective cohort study was conducted on all trauma patients admitted to the emergency ICU of our hospital between 2013 and 2017. Of these patients, adult (age ≥ 18 years) msTBI patients with head Abbreviated Injury Scale ≥ 3 were extracted. We performed univariate/multivariate logistic regression analyses to explore the predictive factors of RH and RTW. Among a total of 146 ICU survivors after msTBI, 107 were included (median follow-up period: 26 months). The RH and RTW rates were 78% and 35%, respectively. Multivariate analyses revealed that the predictive factors of RH were age < 65 years (P < 0.001), HR < 76 bpm (P = 0.015), platelet count ≥ 19 × 10⁴/μL (P = 0.0037), D-dimer < 26 μg/mL (P = 0.034), and Glasgow Coma Scale (GCS) score > 8 (P = 0.0015). Similarly, the predictive factors of RTW were age < 65 years (P < 0.001) and GCS score > 8 (P = 0.0039). This study revealed that “age” and “GCS score on admission” affected RH and RTW for ICU survivors after msTBI.

Does poor methodological quality of prediction modeling studies translate to poor model performance? An illustration in traumatic brain injury

BACKGROUND

Prediction modeling studies often have methodological limitations, which may compromise model performance in new patients and settings. We aimed to examine the relation between methodological quality of model development studies and their performance at external validation.

METHODS

We systematically searched for externally validated multivariable prediction models that predict functional outcome following moderate or severe traumatic brain injury. Risk of bias and applicability of development studies was assessed with the Prediction model Risk Of Bias Assessment Tool (PROBAST). Each model was rated for its presentation with sufficient detail to be used in practice. Model performance was described in terms of discrimination (AUC), and calibration. Delta AUC (dAUC) was calculated to quantify the percentage change in discrimination between development and validation for all models. Generalized estimation equations (GEE) were used to examine the relation between methodological quality and dAUC while controlling for clustering.

RESULTS

We included 54 publications, presenting ten development studies of 18 prediction models, and 52 external validation studies, including 245 unique validations. Two development studies (four models) were found to have low risk of bias (RoB). The other eight publications (14 models) showed high or unclear RoB. The median dAUC was positive in low RoB models (dAUC 8%, [IQR - 4% to 21%]) and negative in high RoB models (dAUC - 18%, [IQR - 43% to 2%]). The GEE showed a larger average negative change in discrimination for high RoB models (- 32% (95% CI: - 48 to - 15) and unclear RoB models (- 13% (95% CI: - 16 to - 10)) compared to that seen in low RoB models.

CONCLUSION

Lower methodological quality at model development associates with poorer model performance at external validation. Our findings emphasize the importance of adherence to methodological principles and reporting guidelines in prediction modeling studies.

External validation of prognostic models predicting outcome after chronic subdural hematoma

BACKGROUND

Several prognostic models for outcomes after chronic subdural hematoma (CSDH) treatment have been published in recent years. However, these models are not sufficiently validated for use in daily clinical practice. We aimed to assess the performance of existing prediction models for outcomes in patients diagnosed with CSDH.

METHODS

We systematically searched relevant literature databases up to February 2021 to identify prognostic models for outcome prediction in patients diagnosed with CSDH. For the external validation of prognostic models, we used a retrospective database, containing data of 2384 patients from three Dutch regions. Prognostic models were included if they predicted either mortality, hematoma recurrence, functional outcome, or quality of life. Models were excluded when predictors were absent in our database or available for < 150 patients in our database. We assessed calibration, and discrimination (quantified by the concordance index C) of the included prognostic models in our retrospective database.

RESULTS

We identified 1680 original publications of which 1656 were excluded based on title or abstract, mostly because they did not concern CSDH or did not define a prognostic model. Out of 18 identified models, three could be externally validated in our retrospective database: a model for 30-day mortality in 1656 patients, a model for 2 months, and another for 3-month hematoma recurrence both in 1733 patients. The models overestimated the proportion of patients with these outcomes by 11% (15% predicted vs. 4% observed), 1% (10% vs. 9%), and 2% (11% vs. 9%), respectively. Their discriminative ability was poor to modest (C of 0.70 [0.63-0.77]; 0.46 [0.35-0.56]; 0.59 [0.51-0.66], respectively).

CONCLUSIONS

None of the examined models showed good predictive performance for outcomes after CSDH treatment in our dataset. This study confirms the difficulty in predicting outcomes after CSDH and emphasizes the heterogeneity of CSDH patients. The importance of developing high-quality models by using unified predictors and relevant outcome measures and appropriate modeling strategies is warranted.

Early death prediction in children with traumatic brain injury using computed tomography scoring systems

PURPOSE

Marshall and Rotterdam are the most commonly used CT scoring systems to predict the outcome following traumatic brain injury (TBI). Although several studies have compared the performance of the two scoring systems in adult patients, none of these studies has evaluated the performance of the two scoring systems in pediatric patients. This study aimed to determine the predictive value of the Marshall and Rotterdam scoring systems in pediatric patients with TBI.

METHODS

This retrospective study included 105 children with admission GCS < 12, with a mean age of 6.2 (±3.5) years. Their initial CT and status at hospital discharge (dead or alive) were reviewed, and both the Marshall and Rotterdam scores were calculated. We examined whether each score was related to the early death of pediatric patients.

RESULTS

The pediatric patients with higher Marshall and Rotterdam scores had a higher mortality rate. There was a good correlation between the Marshall and Rotterdam scoring systems (Spearman's rho = 0.618, significant at the 0.05 level). Both systems demonstrated a high degree of discrimination when predicting early mortality. The Marshall scoring system had reasonable discrimination (AUC 0.782), and the Rotterdam scoring system had good discrimination (AUC 0.729). Comparing the two CT scoring systems, the Marshall scoring system provided a better positive predictive value (90%) for early mortality than the Rotterdam scoring system (78%).

CONCLUSIONS

Both the Marshall and Rotterdam scoring systems have good predictability for assessing mortality in pediatric patients with TBI. The performance of the Marshall scoring system was equal to or slightly better than that of the Rotterdam scoring system.

Behavioral and Cognitive Aspects of Concussion

PURPOSE OF REVIEW

This review provides the reader with an overview of concussion and mild traumatic brain injury (TBI). Key aspects of the pathophysiology, signs, and symptoms, treatment and rehabilitation, and recovery from concussion/mild TBI are reviewed with an emphasis on the variety of factors that may contribute to cognitive concerns following injury.

RECENT FINDINGS

Concussion remains a clinical diagnosis based on symptoms that occur in the immediate aftermath of an applied force and in the hours, days, and weeks thereafter. Although advances have been made in advanced diagnostics, including neuroimaging and fluid biomarkers in hopes of developing objective indicators of injury, such markers currently lack sufficient specificity to be used in clinical diagnostics. The symptoms of concussion are heterogeneous and may be seen to form subtypes, each of which suggests a targeted rehabilitation by the interdisciplinary team. Although the majority of patients with concussion recover within the first 30 to 90 days after injury, some have persistent disabling symptoms. The concept of postconcussion syndrome, implying a chronic syndrome of injury-specific symptoms, is replaced by a broader concept of persistent symptoms after concussion. This concept emphasizes the fact that most persistent symptoms have their basis in complex somatic, cognitive, psychiatric, and psychosocial factors related to risk and resilience. This framework leads to the important conclusion that concussion is a treatable injury from which nearly all patients can be expected to recover.

