Concurrent Types of Intracranial Hemorrhage are Associated with a Higher Mortality Rate in Adult Patients with Traumatic Subarachnoid Hemorrhage: A Cross-Sectional Retrospective Study

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

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

Epidemiological and clinical characteristics predictive of ICU mortality of patients with traumatic brain injury treated at a trauma referral hospital - a cohort study

BACKGROUND

Traumatic brain injury (TBI) has substantial physical, psychological, social and economic impacts, with high rates of morbidity and mortality. Considering its high incidence, the aim of this study was to identify epidemiological and clinical characteristics that predict mortality in patients hospitalized for TBI in intensive care units (ICUs).

METHODS

A retrospective cohort study was carried out with patients over 18 years old with TBI admitted to an ICU of a Brazilian trauma referral hospital between January 2012 and August 2019. TBI was compared with other traumas in terms of clinical characteristics of ICU admission and outcome. Univariate and multivariate analyses were used to estimate the odds ratio for mortality.

RESULTS

Of the 4816 patients included, 1114 had TBI, with a predominance of males (85.1%). Compared with patients with other traumas, patients with TBI had a lower mean age (45.3 ± 19.1 versus 57.1 ± 24.1 years, p < 0.001), higher median APACHE II (19 versus 15, p < 0.001) and SOFA (6 versus 3, p < 0.001) scores, lower median Glasgow Coma Scale (GCS) score (10 versus 15, p < 0.001), higher median length of stay (7 days versus 4 days, p < 0.001) and higher mortality (27.6% versus 13.3%, p < 0.001). In the multivariate analysis, the predictors of mortality were older age (OR: 1.008 [1.002-1.015], p = 0.016), higher APACHE II score (OR: 1.180 [1.155-1.204], p < 0.001), lower GCS score for the first 24 h (OR: 0.730 [0.700-0.760], p < 0.001), greater number of brain injuries and presence of associated chest trauma (OR: 1.727 [1.192-2.501], p < 0.001).

CONCLUSION

Patients admitted to the ICU for TBI were younger and had worse prognostic scores, longer hospital stays and higher mortality than those admitted to the ICU for other traumas. The independent predictors of mortality were older age, high APACHE II score, low GCS score, number of brain injuries and association with chest trauma.

Relationship between platelet indices and red cell distribution width and short-term mortality in traumatic brain injury with 30-day mortality

OBJECTIVE

This study aimed at investigating whether there is a relationship between 7- or 30-day mortality and mean platelet volume, platelet distribution width, platelet count-to-total lymphocyte count ratio, or red cell distribution width in patients with traumatic brain injury.

METHODS

We retrospectively analyzed intensive care unit patients with traumatic brain injury. We recorded patients' ages; genders; diagnoses; Glasgow Coma Scale scores; length of intensive care unit stay (in days); mean platelet volume, platelet distribution width, platelet count-to-total lymphocyte count ratio, and red cell distribution width values upon hospital admission; and health on the 7th and 30th days of their stays.

RESULTS

We analyzed data from 110 patients. Of these, 84 (76.4%) were male and 26 (23.6%) were female. On the 7- and 30-day mortality evaluations, compared to the living patients, the deceased patients had a significantly higher median age and a significantly lower median Glasgow Coma Scale. Thus, increased age and lower Glasgow Coma Scale scores were associated with increased 7- and 30-day mortality rates. mean platelet volume and platelet distribution width values were similar in living and deceased patients. platelet count-to-total lymphocyte count ratio values were lower in deceased patients, but this difference was not statistically significant. Within 30 days after traumatic brain injury, deceased patients' red cell distribution width values were significantly elevated in deceased patients compared to those of living patients.

CONCLUSION

Mean platelet volume, platelet distribution width, and platelet count-to-total lymphocyte count ratio values were not associated with 7- and 30-day mortality, whereas only elevated red cell distribution width was associated with 30-day mortality.

Artificial Intelligence with Statistical Confidence Scores for Detection of Acute or Subacute Hemorrhage on Noncontrast CT Head Scans

Purpose

To present a method that automatically detects, subtypes, and locates acute or subacute intracranial hemorrhage (ICH) on noncontrast CT (NCCT) head scans; generates detection confidence scores to identify high-confidence data subsets with higher accuracy; and improves radiology worklist prioritization. Such scores may enable clinicians to better use artificial intelligence (AI) tools.

