Sedation Intensity in Patients with Moderate to Severe Traumatic Brain Injury in the Intensive Care Unit: A TRACK-TBI Cohort Study.

BACKGROUND: Interventions to reduce intracranial pressure (ICP) in patients with traumatic brain injury (TBI) are multimodal but variable, including sedation-dosing strategies. This article quantifies the different sedation intensities administered in patients with moderate to severe TBI (msTBI) using the therapy intensity level (TIL) across different intensive care units (ICUs), including the use of additional ICP-lowering therapies. METHODS: Within the prospective Transforming Research and Clinical Knowledge in TBI (TRACK-TBI) study, we performed a retrospective analysis of adult patients with msTBI admitted to an ICU for a least 5 days from seven US level 1 trauma centers who received invasive ICP monitoring and intravenous sedation. Sedation intensity was classified prospectively as one of three ordinal levels as part of the validated TIL score, which were collected at least once a day. RESULTS: A total of 127 patients met inclusion criteria (mean age 41.6 ± 17.7 years; 20% female). The median Injury Severity Score was 27 (interquartile range 17-33), with a median admission Glasgow Coma Score of 3 (interquartile range 3-7); 104 patients had severe TBI (82%), and 23 patients had moderate TBI (18%). The sedation intensity score was highest on the first ICU day (2.69 ± 1.78), independent of patient severity. Time to reaching each sedation intensity level varied by site. Sedation level I was reached within 24 h for all sites, but sedation levels II and III were reached variably between days 1 and 3. Sedation level III was never reached by two of seven sites. The total TIL score was highest on the first ICU day, with a modest decrease for each subsequent ICU day, but there was high site-specific practice-pattern variation. CONCLUSIONS: Intensity of sedation and other therapies for elevated ICP for patients with msTBI demonstrate large practice-pattern variation across level 1 trauma centers within the TRACK-TBI cohort study, independent of patient severity. Optimizing sedation strategies using patient-specific physiologic and pathoanatomic information may optimize patient outcomes.

Discovery of generalizable TBI phenotypes using multivariate time-series clustering.

Traumatic Brain Injury (TBI) presents a broad spectrum of clinical presentations and outcomes due to its inherent heterogeneity, leading to diverse recovery trajectories and varied therapeutic responses. While many studies have delved into TBI phenotyping for distinct patient populations, identifying TBI phenotypes that consistently generalize across various settings and populations remains a critical research gap. Our research addresses this by employing multivariate time-series clustering to unveil TBI's dynamic intricates. Utilizing a self-supervised learning-based approach to clustering multivariate time-Series data with missing values (SLAC-Time), we analyzed both the research-centric TRACK-TBI and the real-world MIMIC-IV datasets. Remarkably, the optimal hyperparameters of SLAC-Time and the ideal number of clusters remained consistent across these datasets, underscoring SLAC-Time's stability across heterogeneous datasets. Our analysis revealed three generalizable TBI phenotypes (α, ß, and γ), each exhibiting distinct non-temporal features during emergency department visits, and temporal feature profiles throughout ICU stays. Specifically, phenotype α represents mild TBI with a remarkably consistent clinical presentation. In contrast, phenotype ß signifies severe TBI with diverse clinical manifestations, and phenotype γ represents a moderate TBI profile in terms of severity and clinical diversity. Age is a significant determinant of TBI outcomes, with older cohorts recording higher mortality rates. Importantly, while certain features varied by age, the core characteristics of TBI manifestations tied to each phenotype remain consistent across diverse populations.

Clinical Impact of Standardized Interpretation and Reporting of Multimodality Neuromonitoring Data.

OBJECTIVE: Evaluate the consistency and clinical impact of standardized multimodality neuromonitoring (MNM) interpretation and reporting within a system of care for patients with severe traumatic brain injury (sTBI). DESIGN: Retrospective, observational historical case-control study. SETTING: Single-center academic level I trauma center. INTERVENTIONS: Standardized interpretation of MNM data summarized within daily reports. MEASUREMENTS MAIN RESULTS: Consecutive patients with sTBI undergoing MNM were included. Historical controls were patients monitored before implementation of standardized MNM interpretation; cases were defined as patients with available MNM interpretative reports. Patient characteristics, physiologic data, and clinical outcomes were recorded, and clinical MNM reporting elements were abstracted. The primary outcome was the Glasgow Outcome Scale score 3-6 months postinjury. One hundred twenty-nine patients were included (age 42 ± 18 yr, 82% men); 45 (35%) patients were monitored before standardized MNM interpretation and reporting, and 84 (65%) patients were monitored after that. Patients undergoing standardized interpretative reporting received fewer hyperosmotic agents (3 [1-6] vs. 6 [1-8]; p = 0.04) and spent less time above an intracranial threshold of 22 mm Hg (22% ± 26% vs. 28% ± 24%; p = 0.05). The MNM interpretation cohort had a lower proportion of anesthetic days (48% [24-70%] vs. 67% [33-91%]; p = 0.02) and higher average end-tidal carbon dioxide during monitoring (34 ± 6 mm Hg vs. 32 ± 6 mm Hg; p < 0.01; d = 0.36). After controlling for injury severity, patients undergoing standardized MNM interpretation and reporting had an odds of 1.5 (95% CI, 1.37-1.59) for better outcomes. CONCLUSIONS: Standardized interpretation and reporting of MNM data are a novel approach to provide clinical insight and to guide individualized critical care. In patients with sTBI, independent MNM interpretation and communication to bedside clinical care teams may result in improved intracranial pressure control, fewer medical interventions, and changes in ventilatory management. In this study, the implementation of a system for management, including standardized MNM interpretation, was associated with a significant improvement in outcome.