SUMMARY

Concussion/mild TBI is a significant public health problem in civilian, military, and organized athletic settings. Recent advances have led to a better understanding of underlying pathophysiology and symptom presentation and efficacious treatment and rehabilitation of the resulting symptoms. An interdisciplinary team is well-positioned to provide problem-oriented, integrated care to facilitate recovery and to advance the evidence base supporting effective practice in diagnosis, treatment, and prevention.

Outcome Prediction in Patients of Traumatic Brain Injury Based on Midline Shift on CT Scan of Brain

Background Clinicians treating patients with head injury often take decisions based on their assessment of prognosis. Assessment of prognosis could help communication with a patient and the family. One of the most widely used clinical tools for such prediction is the Glasgow coma scale (GCS); however, the tool has a limitation with regard to its use in patients who are under sedation, are intubated, or under the influence of alcohol or psychoactive drugs. CT scan findings such as status of basal cistern, midline shift, associated traumatic subarachnoid hemorrhage (SAH), and intraventricular hemorrhage are useful indicators in predicting outcome and also considered as valid options for prognostication of the patients with traumatic brain injury (TBI), especially in emergency setting.

Materials and Methods 108 patients of head injury were assessed at admission with clinical examination, history, and CT scan of brain. CT findings were classified according to type of lesion and midline shift correlated to GCS score at admission. All the subjects in this study were managed with an identical treatment protocol. Outcome of these patients were assessed on GCS score at discharge.

Results Among patients with severe GCS, 51% had midline shift. The degree of midline shift in CT head was a statistically significant determinant of outcome (p = 0.023). Seventeen out of 48 patients (35.4%) with midline shift had poor outcome as compared with 8 out of 60 patients (13.3%) with no midline shift.

Conclusion In patients with TBI, the degree of midline shift on CT scan was significantly related to the severity of head injury and resulted in poor clinical outcome.

Densitometric analysis of brain computed tomography as a new prognostic factor in patients with acute subdural hematoma

OBJECTIVE

Acute subdural hematoma (ASDH) is a major cause of mortality and morbidity after traumatic brain injury (TBI). Surgical evacuation is the mainstay of treatment in patients with altered neurological status or significant mass effect. Nevertheless, concerns regarding surgical indication still persist. Given that clinicians often make therapeutic decisions on the basis of their prognosis assessment, to accurately evaluate the prognosis is of great significance. Unfortunately, there is a lack of specific and reliable prognostic models. In addition, the interdependence of certain well-known predictive variables usually employed to guide surgical decision-making in ASDH has been proven. Because gray matter and white matter are highly susceptible to secondary insults during the early phase after TBI, the authors aimed to assess the extent of these secondary insults with a brain parenchyma densitometric quantitative CT analysis and to evaluate its prognostic capacity.

METHODS

The authors performed a retrospective analysis among their prospectively collected cohort of patients with moderate to severe TBI. Patients with surgically evacuated, isolated, unilateral ASDH admitted between 2010 and 2017 were selected. Thirty-nine patients were included. For each patient, brain parenchyma density in Hounsfield units (HUs) was measured in 10 selected slices from the supratentorial region. In each slice, different regions of interest (ROIs), including and excluding the cortical parenchyma, were defined. The injured hemisphere, the contralateral hemisphere, and the absolute differences between them were analyzed. The outcome was evaluated using the Glasgow Outcome Scale-Extended at 1 year after TBI.

RESULTS

Fifteen patients (38.5%) had a favorable outcome. Collected demographic, clinical, and radiographic data did not show significant differences between favorable and unfavorable outcomes. In contrast, the densitometric analysis demonstrated that greater absolute differences between both hemispheres were associated with poor outcome. These differences were detected along the supratentorial region, but were greater at the high convexity level. Moreover, these HU differences were far more marked at the cortical parenchyma. It was also detected that these differences were more prone to ischemic and/or edematous insults than to hyperemic changes. Age was significantly correlated with the side-to-side HU differences in patients with unfavorable outcome.

CONCLUSIONS

The densitometric analysis is a promising prognostic tool in patients diagnosed with ASDH. The supplementary prognostic information provided by the densitometric analysis should be evaluated in future studies.

A hierarchical expert-guided machine learning framework for clinical decision support systems: an application to traumatic brain injury prognostication

Prognosis of the long-term functional outcome of traumatic brain injury is essential for personalized management of that injury. Nonetheless, accurate prediction remains unavailable. Although machine learning has shown promise in many fields, including medical diagnosis and prognosis, such models are rarely deployed in real-world settings due to a lack of transparency and trustworthiness. To address these drawbacks, we propose a machine learning-based framework that is explainable and aligns with clinical domain knowledge. To build such a framework, additional layers of statistical inference and human expert validation are added to the model, which ensures the predicted risk score’s trustworthiness. Using 831 patients with moderate or severe traumatic brain injury to build a model using the proposed framework, an area under the receiver operating characteristic curve (AUC) and accuracy of 0.8085 and 0.7488 were achieved, respectively, in determining which patients will experience poor functional outcomes. The performance of the machine learning classifier is not adversely affected by the imposition of statistical and domain knowledge “checks and balances”. Finally, through a case study, we demonstrate how the decision made by a model might be biased if it is not audited carefully.

Radiographic predictors of clinical outcome in traumatic brain injury after decompressive craniectomy

BACKGROUND

Primary decompressive craniectomy (DC) is considered for traumatic brain injury (TBI) patients with clinical deterioration, presenting large amounts of high-density lesions on computed tomography (CT). Postoperative CT findings may be suitable for prognostic evaluation. This study evaluated the radiographic predictors of clinical outcome and survival using pre- and postoperative CT scans of such patients.

METHODS

We enrolled 150 patients with moderate to severe TBI who underwent primary DC. They were divided into two groups based on the 6-month postoperative Glasgow Outcome Scale Extended scores (1-4, unfavorable; 5-8, favorable). Radiographic parameters, including hemorrhage type, location, presence of skull fracture, midline shifting, hemispheric diameter, effacement of cisterns, parenchymal hypodensity, and craniectomy size, were reviewed. Stepwise logistic regression analysis was used to identify the prognostic factors of clinical outcome and 6-month mortality.

RESULTS

Multivariable logistic regression analysis revealed that age (odds ratio [OR] = 1.09; 95% confidence interval [CI] 1.032-1.151; p = 0.002), postoperative low density (OR = 12.58; 95% CI 1.247-126.829; p = 0.032), and postoperative effacement of the ambient cistern (OR = 14.52; 95% CI 2.234-94.351; p = 0.005) and the crural cistern (OR = 4.90; 95% CI 1.359-17.678; p = 0.015) were associated with unfavorable outcomes. Postoperative effacement of the crural cistern was the strongest predictor of 6-month mortality (OR = 8.93; 95% CI 2.747-29.054; p = 0.000).

CONCLUSIONS

Hemispheric hypodensity and effacement of the crural and ambient cisterns on postoperative CT after primary DC seems to associate with poor outcome in patients with TBI.

Cerebrolysin after moderate to severe traumatic brain injury: prospective meta-analysis of the CAPTAIN trial series

INTRODUCTION

This prospective meta-analysis summarizes results from the CAPTAIN trial series, evaluating the effects of Cerebrolysin for moderate-severe traumatic brain injury, as an add-on to usual care.

