Intracranial hypertension (IH) is a key driver of secondary brain injury in patients with traumatic brain injury. Lowering intracranial pressure (ICP) as soon as IH occurs is important, but a preemptive approach would be more beneficial. We systematically reviewed the artificial intelligence (AI) models, variables, performances, risks of bias, and clinical machine learning (ML) readiness levels of IH prediction models using AI. We conducted a systematic search until 12-03-2023 in three databases. Only studies predicting IH or ICP in patients with traumatic brain injury with a validation of the AI model were included. We extracted type of AI model, prediction variables, model performance, validation type, and prediction window length. Risk of bias was assessed with the Prediction Model Risk of Bias Assessment Tool, and we determined the clinical ML readiness level. Eleven out of 399 nonduplicate publications were included. A gaussian processes model using ICP and mean arterial pressure was most common. The maximum reported area under the receiver operating characteristic curve was 0.94. Four studies conducted external validation, and one study a prospective clinical validation. The prediction window length preceding IH varied between 30 and 60 min. Most studies (73%) had high risk of bias. The highest clinical ML readiness level was 6 of 9, indicating "real-time model testing" stage in one study. Several IH prediction models using AI performed well, were externally validated, and appeared ready to be tested in the clinical workflow (clinical ML readiness level 5 of 9). A Gaussian processes model was most used, and ICP and mean arterial pressure were frequently used variables. However, most studies showed a high risk of bias. Our findings may help position AI for IH prediction on the path to ultimate clinical integration and thereby guide researchers plan and design future studies.

Background: Suicidality is common among patients with traumatic brain injury (TBI). This meta-analysis aimed to assess the prevalence of suicidal ideation (SI) and suicide attempt (SA) in TBI patients. Methods: Databases including PubMed, Cochrane Library, Embase, CINAHL, Web of Science, PsycINFO, China National Knowledge Infrastructure (CNKI), VIP, WanFang Data, and Chinese Biomedical Literature Database (CBM) were searched for relevant literature from inception to 9th January 2021. Random effects models were used to estimate the prevalence of SI and SA. The methodological quality of the included studies was assessed using the modified Newcastle-Ottawa Scale. Subgroup analysis was performed based on the geographical location of the institutions. Results: A total of 16 studies involving 1,146,271 patients with TBI were included in our meta-analysis. The prevalence of SI and SA were 19.1% (95% CI 13.6-24.6%) and 2.1% (95% CI 1.8-2.4%), respectively, while the prevalence of SA in Asian and non-Asian countries were 2.0% (95% CI 0.3-3.7%) and 2.0% (95% CI 1.6-2.3%). Meta-regression analyses revealed that the publication year and age were positively associated, while the geographical location of the institution and sample size were negatively associated with the pooled prevalence of SI. On the other hand, the geographical location of the institution, sample size and age were negatively associated with the pooled prevalence of SA. Limitations: The overall heterogeneity between studies was high. Conclusion: SI and SA are common among TBI patients. Therefore, targeted preventive measures are paramount to manage TBI-related suicide.

Objective: To document the characteristics, measured outcomes, and effectiveness of physical activity (PA) interventions designed to improve health-related outcomes in individuals with a mild traumatic brain injury (mTBI) to assist in rehabilitation quality improvement efforts of a TBI rehabilitation program. Methods: A scoping review following a 6-step iterative framework search across 5 databases (MEDLINE, CINAHL, PsycINFO, SPORTDiscuss, and EMBASE) and the gray literature (Google) was performed. Selected PA interventions were designed for individuals of all ages and any mechanism of injury (eg, sports-related and falls). Data were charted, collated, and summarized according to the Consensus on Exercise Reporting Template checklist and domains of the International Classification of Functioning, Disability and Health. Involvement of clinical experts ensured tailoring of the knowledge synthesis to meet clinical needs. Results: Thirty-five articles and 14 gray literature records were retained. Five types of PA interventions were identified with the majority being multimodal. Reporting of PA intervention characteristics was highly variable across studies; many details necessary for intervention replication are missing. Study outcomes focused primarily on improving body functions and symptoms of mTBI, and less frequently on activities, participation, and health-related quality of life. The methodological quality of studies varies. Conclusions: Identified PA intervention types offer various management options for healthcare providers. PA interventions may improve a wide range of health-related outcomes supporting the inclusion of PA in the management of individuals of all ages with mTBI. Higher-quality research and better reporting about intervention characteristics is however needed.

Objective: To undertake a systematic review of the evidence for the effect of acetylcholinesterase inhibitors (AChEIs) on cognition late after moderate or severe traumatic brain injury. Background: Cognitive impairment after traumatic brain injury has significant consequences for the individual and society. Cholinergic pathways play an important role in cognitive processing and a hypocholinergic state exists in the chronic phases after traumatic brain injury. AChEIs are already used off-label to treat patients with traumatic brain injury. Data sources and study selection: PubMed, CINAHL, Psyc-INFO, the Cochrane Collaboration and Web of Science were searched with pre-specified criteria between 1999 and June 2015. Data extraction and synthesis: A total of 153 studies were identified. None met pre-specified criteria. The criteria were revised in order to identify studies that may provide useful information despite some risk of bias. Three studies met the revised criteria and were evaluated by 2 reviewers using the Swedish Council on Health Technology Assessment grading system, based on GRADE. Key findings and limitations were tabulated. One study found no effect and 2 found limited effect. Conclusion: Large randomized controlled studies are needed to establish whether AChEIs are effective for cognition late after moderate or severe traumatic brain injury. Clinicians should be aware of the weak evidence base when considering the off-label use of AChEIs.

OBJECTIVE: To identify literature which discusses the barriers and enablers of eHealth technology and which evaluates its role in facilitating interdisciplinary team work for the care of people with a traumatic brain injury (TBI). DESIGN: Systematic review. DATA SOURCES: Studies were identified by searching CINAHL, Embase, Medline, PsycINFO, Scopus, and Web of Science. STUDY SELECTION: Studies included in the review were required to feature an eHealth intervention which assisted interdisciplinary care for people with TBI. DATA EXTRACTION: Descriptive data for each study described the eHealth intervention, interdisciplinary team, outcomes, and barriers and facilitators in implementing eHealth interventions. RESULTS: The search resulted in 1389 publications, of which 35 were retrieved and scanned in full. Six studies met all the inclusion criteria for the review. Four different eHealth interventions were identified: (i) an electronic goals systems, (ii) telerehabilitation, (iii) videoconferencing, and (iv) a point-of-care team-based information system. Various barriers and facilitators were identified in the use of eHealth. CONCLUSION: eHealth interventions have been reported to support interdisciplinary teams for the care of TBI. However, there is a substantial gap in existing literature regarding the barriers and enablers which characterize a successful interdisciplinary eHealth model for people with TBI.