Materials and Methods

This retrospective study included 46 057 studies from seven "internal" centers for development (training, architecture selection, hyperparameter tuning, and operating-point calibration; n = 25 946) and evaluation (n = 2947) and three "external" centers for calibration (n = 400) and evaluation (n = 16764). Internal centers contributed developmental data, whereas external centers did not. Deep neural networks predicted the presence of ICH and subtypes (intraparenchymal, intraventricular, subarachnoid, subdural, and/or epidural hemorrhage) and segmentations per case. Two ICH confidence scores are discussed: a calibrated classifier entropy score and a Dempster-Shafer score. Evaluation was completed by using receiver operating characteristic curve analysis and report turnaround time (RTAT) modeling on the evaluation set and on confidence score-defined subsets using bootstrapping.

Results

The areas under the receiver operating characteristic curve for ICH were 0.97 (0.97, 0.98) and 0.95 (0.94, 0.95) on internal and external center data, respectively. On 80% of the data stratified by calibrated classifier and Dempster-Shafer scores, the system improved the Youden indexes, increasing them from 0.84 to 0.93 (calibrated classifier) and from 0.84 to 0.92 (Dempster-Shafer) for internal centers and increasing them from 0.78 to 0.88 (calibrated classifier) and from 0.78 to 0.89 (Dempster-Shafer) for external centers (P < .001). Models estimated shorter RTAT for AI-prioritized worklists with confidence measures than for AI-prioritized worklists without confidence measures, shortening RTAT by 27% (calibrated classifier) and 27% (Dempster-Shafer) for internal centers and shortening RTAT by 25% (calibrated classifier) and 27% (Dempster-Shafer) for external centers (P < .001).

Conclusion

AI that provided statistical confidence measures for ICH detection on NCCT scans reliably detected and subtyped hemorrhages, identified high-confidence predictions, and improved worklist prioritization in simulation.Keywords: CT, Head/Neck, Hemorrhage, Convolutional Neural Network (CNN) Supplemental material is available for this article. © RSNA, 2022.

Risk factor guided early discharge and potential resource allocation benefits in patients with traumatic subarachnoid hemorrhage

OBJECTIVES

We sought to develop screening criteria predicting the lack of poor neurological outcomes in patients presenting with traumatic subarachnoid hemorrhage (tSAH), while evaluating their potential to improve resource-allocation in these cases.

METHODS

We retrospectively reviewed patients presenting with tSAH to the emergency department (ED) of a tertiary care institution from 2016-2018. We defined good neurological outcomes as patients with stable/improving neurological status, did not require neurosurgical intervention, no expanding bleed, and no hospital readmission. Univariate and multivariate models were generated to predict risk factors inversely associated with good neurological outcome.

RESULTS

167 patients presented with tSAH from 2016-2018. The presence of depressed skull fracture, concomitant spinal fracture, low GCS, cranial nerve palsies, disorientation, concomitant hemorrhages, midline shift (MLS), elevated INR, and emergent medical intervention were inversely correlated with likelihood of good neurological outcome upon univariate analysis. Multivariate regression demonstrated that midline shift [OR=0.22 (0.05-0.89), p=0.04], GCS <13 [OR=0.22 (0.05-0.99), p=0.05], elevated INR [OR=0.18 (0.03-0.85), p=0.04], and emergent medical intervention [OR=0.18 (0.04-0.63), p=0.01] were independently associated with lower likelihood of good neurological outcome. 46 patients without any factors had good outcomes but were held in the ED or admitted to the hospital. These patients - if instead discharged directly - translated to a potential cost savings of $179,172.

CONCLUSIONS

In our study we found multiple risk factors inversely associated with good neurological outcome, namely low GCS, midline shift, emergent medical intervention, and INR ≥ 1.4. Our findings may aid clinicians in determining which tSAH patients are candidates for safe early discharge.

Traumatic Subarachnoid Hemorrhage: A Scoping Review

Sixty-nine million people have a traumatic brain injury (TBI) each year, and TBI is the most common cause of subarachnoid hemorrhage (SAH). Traumatic SAH (TSAH) has been described as an adverse prognostic factor leading to progressive neurological deterioration and increased morbidity and mortality. A limited number of studies, however, evaluate recent trends in the diagnosis and management of SAH in the context of trauma. The objective of this scoping review was to understand the extent and type of evidence concerning the diagnostic criteria and management of TSAH. This scoping review was conducted following the Joanna Briggs Institute methodology for scoping reviews. The review included adults with SAH secondary to trauma, where isolated TSAH (iTSAH) refers to the presence of SAH in the absence of any other traumatic radiographic intracranial pathology, and TSAH refers to the presence of SAH with the possibility or presence of additional traumatic radiographic intracranial pathology. Data extracted from each study included study aim, country, methodology, population characteristics, outcome measures, a summary of findings, and future directives. Thirty studies met inclusion criteria. Studies were grouped into five categories by topic: TSAH associated with mild TBI (mTBI), n = 13), and severe TBI (n = 3); clinical management and diagnosis (n = 9); imaging (n = 3); and aneurysmal TSAH (n = 1). Of the 30 studies, two came from a low- and middle-income country (LMIC), excluding China, nearly a high-income country. Patients with TSAH associated with mTBI have a very low risk of clinical deterioration and surgical intervention and should be treated conservatively when considering intensive care unit admission. The Helsinki and Stockholm computed tomography scoring systems, in addition to the American Injury Scale, creatinine level, age decision tree, may be valuable tools to use when predicting outcome and death.