MATERIALS AND METHODS

The study included two phase IIIb/IV prospective, randomized, double-blind, placebo-controlled clinical trials. Eligible patients with a Glasgow Coma Score (GCS) between 6 and 12 received study medication (50 mL of Cerebrolysin or physiological saline solution per day for ten days, followed by two additional treatment cycles with 10 mL per day for 10 days) in addition to usual care. The meta-analysis comprises the primary ensembles of efficacy criteria for 90, 30, and 10 days after TBI with a priori ordered hypotheses based on multivariate, directional tests.

RESULTS

A total 185 patients underwent meta-analysis (mean admission GCS = 10.3, mean age = 45.3, and mean Baseline Prognostic Risk Score = 2.8). The primary endpoint, a multidimensional ensemble of functional and neuropsychological outcome scales indicated a "small-to-medium" sized effect in favor of Cerebrolysin, statistically significant at Day 30 and at Day 90 (Day 30: MW_combined = 0.60, 95%CI 0.52 to 0.66, p = 0.0156; SMD = 0.31; OR = 1.69; Day 90: MW_combined = 0.60, 95%CI 0.52 to 0.68, p = 0.0146; SMD = 0.34, OR = 1.77). Treatment groups showed comparable safety and tolerability profiles.

DISCUSSION

The meta-analysis of the CAPTAIN trials confirms the safety and efficacy of Cerebrolysin after moderate-severe TBI, opening a new horizon for neurorecovery in this field. Integration of Cerebrolysin into existing guidelines should be considered after careful review of internationally applicable criteria.

Development and temporal external validation of a simple risk score tool for prediction of outcomes after severe head injury based on admission characteristics from level-1 trauma centre of India using retrospectively collected data

OBJECTIVE

To develop and validate a simple risk scores chart to estimate the probability of poor outcomes in patients with severe head injury (HI).

DESIGN

Retrospective.

SETTING

Level-1, government-funded trauma centre, India.

PARTICIPANTS

Patients with severe HI admitted to the neurosurgery intensive care unit during 19 May 2010-31 December 2011 (n=946) for the model development and further, data from same centre with same inclusion criteria from 1 January 2012 to 31 July 2012 (n=284) for the external validation of the model.

OUTCOMES

In-hospital mortality and unfavourable outcome at 6 months.

RESULTS

A total of 39.5% and 70.7% had in-hospital mortality and unfavourable outcome, respectively, in the development data set. The multivariable logistic regression analysis of routinely collected admission characteristics revealed that for in-hospital mortality, age (51-60, >60 years), motor score (1, 2, 4), pupillary reactivity (none), presence of hypotension, basal cistern effaced, traumatic subarachnoid haemorrhage/intraventricular haematoma and for unfavourable outcome, age (41-50, 51-60, >60 years), motor score (1-4), pupillary reactivity (none, one), unequal limb movement, presence of hypotension were the independent predictors as its 95% confidence interval (CI) of odds ratio (OR)_did not contain one. The discriminative ability (area under the receiver operating characteristic curve (95% CI)) of the score chart for in-hospital mortality and 6 months outcome was excellent in the development data set (0.890 (0.867 to 912) and 0.894 (0.869 to 0.918), respectively), internal validation data set using bootstrap resampling method (0.889 (0.867 to 909) and 0.893 (0.867 to 0.915), respectively) and external validation data set (0.871 (0.825 to 916) and 0.887 (0.842 to 0.932), respectively). Calibration showed good agreement between observed outcome rates and predicted risks in development and external validation data set (p>0.05).

CONCLUSION

For clinical decision making, we can use of these score charts in predicting outcomes in new patients with severe HI in India and similar settings.

The Effect of Cerebrolysin on the Predictive Value of Baseline Prognostic Risk Score in Moderate and Severe Traumatic Brain Injury

Cognitive dysfunction is a significant complaint among patients after moderate to severe traumatic brain injury (TBI), with devastating consequences on functional recovery and quality of life. Prognostic models allow a better assessment and management of neurotrauma patients.

The aim of the study was to demonstrate the predictive value of the Baseline Prognostic Risk Score (BPRS) in moderate to severe TBI, in a sample of patients treated with neurotrophic factors.

Eighty patients with moderate-severe TBI from the CAPTAIN II study were included in secondary data analysis. Patients received active treatment with Cerebrolysin, 50 mL per day for ten days, followed by two treatment cycles with 10 mL per day for ten days. BPRS was determined on admission; the age was recorded, and patients were evaluated using the following neurocognitive tests: Mini-Mental State Essay (MMSE), Wechsler Adult Intelligence Scale-Third Edition Processing Speed Index (WAIS-III PSI) and Stroop Colour Word Test-Victoria Version at 10, 30 and 90 days. Hierarchical regression analysis was performed to investigate the unique predictive value of BPRS on cognitive evolution, independent of age. BPRS independently predicted scores on the WAIS-III PSI DSCales and the Word subscale of the Stroop Colour Word Test at 90 days. Age was a significant predictor for all the investigated scales at 10, 30, and 90 days.

This study demonstrates the predictive value of a validated prognostic model (BPRS) for medium-term neurocognitive outcomes in a sample of moderate-severe traumatic brain injury treated with neurotrophic factors.

Factors with the strongest prognostic value associated with in-hospital mortality rate among patients operated for acute subdural and epidural hematoma

INTRODUCTION

Traumatic brain injury (TBI) still remains a serious health problem and is called a "silent epidemic". Each year in Europe 262 per 100,000 individuals suffer from TBI. The most common consequence of severe head injuries include acute subdural (SDH) and epidural hematomas (EDH), which usually require immediate surgically treatment. The aim of our study is to identify factors which have the strongest prognostic value in relation to in-hospital mortality rate among of patients undergoing surgery for EDH and SDH.

PATIENTS AND METHODS

Cohort included 128 patients with isolated craniocerebral injuries who underwent surgery for EDH (28 patients) and SDH (100 patients) in a single, tertiary care Department of Neurosurgery. The data were collected on admission of patients to the Emergency Department and retrospectively analyzed. The following factors were analyzed: demographic data, physiological parameters, laboratory variables, computed tomography scan characteristics and the time between trauma and surgery. Likewise, we have investigated the in-hospital mortality of patients at the time of discharge.

RESULTS

We found that the factors with the strongest prognostic values were: the initial GCS score, respiratory rate, glycaemia, blood saturation, systolic blood pressure, midline shift and type of hematoma. Additionally, we proved that a drop by one point in the GCS score almost doubles the risk of in-hospital death while the presence of coagulopathy increases the risk of in-hospital death almost six times.

CONCLUSION

Most of the factors with the strongest prognostic value are factors that the emergency team can treat prior to the hospital admission. Coagulopathy, however that has the strongest influence on in-hospital death rate can only be efficiently treated in a hospital setting.

Prognostic Value of Spreading Depolarizations in Patients With Severe Traumatic Brain Injury

Importance

Advances in treatment of traumatic brain injury are hindered by the inability to monitor pathological mechanisms in individual patients for targeted neuroprotective treatment. Spreading depolarizations, a mechanism of lesion development in animal models, are a novel candidate for clinical monitoring in patients with brain trauma who need surgery.

Objective

To test the null hypothesis that spreading depolarizations are not associated with worse neurologic outcomes.

Design, Setting, and Participants

This prospective, observational, multicenter cohort study was conducted from February 2009 to August 2013 in 5 level 1 trauma centers. Consecutive patients who required neurological surgery for treatment of acute brain trauma and for whom research consent could be obtained were enrolled; participants were excluded because of technical problems in data quality, patient withdrawal, or loss to follow-up. Primary statistical analysis took place from April to December 2018. Evaluators of outcome assessments were blinded to other measures.

Interventions

A 6-contact electrode strip was placed on the brain surface during surgery for continuous electrocorticography during intensive care.

Main Outcomes and Measures

Electrocorticography was scored for depolarizations, following international consensus procedures. Six-month outcomes were assessed by the Glasgow Outcome Scale-Extended score.

Results

A total of 157 patients were initially enrolled; 19 were subsequently excluded. The 138 remaining patients (104 men [75%]; median [interquartile range] age, 45 [29-64] years) underwent a median (interquartile range) of 75.5 (42.2-117.1) hours of electrocorticography. A total of 2837 spreading depolarizations occurred in 83 of 138 patients (60.1% incidence) who, compared with patients who did not have spreading depolarizations, had lower prehospital systolic blood pressure levels (mean [SD], 133 [31] mm Hg vs 146 [33] mm Hg; P = .03), more traumatic subarachnoid hemorrhage (depolarization incidences of 17 of 37 [46%], 18 of 32 [56%], 22 of 33 [67%], and 23 of 30 patients [77%] for Morris-Marshall Grades 0, 1, 2, and 3/4, respectively; P = .047), and worse radiographic pathology (in 38 of 73 patients [52%] and 42 of 60 patients [70%] for Rotterdam Scores 2-4 vs 5-6, respectively; P = .04). Of patients with depolarizations, 32 of 83 (39%) had only sporadic events that induced cortical spreading depression of spontaneous electrical activity, whereas 51 of 83 patients (61%) exhibited temporal clusters of depolarizations (≥3 in a 2-hour span). Nearly half of those with clusters (23 of 51 [45%]) also had depolarizations in an electrically silent area of the cortex (isoelectric spreading depolarization). Patients with clusters did not improve in motor neurologic examinations from presurgery to postelectrocorticography, while other patients did improve. In multivariate ordinal regression adjusting for baseline prognostic variables, the occurrence of depolarization clusters had an odds ratio of 2.29 (95% CI, 1.13-4.65; P = .02) for worse outcomes.

Conclusions and Relevance

In this cohort study of patients with acute brain trauma, spreading depolarizations were predominant but heterogeneous and independently associated with poor neurologic recovery. Monitoring the occurrence of spreading depolarizations may identify patients most likely to benefit from targeted management strategies.

Improvement of quality of life (QOL) between 1 and 3-4 years after traumatic brain injury (TBI) in ICU patients

BACKGROUND

Evaluation of changes in quality of life (QOL) in ICU patients several years after traumatic brain injury (TBI) is not well documented.

METHODS

A prospective cohort study was conducted in all patients with TBI admitted between 2004 and 2008 to the ICU of Regional Hospital of Malaga (Spain). Functional status was evaluated by Glasgow Outcome Scale (GOS) and QOL by PAECC (Project for the Epidemiologic Analysis of Critical Care patients) questionnaire between 0 (normal QOL) to 29 points (worst QOL).

RESULTS

A total of 531 patients. Median(Quartile1,Quartile 3) age: 35 (22, 56) years. After 3-4 years, 175 died (33%). Survivor QOL was deteriorated (median total PAECC score: 5 (0, 11) points) although 75.76% of patients who survived showed good functional situation (GOS normal or mild dysfunction). An improvement in QOL scores between 1 and 3-4 years was observed (median PAECC score differences between 3-4 years and 1 year: - 1(- 4, 0) points). QOL score improved during this interval of time: 62.6% of patients. Change in QOL was related by multivariate analysis to admission cranial-computed tomography scan (Marshall's classification), age, and Injury Severity Score (ISS), with the biggest improvement seen in younger patients and with more severe ISS. Basic physiological activities were maintained in the majority of patients. Subjective aspects and working activities improved between 1 and 3-4 years but with a high proportion still impaired in these items after 3-4 years.

CONCLUSIONS

ICU patients with TBI after 1 year show improvement in QOL between 1 and 3-4 years, with the biggest improvement in QOL seen in younger patients and in those with more severe ISS.

Factors Associated with Poor Outcomes in Patients with Mild or Moderate Acute Subdural Hematomas

The factors influencing the outcomes of mild/moderate acute subdural hematoma (ASDH) are still unclear. Retrospective analyses were performed to identify such factors. The medical records of all patients who were admitted to Saiseikai Shiga Hospital with mild (Glasgow Coma Scale [GCS] score of 14–15) or moderate (GCS score of 9–13) ASDH between April 2008 and March 2017 were reviewed. Comparisons between the patients who exhibited favorable and poor outcomes were performed. Then, independent factors that contributed to poor outcomes were identified via logistic regression analyses. A total of 266 patients with a mean age of 70.2 were included in this study. The most common concomitant injuries were subarachnoid hemorrhages (SAHs; 56.8%). The patients’ Injury Severity Scores (ISS) ranged from 16 to 75 (median: 21). The 66 moderate ASDH patients exhibited significantly higher frequencies of surgery and mortality (24.2% and 13.6%, respectively) than the 200 mild ASDH patients (8.0% and 4.5%, respectively). The factors associated with poor outcomes were age (odds ratio [OR]: 1.06) and the ISS (OR: 1.24) in the mild ASDH patients, and older age (OR: 1.09) and the higher ISS (OR: 1.15) in the moderate group, too.

Enteral combined with parenteral nutrition improves clinical outcomes in patients with traumatic brain injury

Objective: To investigate the effect of nutritional support on nutritional status and clinical outcomes of patients with traumatic brain injury (TBI).Methods: Sixty-one patients with TBI from the intensive care unit and neurosurgery of Xianyang Central Hospital from 2017 to 2019 were retrospectively included. General and clinical data of the study subjects were collected. The control group (n = 28) received parenteral nutrition alone, and the observation group (n = 33) received parenteral nutrition combined with enteral nutrition. The general conditions and biochemical indicators of both groups of patients were divided into two groups of ≤8 and ≥9 for stratified analysis to compare the nutritional support status and infection complications during hospitalization Occurrence, ICU length of stay, total length of stay, total cost of stay, and prognostic indicators of the patients were analyzed and compared.Results: There were no significant differences in biochemical indicators between both groups of patients when they were discharged. Among patients with GCS ≤8 points, the incidence of lung infection in the observer was significantly higher than that in the control group (P < 0.001), but the incidence of intracranial infection, stress ulcers, and diarrhea was not statistically different from that in the control group (P = 0.739). No significant differences were observed in hospitalization time and hospitalization costs between both groups (P = 0.306 and P = 0.079, respectively). The observation group was significantly better than the control group in GSC score and long-term quality of life score (P = 0.042 and P = 0.025, respectively). When GCS was ≥ 9 points, there was no statistical difference in the incidence of lung infections and intracranial infections between both groups of patients (P = 0.800 and P = 0.127, respectively). The observation group was significantly higher than the control group in terms of length of hospital stay, nasal feeding time and hospitalization costs (P < 0.001, P < 0.001 and P = 0.006, respectively). The observation group was significantly better than the control group in GSC score and long-term quality of life score (P = 0.001 and P = 0.015, respectively). There was no significant difference in the incidence of pulmonary infection and intracranial infection between both groups of patients (P = 0.800 and P = 0.127, respectively).Conclusion: Enteral nutrition combined with parenteral nutrition intervention has a positive effect on the clinical prognosis of TBI patients.