Investigating the Association between Orthopedic Fractures and Head Injury due to Road Traffic Accidents

BACKGROUND

Traumatic head injury (THI) due to road traffic accidents (RTAs) is a global health problem. Studies exploring the association between RTA-related THI and concurrent orthopedic fractures are lacking. We aim to provide a detailed analysis of this association and its impact on inhospital outcomes.

METHODS

Retrospective analysis of RTA-related THI associated with orthopedic fractures admitted to a large tertiary center, Southwest, Saudi Arabia, over ten years. Descriptive statistics for participant demographics and clinical outcomes were represented by percentages. The associations between head injury diagnosis or orthopedic fractures region and patient demographics are analyzed using the Chi-square test. Post hoc analysis for the significant Chi-square values was carried out by calculating the significant adjusted residuals. Adjust p value was obtained by using the Benjamini-Hochberg procedure to control for multiplicity testing. A p value less than 0.05 was considered statistically significant.

RESULTS

Concurrent orthopedic fractures are present in one-tenth of RTA-related THI. The cohort was dominated by young males, with 46.5% of the population between 18 and 29 years old. There was a significant association between the head injury diagnosis and the region of orthopedic fracture (p = 0.028). The type of head injury had significant associations with mortality and duration of hospital stay (p = 0.039 and p = 0.037, respectively). The region of orthopedic fracture significantly (p = 0.018) affected the duration of hospital stay, with fractures in the clavicle/shoulder region significantly (p = 0.035) having a short course of hospital admission.

CONCLUSION

Orthopedic fractures concomitant with RTA-related THI are common. The associations between the two injuries tend to happen in specific patterns. The inhospital stay duration and mortality significantly correlated with the site of the head or orthopedic injury. Knowledge of these patterns improves the care of THI victims, triaging, and resource allocations.

Predictors of Mortality in Traumatic Intracranial Hemorrhage: A National Trauma Data Bank Study

Background/Objective: Traumatic intracranial hemorrhage (tICH) accounts for significant trauma morbidity and mortality. Several studies have developed prognostic models for tICH outcomes, but previous models face limitations, including poor generalizability and limited accuracy. The objective was to develop a prognostic model and determine predictors of mortality using the largest trauma database in the U.S., applying rigorous analytical methodology with true hold-out-set model validation.

Methods: We identified 248,536 patients in the National Trauma Data Bank (NTDB) from 2012 to 2016 with a diagnosis code associated with tICH. For each admission, we collected demographic information, systolic blood pressure, blood alcohol level (BAL), Glasgow Coma Score (GCS), Injury Severity Score (ISS), presence of epidural/subdural/subarachnoid/intraparenchymal hemorrhage, comorbidities, complications, trauma center level, and trauma center region. Our final study population was 212,666 patients following exclusion of records with missing data. The dependent variable was patient death. Linear support vector machine (SVM) classification was carried out with recursive feature selection. Model performance was assessed using holdout 10-fold cross-validation.

Results: Cross-validation demonstrated a mean accuracy of 0.792 (95% CI 0.783–0.799). Accuracy, precision, recall, and AUC were 0.827, 0.309, 0.750, and 0.791, respectively. In the final model, high ISS, advanced age, subdural hemorrhage, and subarachnoid hemorrhage were associated with increased mortality, while high GCS verbal and motor subscores, current smoker, BAL beyond the legal limit, and level 1 trauma center were associated with decreased mortality.

Conclusions: A linear SVM model was developed for tICH, with nine features selected as predictors of mortality. These findings are applicable to multiple hemorrhage subtypes and may benefit the triage of high risk patients upon admission. While many studies have attempted to create models to predict mortality in TBI, we sought to confirm those predictors using modern modeling approaches, machine learning, and true hold-out test sets, using the largest available TBI database in the U.S. We find that while the predictors we identify are consistent with prior reports, overall prediction accuracy is somewhat lower than prior reports when assessed more rigorously.