The Initial Factors with Strong Predictive Value in Relation to Six-Month Outcome among Patients Operated due to Extra-Axial Hematomas

Introduction: Traumatic brain injuries (TBI) are a real social problem, with an upward trend worldwide. The most frequent consequence of a traumatic brain injury is extra-axial hemorrhage, i.e., an acute subdural (SDH) and epidural hematoma (EDH). Most of the factors affecting the prognosis have been analyzed on a wide group of traumatic brain injuries. Nonetheless, there are few studies analyzing factors influencing the prognosis regarding patients undergoing surgery due to acute subdural and epidural hematoma. The aim of this study was to identify the factors which have the strongest prognostic value in relation to the 6-month outcome of the patients undergoing surgery for SDH and EDH. Patients and methods: The study included a group of 128 patients with isolated craniocerebral injuries. Twenty eight patients were operated upon due to EDH, and a group of 100 patients were operated upon due to SDH. The following factors from the groups were analyzed: demographic data, physiological factors, laboratory factors, computed tomography scan characteristics, and time between the trauma and the surgery. All of these factors were correlated in a multivariate analysis with the six-month outcome in the Glasgow outcome scale. Results: The factors with the strongest prognostic value are GCS score, respiration rate, saturation, glycaemia and systolic blood pressure. Conclusion: Initial GCS score, respiratory rate, saturation, glycaemia and systolic blood pressure were the factors with the strongest prognostic value.

Epidemiology and outcome of trauma victims admitted in trauma centers of tertiary care hospitals - A multicentric study in India

Background:

Roadside trauma in India is an increasingly significant problem, particularly because of bad roads, irregular road signs, overcrowding, overspeeding, and bad traffic etiquettes. Adequate information on the characteristics of victims, causes of accidents, frequency, vehicles involved, alcohol intake, and outcome of management is essential for understanding and planning for better management.

Aim:

This study aimed to determine the characteristics of trauma (roadside accidents) victims admitted to various trauma centers in India. The purpose of this study is to examine the epidemiology of trauma within a local community in India through data gained from the different emergency centers and to analyze trauma patients to find the predictors that led to the deaths of trauma patients.

Materials and Methods:

The present observational study involved trauma victims over 1-year period in three centers. Demographical details recorded were age, sex, alcohol intake, systolic blood pressure on arrival, respiratory rate, Glasgow Coma Scale (GCS) score, the interval between injury and admission, Injury Severity Score (ISS) risk factors, hospital stay, and outcome.

Results:

A total of 2650 injuries were recorded in 2466 patients. The mean age was 42.45 ± 15.7 years, the mean ISS was 13.82 ± 6.2, and the mean GCS was 12.20 ± 4.1. The mean time to admission at different trauma centres was 48.41 ± 172.8 h. The head injury was the most common (29.52%).

Conclusion:

Road side accidents due to overspeeding was the most common cause whereas driving under the effect of alcohol was the second most common cause. Accidents are common because of bad traffic etiquette on Indian roads.

Prediction of Early TBI Mortality Using a Machine Learning Approach in a LMIC Population

Background: In a time when the incidence of severe traumatic brain injury (TBI) is increasing in low- to middle-income countries (LMICs), it is important to understand the behavior of predictive variables in an LMIC's population. There are few previous attempts to generate prediction models for TBI outcomes from local data in LMICs. Our study aim is to design and compare a series of predictive models for mortality on a new cohort in TBI patients in Brazil using Machine Learning.

Methods: A prospective registry was set in São Paulo, Brazil, enrolling all patients with a diagnosis of TBI that require admission to the intensive care unit. We evaluated the following predictors: gender, age, pupil reactivity at admission, Glasgow Coma Scale (GCS), presence of hypoxia and hypotension, computed tomography findings, trauma severity score, and laboratory results.

Results: Overall mortality at 14 days was 22.8%. Models had a high prediction performance, with the best prediction for overall mortality achieved through Naive Bayes (area under the curve = 0.906). The most significant predictors were the GCS at admission and prehospital GCS, age, and pupil reaction. When predicting the length of stay at the intensive care unit, the Conditional Inference Tree model had the best performance (root mean square error = 1.011), with the most important variable across all models being the GCS at scene.

Conclusions: Models for early mortality and hospital length of stay using Machine Learning can achieve high performance when based on registry data even in LMICs. These models have the potential to inform treatment decisions and counsel family members.

Level of evidence: This observational study provides a level IV evidence on prognosis after TBI.

Continuous Electroencephalography After Moderate to Severe Traumatic Brain Injury

OBJECTIVES

After traumatic brain injury, continuous electroencephalography is widely used to detect electrographic seizures. With the development of standardized continuous electroencephalography terminology, we aimed to describe the prevalence and burden of ictal-interictal patterns, including electrographic seizures after moderate-to-severe traumatic brain injury and to correlate continuous electroencephalography features with functional outcome.

DESIGN

Post hoc analysis of the prospective, randomized controlled phase 2 multicenter INTREPID study (ClinicalTrials.gov: NCT00805818). Continuous electroencephalography was initiated upon admission to the ICU. The primary outcome was the 3-month Glasgow Outcome Scale-Extended. Consensus electroencephalography reviews were performed by raters certified in standardized continuous electroencephalography terminology blinded to clinical data. Rhythmic, periodic, or ictal patterns were referred to as "ictal-interictal continuum"; severe ictal-interictal continuum was defined as greater than or equal to 1.5 Hz lateralized rhythmic delta activity or generalized periodic discharges and any lateralized periodic discharges or electrographic seizures.

SETTING

Twenty U.S. level I trauma centers.

PATIENTS

Patients with nonpenetrating traumatic brain injury and postresuscitation Glasgow Coma Scale score of 4-12 were included.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

Among 152 patients with continuous electroencephalography (age 34 ± 14 yr; 88% male), 22 (14%) had severe ictal-interictal continuum including electrographic seizures in four (2.6%). Severe ictal-interictal continuum burden correlated with initial prognostic scores, including the International Mission for Prognosis and Analysis of Clinical Trials in Traumatic Brain Injury (r = 0.51; p = 0.01) and Injury Severity Score (r = 0.49; p = 0.01), but not with functional outcome. After controlling clinical covariates, unfavorable outcome was independently associated with absence of posterior dominant rhythm (common odds ratio, 3.38; 95% CI, 1.30-9.09), absence of N2 sleep transients (3.69; 1.69-8.20), predominant delta activity (2.82; 1.32-6.10), and discontinuous background (5.33; 2.28-12.96) within the first 72 hours of monitoring.

CONCLUSIONS

Severe ictal-interictal continuum patterns, including electrographic seizures, were associated with clinical markers of injury severity but not functional outcome in this prospective cohort of patients with moderate-to-severe traumatic brain injury. Importantly, continuous electroencephalography background features were independently associated with functional outcome and improved the area under the curve of existing, validated predictive models.

Neurostereologic Lesion Volumes and Spreading Depolarizations in Severe Traumatic Brain Injury Patients: A Pilot Study

BACKGROUND

Spreading depolarizations (SDs) occur in 50-60% of patients after surgical treatment of severe traumatic brain injury (TBI) and are independently associated with unfavorable outcomes. Here we performed a pilot study to examine the relationship between SDs and various types of intracranial lesions, progression of parenchymal damage, and outcomes.

METHODS

In a multicenter study, fifty patients (76% male; median age 40) were monitored for SD by continuous electrocorticography (ECoG; median duration 79 h) following surgical treatment of severe TBI. Volumes of hemorrhage and parenchymal damage were estimated using unbiased stereologic assessment of preoperative, postoperative, and post-ECoG serial computed tomography (CT) studies. Neurologic outcomes were assessed at 6 months by the Glasgow Outcome Scale-Extended.

RESULTS

Preoperative volumes of subdural and subarachnoid hemorrhage, but not parenchymal damage, were significantly associated with the occurrence of SDs (P's < 0.05). Parenchymal damage increased significantly (median 34 ml [Interquartile range (IQR) - 2, 74]) over 7 (5, 8) days from preoperative to post-ECoG CT studies. Patients with and without SDs did not differ in extent of parenchymal damage increase [47 ml (3, 101) vs. 30 ml (- 2, 50), P = 0.27], but those exhibiting the isoelectric subtype of SDs had greater initial parenchymal damage and greater increases than other patients (P's < 0.05). Patients with temporal clusters of SDs (≥ 3 in 2 h; n = 10 patients), which included those with isoelectric SDs, had worse outcomes than those without clusters (P = 0.03), and parenchymal damage expansion also correlated with worse outcomes (P = 0.01). In multivariate regression with imputation, both clusters and lesion expansion were significant outcome predictors.

CONCLUSIONS

These results suggest that subarachnoid and subdural blood are important primary injury factors in provoking SDs and that clustered SDs and parenchymal lesion expansion contribute independently to worse patient outcomes. These results warrant future prospective studies using detailed quantification of TBI lesion types to better understand the relationship between anatomic and physiologic measures of secondary injury.

Efficacy and safety of cerebrolysin in neurorecovery after moderate-severe traumatic brain injury: results from the CAPTAIN II trial

INTRODUCTION

The objective of this trial was to evaluate the efficacy and safety of Cerebrolysin in treating patients after moderate to severe traumatic brain injury (TBI) as an adjunct to standard care protocols. The trial was designed to investigate the clinical effects of Cerebrolysin in the acute (neuroprotective) stage and during early and long-term recovery as part of a neurorestorative strategy.

MATERIALS AND METHODS

The study was a phase IIIb/IV single-center, prospective, randomized, double-blind, placebo-controlled clinical trial. Eligible patients with a Glasgow Coma Score (GCS) between 7 and 12 received study medication (50 ml of Cerebrolysin or physiological saline solution per day for 10 days, followed by two additional treatment cycles with 10 ml per day for 10 days) in addition to standard care. We tested ensembles of efficacy criteria for 90, 30, and 10 days after TBI with a priori ordered hypotheses using a multivariate, directional test, to reflect the global status of patients after TBI.

RESULTS

The study enrolled 142 patients, of which 139 underwent formal analysis (mean age = 47.4, mean admission GCS = 10.4, and mean Baseline Prognostic Risk Score = 2.6). The primary endpoint, a multidimensional ensemble of 13 outcome scales, indicated a "small-to-medium"-sized effect in favor of Cerebrolysin, statistically significant at day 90 (MW_combined = 0.59, 95% CI 0.52 to 0.66, P = 0.0119). Safety and tolerability observations were comparable between treatment groups.

CONCLUSION

Our trial confirms previous beneficial effects of the multimodal, biological agent Cerebrolysin for overall outcome after moderate to severe TBI, as measured by a multidimensional approach. Study findings must be appraised and aggregated in conjunction with existing literature, as to improve the overall level of insight regarding therapeutic options for TBI patients. The widely used pharmacologic intervention may benefit from a large-scale observational study to map its use and to establish comparative effectiveness in real-world clinical settings.

Prognostic performance of computerized tomography scoring systems in civilian penetrating traumatic brain injury: an observational study

Background

The prognosis of penetrating traumatic brain injury (pTBI) is poor yet highly variable. Current computerized tomography (CT) severity scores are commonly not used for pTBI prognostication but may provide important clinical information in these cohorts.

Methods

All consecutive pTBI patients from two large neurotrauma databases (Helsinki 1999–2015, Stockholm 2005–2014) were included. Outcome measures were 6-month mortality and unfavorable outcome (Glasgow Outcome Scale 1–3). Admission head CT scans were assessed according to the following: Marshall CT classification, Rotterdam CT score, Stockholm CT score, and Helsinki CT score. The discrimination (area under the receiver operating curve, AUC) and explanatory variance (pseudo-R²) of the CT scores were assessed individually and in addition to a base model including age, motor response, and pupil responsiveness.

Results

Altogether, 75 patients were included. Overall 6-month mortality and unfavorable outcome were 45% and 61% for all patients, and 31% and 51% for actively treated patients. The CT scores’ AUCs and pseudo-R²s varied between 0.77–0.90 and 0.35–0.60 for mortality prediction and between 0.85–0.89 and 0.50–0.57 for unfavorable outcome prediction. The base model showed excellent performance for mortality (AUC 0.94, pseudo-R² 0.71) and unfavorable outcome (AUC 0.89, pseudo-R² 0.53) prediction. None of the CT scores increased the base model’s AUC (p > 0.05) yet increased its pseudo-R² (0.09–0.15) for unfavorable outcome prediction.

Conclusion

Existing head CT scores demonstrate good-to-excellent performance in 6-month outcome prediction in pTBI patients. However, they do not add independent information to known outcome predictors, indicating that a unique score capturing the intracranial severity in pTBI may be warranted.

Electronic supplementary material

The online version of this article (10.1007/s00701-019-04074-1) contains supplementary material, which is available to authorized users.

Predictors of recovery in moderate to severe traumatic brain injury

OBJECTIVE

Traumatic brain injury (TBI) is a significant cause of morbidity and mortality worldwide. Clinical outcomes in TBI are determined by the severity of injury, which is dependent on the primary and secondary brain injury processes. Whereas primary brain injury lesions are related to the site of impact, secondary brain injury results from physiological changes caused by oxidative stress and inflammatory responses that occur after the primary insult. The aim of this study was to identify important clinical and biomarker profiles that were predictive of recovery after moderate to severe TBI. A good functional outcome was defined as a Glasgow Outcome Scale (GOS) score of ≥ 4.

METHODS

This was a prospective study of patients with moderate to severe TBI managed at the Nelson Mandela Academic Hospital during the period between March 2014 and March 2016. Following admission and initial management, the patient demographic data (sex, age) and admission Glasgow Coma Scale score were recorded. Oxidative stress and inflammatory biomarkers in blood and CSF were sampled on days 1-7. On day 14, only blood was sampled for the same biomarkers. The primary outcome was the GOS score-due to its simplicity, the GOS was used to assess clinical outcomes at day 90. Because of difficulty in performing regular follow-up due to the vastness of the region, difficult terrain, and long travel distances, a 3-month follow-up period was used to avoid default.

RESULTS

Sixty-four patients with Glasgow Coma Scale scores of ≤ 12 were seen and managed. Among the 56 patients who survived, 42 showed significant recovery (GOS score ≥ 4) at 3 months. Important predictors of recovery included antioxidant activity in the CSF (superoxide dismutase and total antioxidant capacity).

CONCLUSIONS

Recovery after TBI was dependent on the resolution of oxidative stress imbalance.

Neurosurgical Outcomes of Isolated Hemorrhagic Mild Traumatic Brain Injury

Introduction Mild traumatic brain injury (TBI) is common but its management is variable. Objectives To describe the acute natural history of isolated hemorrhagic mild TBI. Methods This was a single-center, retrospective chart review of 661 patients. Inclusion criteria were consecutive patients with hemorrhagic mild TBI. Exclusion criteria were any other acute traumatic injury and significant comorbidities. Variables recorded included neurosurgical intervention and timing, mortality, emergency room disposition, intensive care unit (ICU) length of stay (LOS), discharge disposition, repeat computed tomography head (CTH) indications and results, neurologic exam, age, sex, Glasgow Coma Scale (GCS) score, and hemorrhage type. Results Overall intervention and unexpected delayed intervention rates were 9.4% and 1.5%, respectively. The mortality rate was 2.4%. A 10-year age increase had 26% greater odds of intervention (95% CI, 9.6-45%; P<.001) and 53% greater odds of mortality (95% CI, 11-110%; P=.009). A one-point GCS increase had 49% lower odds of intervention (95% CI, 25-66%; P<.001) and 50% lower odds of mortality (95% CI, 1-75%; P=.047). Subdural and epidural hemorrhages were more likely to require intervention (P=.02). ICU admission was associated with discharge to an acute care facility (OR, 2.9; 95% CI, 1.4-6.0; P=.003). Neurologic exam changes were associated with a worsened CTH scan (OR, 12.3; 95% CI, 7.0-21.4; P<.001) and intervention (OR, 15.1; 95% CI, 8.4-27.2; P<.001). Conclusions Isolated hemorrhagic mild TBI patients are at a low, but not clinically insignificant, risk of intervention and mortality.

Safety and efficacy of Cerebrolysin in acute brain injury and neurorecovery: CAPTAIN I-a randomized, placebo-controlled, double-blind, Asian-Pacific trial

OBJECTIVE

To evaluate the safety and efficacy of Cerebrolysin as an add-on therapy to local standard treatment protocol in patients after moderate-to-severe traumatic brain injury.

METHODS

The patients received the study medication in addition to standard care (50 mL of Cerebrolysin or physiological saline solution daily for 10 days, followed by two additional treatment cycles with 10 mL daily for 10 days) in a prospective, randomized, double-blind, placebo-controlled, parallel-group, multi-centre phase IIIb/IV trial. The primary endpoint was a multidimensional ensemble of 14 outcome scales pooled to be analyzed by means of the multivariate, correlation-sensitive Wei-Lachin procedure.

RESULTS

In 46 enrolled TBI patients (Cerebrolysin 22, placebo 24), three single outcomes showed stand-alone statistically significant superiority of Cerebrolysin [Stroop Word/Dots Interference (p = 0.0415, Mann-Whitney(MW) = 0.6816, 95% CI 0.51-0.86); Color Trails Tests 1 and 2 (p = 0.0223/0.0170, MW = 0.72/0.73, 95% CI 0.53-0.90/0.54-0.91), both effect sizes lying above the benchmark for "large" superiority (MW > 0.71)]. While for the primary multivariate ensemble, statistical significance was just missed in the intention-to-treat population (p_Wei-Lachin < 0.1, MW_combined = 0.63, 95% CI 0.48-0.77, derived standardized mean difference (SMD) 0.45, 95% CI -0.07 to 1.04, derived OR 2.1, 95% CI 0.89-5.95), the per-protocol analysis showed a statistical significant superiority of Cerebrolysin (p_Wei-Lachin = 0.0240, MW_combined = 0.69, 95% CI 0.53 to 0.85, derived SMD 0.69, 95% CI 0.09 to 1.47, derived OR 3.2, 95% CI 1.16 to 12.8), with effect sizes of six single outcomes lying above the benchmark for "large" superiority. Safety aspects were comparable to placebo.

CONCLUSION

Our trial suggests beneficial effects of Cerebrolysin on outcome after TBI. Results should be confirmed by a larger RCT with a comparable multidimensional approach.

Comparison between logistic regression and machine learning algorithms on survival prediction of traumatic brain injuries

PURPOSE

To compare twenty-two machine learning (ML) models against logistic regression on survival prediction in severe traumatic brain injury (STBI) patients in a single center study.

MATERIALS AND METHODS

Data was collected from STBI patients admitted to the Sichuan Provincial People's Hospital between December 2009 and November 2011. Twenty-two machine learning (ML) models were tested, and their predictive performance compared with logistic regression (LR) model. Receiver operating characteristics (ROC), area under curve (AUC), accuracy, F-score, precision, recall and Decision Curve Analysis (DCA) were used as performance metrics.

RESULTS

A total of 117 patients were enrolled. AUC of all ML models ranged from 86.3% to 94%. AUC of LR was 83%, and accuracy was 88%. The AUC of Cubic SVM, Quadratic SVM and Linear SVM were higher than that of LR. The precision ratio of LR was 95% and recall ratio was 91%, both were lower than most ML models. The F-Score of LR was 0.93, which was only slightly better than that of Linear Discriminant and Quadratic Discriminant.

CONCLUSIONS

The twenty-two ML models selected have capabilities comparable to classical LR model for outcome prediction in STBI patients. Of these, Cubic SVM, Quadratic SVM, Linear SVM performed significantly better than LR.

Prognosis in Moderate and Severe Traumatic Brain Injury: A Systematic Review of Contemporary Models and Validation Studies

Outcome prognostication in traumatic brain injury (TBI) is important but challenging due to heterogeneity of the disease. The aim of this systematic review is to present the current state-of-the-art on prognostic models for outcome after moderate and severe TBI and evidence on their validity. We searched for studies reporting on the development, validation or extension of prognostic models for functional outcome after TBI with Glasgow Coma Scale (GCS) ≤12 published between 2006-2018. Studies with patients age ≥14 years and evaluating a multi-variable prognostic model based on admission characteristics were included. Model discrimination was expressed with the area under the receiver operating characteristic curve (AUC), and model calibration with calibration slope and intercept. We included 58 studies describing 67 different prognostic models, comprising the development of 42 models, 149 external validations of 31 models, and 12 model extensions. The most common predictors were GCS (motor) score (n = 55), age (n = 54), and pupillary reactivity (n = 48). Model discrimination varied substantially between studies. The International Mission for Prognosis and Analysis of Clinical Trials (IMPACT) and Corticoid Randomisation After Significant Head injury (CRASH) models were developed on the largest cohorts (8509 and 10,008 patients, respectively) and were most often externally validated (n = 91), yielding AUCs ranging between 0.65-0.90 and 0.66-1.00, respectively. Model calibration was reported with a calibration intercept and slope for seven models in 53 validations, and was highly variable. In conclusion, the discriminatory validity of the IMPACT and CRASH prognostic models is supported across a range of settings. The variation in calibration, reflecting heterogeneity in reliability of predictions, motivates continuous validation and updating if clinical implementation is pursued.

Mild, moderate and severe: terminology implications for clinical and experimental traumatic brain injury

PURPOSE OF REVIEW

When describing clinical or experimental traumatic brain injury (TBI), the adjectives 'mild,' 'moderate' and 'severe' are misleading. 'Mild' clinical TBI frequently results in long-term disability. 'Severe' rodent TBI actually resembles mild or complicated mild clinical TBI.

RECENT FINDINGS

Many mild TBI patients appear to have recovered completely but have postconcussive symptoms, deficits in cognitive and executive function and reduced cerebral blood flow. After moderate TBI, 31.8% of patients died or were discharged to skilled nursing or hospice. Among survivors of moderate and severe TBI, 44% were unable to return to work. On MRI, 88% of mild TBI patients have evidence of white matter damage, based on measurements of fractional anisotropy and mean diffusivity/apparent diffusion coefficient. After sports concussion, clinically recovered patients have abnormalities in functional connectivity on functional MRI. Methylphenidate improved fatigue and cognitive impairment and, combined with cognitive rehabilitation, improved memory and executive functioning. In comparison to clinical TB, because the entire spectrum of experimental rodent TBI, although defined as moderate or severe, more closely resembles mild or complicated mild clinical TBI.

SUMMARY

Many patients after mild or moderate TBI suffer long-term sequelae and should be considered a major target for translational research. Treatments that improve outcome in rodent TBI, even when the experimental injuries are defined as severe, might be most applicable to mild or moderate TBI.

Isolated subdural hematomas in mild traumatic brain injury. Part 1: the association between radiographic characteristics and neurosurgical intervention

OBJECTIVE

Isolated subdural hematomas (iSDHs) are one of the most common intracranial hemorrhage (ICH) types in the population with mild traumatic brain injury (mTBI; Glasgow Coma Scale score 13-15), account for 66%-75% of all neurosurgical procedures, and have one of the highest neurosurgical intervention rates. The objective of this study was to examine how quantitative hemorrhage characteristics of iSDHs in patients with mTBI at admission are associated with subsequent neurosurgical intervention.

METHODS

This was a 3.5-year, retrospective observational cohort study at a Level I trauma center. All adult trauma patients with mTBI and iSDHs were included in the study. Maximum length and thickness (in mm) of acute SDHs, the presence of acute-on-chronic SDH, mass effect, and other hemorrhage-related variables were double-data entered; discrepant results were adjudicated after a maximum of 4 reviews. Patients with coagulopathy, skull fractures, no acute hemorrhage, a non-SDH ICH, or who did not undergo imaging on admission were excluded. The primary outcome was neurosurgical intervention (craniotomy, burr hole, catheter drainage of SDH, placement of intracranial pressure monitor, shunt, or ventriculostomy). Multivariate stepwise logistic regression was used to identify significant covariates and to assess interactions.

RESULTS

A total of 176 patients were included in our study: 28 patients did and 148 patients did not receive a neurosurgical intervention. Increasing head Abbreviated Injury Scale score was significantly associated with neurosurgical interventions. There was a strong correlation between the first 3 reviews on maximum hemorrhage length (R2 = 0.82) and maximum hemorrhage thickness (R2 = 0.80). The neurosurgical intervention group had a mean maximum SDH length and thickness that were 61 mm longer and 13 mm thicker than those of the nonneurosurgical intervention group (p < 0.001 for both). After adjusting for the presence of an acute-on-chronic hemorrhage, for every 1-mm increase in the thickness of an iSDH, the odds of a neurosurgical intervention increase by 32% (95% CI 1.16-1.50). There were no interventions for any SDH with a maximum thickness ≤ 5 mm on initial presenting scan.

CONCLUSIONS

This is the first study to quantify the odds of a neurosurgical intervention based on hemorrhage characteristics in patients with an iSDH and mTBI. Once validated in a second population, these data can be used to better inform patients and families of the risk of future neurosurgical intervention, and to evaluate the necessity of interhospital transfers.

Isolated subdural hematomas in mild traumatic brain injury. Part 2: a preliminary clinical decision support tool for neurosurgical intervention

OBJECTIVE

A paucity of studies have examined neurosurgical interventions in the mild traumatic brain injury (mTBI) population with intracranial hemorrhage (ICH). Furthermore, it is not understood how the dimensions of an ICH relate to the risk of a neurosurgical intervention. These limitations contribute to a lack of treatment guidelines. Isolated subdural hematomas (iSDHs) are the most prevalent ICH in mTBI, carry the highest neurosurgical intervention rate, and account for an overwhelming majority of all neurosurgical interventions. Decision criteria in this population could benefit from understanding the risk of requiring neurosurgical intervention. The aim of this study was to quantify the risk of neurosurgical intervention based on the dimensions of an iSDH in the setting of mTBI.

METHODS

This was a 3.5-year, retrospective observational cohort study at a Level I trauma center. All adult (≥ 18 years) trauma patients with mTBI and iSDH were included in the study. Maximum length and thickness (in mm) of acute SDHs, the presence of acute-on-chronic (AOC) SDH, mass effect, and other hemorrhage-related variables were double-data entered; discrepant results were adjudicated after a maximum of 4 reviews. Patients with coagulopathy, skull fractures, no acute hemorrhage, a non-SDH ICH, or who did not undergo imaging on admission were excluded. Tentorial SDHs were not measured. The primary outcome was neurosurgical intervention (craniotomy, burr holes, intracranial pressure monitor placement, shunt, ventriculostomy, or SDH evacuation). Multivariate stepwise logistic regression was used to identify significant covariates, to assess interactions, and to create the scoring system.

RESULTS

There were a total of 176 patients included in our study: 28 patients did and 148 did not receive a neurosurgical intervention. There were no significant differences between neurosurgical intervention groups in 11 demographic and 22 comorbid variables. Patients with neurosurgical intervention had significantly longer and thicker SDHs than nonsurgical controls. Logistic regression identified thickness and AOC hemorrhage as being the most important variables in predicting neurosurgical intervention; SDH length was not. Risk of neurosurgical intervention was calculated based on the SDH thickness and presence of an AOC hemorrhage from a multivariable logistic regression model (area under the receiver operating characteristic curve 0.94, 95% CI 0.90-0.97; p < 0.001). With a decision point of 2.35% risk, we predicted neurosurgical intervention with 100% sensitivity, 100% negative predictive value, and 53% specificity.

CONCLUSIONS

This is the first study to quantify the risk of neurosurgical intervention based on hemorrhage characteristics in patients with mTBI and iSDH. Once validated in a second population, these data can be used to inform the necessity of interhospital transfers and neurosurgical consultations.

The development and internal validation of a model to predict functional recovery after trauma

Objective

To develop and internally validate the PROgnosis of functional recovery after Trauma (PRO-Trauma) prediction model.

Design

A prospective single-center longitudinal cohort study. Patients were assessed at 6 weeks and 12 months post-injury.

Methods

Patients that presented at the emergency department with an acute traumatic injury, were prompted for participation. Patients that completed the assessments at 6 weeks and 12 months post injury were included. Exclusion criteria: age < 18, age > 65, pathologic fractures, injuries that resulted in severe neurologic deficits. The predicted outcome, functional recovery, was defined as a Short Musculoskeletal Function Assessment (SMFA-NL) Problems with Daily Activities (PDA) subscale ≤ 12.2 points at 12 months post-injury (Dutch population norm). Predictors were: gender, age, living with partner, number of chronic health conditions, SMFA-NL PDA score 6 weeks post-injury, ICU admission, length of stay in hospital, injury severity score, occurrence of complications and treatment type. All predictors were obtained before 6 weeks post-injury. Missing data were multiply imputed. Predictor variables were selected using backward stepwise multivariable logistic regression. Hosmer-Lemeshow tests were used to evaluate calibration. Bootstrap resampling was used to internally validate the final model.

Results

A total of 246 patients were included, of which 104 (44%) showed functional recovery. The predictors in the final PRO-Trauma model were: living with partner, the number of chronic health conditions, SMFA-NL PDA subscale score at 6 weeks post-injury and length of stay in hospital. The apparent R² was 0.33 [0.33;0.34], the c-statistic was 0.79 [0.79;0.80]. Hosmer-Lemeshow test indicated good calibration (p = 0.92). Optimism-corrected R² was 0.28 [0.27;0.29] and the optimism-corrected Area Under the Curve was 0.77 [0.77;0.77].

Conclusion

The PRO-Trauma prediction model can be used to obtain valid predictions of attaining functional recovery after trauma at 12 months post-injury. The PRO-Trauma prediction model showed acceptable calibration and discrimination.