Prior traumatic brain injury is a risk factor for in-hospital mortality in moderate to severe traumatic brain injury: a TRACK-TBI cohort study.

Objectives: An estimated 14-23% of patients with traumatic brain injury (TBI) incur multiple lifetime TBIs. The relationship between prior TBI and outcomes in patients with moderate to severe TBI (msTBI) is not well delineated. We examined the associations between prior TBI, in-hospital mortality, and outcomes up to 12 months after injury in a prospective US msTBI cohort. Methods: Data from hospitalized subjects with Glasgow Coma Scale score of 3-12 were extracted from the Transforming Research and Clinical Knowledge in Traumatic Brain Injury Study (enrollment period: 2014-2019). Prior TBI with amnesia or alteration of consciousness was assessed using the Ohio State University TBI Identification Method. Competing risk regressions adjusting for age, sex, psychiatric history, cranial injury and extracranial injury severity examined the associations between prior TBI and in-hospital mortality, with hospital discharged alive as the competing risk. Adjusted HRs (aHR (95% CI)) were reported. Multivariable logistic regressions assessed the associations between prior TBI, mortality, and unfavorable outcome (Glasgow Outcome Scale-Extended score 1-3 (vs. 4-8)) at 3, 6, and 12 months after injury. Results: Of 405 acute msTBI subjects, 21.5% had prior TBI, which was associated with male sex (87.4% vs. 77.0%, p=0.037) and psychiatric history (34.5% vs. 20.7%, p=0.010). In-hospital mortality was 10.1% (prior TBI: 17.2%, no prior TBI: 8.2%, p=0.025). Competing risk regressions indicated that prior TBI was associated with likelihood of in-hospital mortality (aHR=2.06 (1.01-4.22)), but not with hospital discharged alive. Prior TBI was not associated with mortality or unfavorable outcomes at 3, 6, and 12 months. Conclusions: After acute msTBI, prior TBI history is independently associated with in-hospital mortality but not with mortality or unfavorable outcomes within 12 months after injury. This selective association underscores the importance of collecting standardized prior TBI history data early after acute hospitalization to inform risk stratification. Prospective validation studies are needed. Level of evidence: IV. Trial registration number: NCT02119182.

Deriving Automated Device Metadata From Intracranial Pressure Waveforms: A Transforming Research and Clinical Knowledge in Traumatic Brain Injury ICU Physiology Cohort Analysis.

IMPORTANCE: Treatment for intracranial pressure (ICP) has been increasingly informed by machine learning (ML)-derived ICP waveform characteristics. There are gaps, however, in understanding how ICP monitor type may bias waveform characteristics used for these predictive tools since differences between external ventricular drain (EVD) and intraparenchymal monitor (IPM)-derived waveforms have not been well accounted for. OBJECTIVES: We sought to develop a proof-of-concept ML model differentiating ICP waveforms originating from an EVD or IPM. DESIGN, SETTING, AND PARTICIPANTS: We examined raw ICP waveform data from the ICU physiology cohort within the prospective Transforming Research and Clinical Knowledge in Traumatic Brain Injury multicenter study. MAIN OUTCOMES AND MEASURES: Nested patient-wise five-fold cross-validation and group analysis with bagged decision trees (BDT) and linear discriminant analysis were used for feature selection and fair evaluation. Nine patients were kept as unseen hold-outs for further evaluation. RESULTS: ICP waveform data totaling 14,110 hours were included from 82 patients (EVD, 47; IPM, 26; both, 9). Mean age, Glasgow Coma Scale (GCS) total, and GCS motor score upon admission, as well as the presence and amount of midline shift, were similar between groups. The model mean area under the receiver operating characteristic curve (AU-ROC) exceeded 0.874 across all folds. In additional rigorous cluster-based subgroup analysis, targeted at testing the resilience of models to cross-validation with smaller subsets constructed to develop models in one confounder set and test them in another subset, AU-ROC exceeded 0.811. In a similar analysis using propensity score-based rather than cluster-based subgroup analysis, the mean AU-ROC exceeded 0.827. Of 842 extracted ICP features, 62 were invariant within every analysis, representing the most accurate and robust differences between ICP monitor types. For the nine patient hold-outs, an AU-ROC of 0.826 was obtained using BDT. CONCLUSIONS AND RELEVANCE: The developed proof-of-concept ML model identified differences in EVD- and IPM-derived ICP signals, which can provide missing contextual data for large-scale retrospective datasets, prevent bias in computational models that ingest ICP data indiscriminately, and control for confounding using our model's output as a propensity score by to adjust for the monitoring method that was clinically indicated. Furthermore, the invariant features may be leveraged as ICP features for anomaly detection.

Post-Intensive Care Syndrome in Neurocritical Care Patients.

Post-intensive care syndrome (PICS) refers to unintended consequences of critical care that manifest as new or worsening impairments in physical functioning, cognitive ability, or mental health. As intensive care unit (ICU) survival continues to improve, PICS is becoming increasingly recognized as a public health problem. Studies that focus on PICS have typically excluded patients with acute brain injuries and chronic neurodegenerative problems. However, patients who require neurocritical care undoubtedly suffer from impairments that overlap substantially with those encompassed by PICS. A major challenge is to distinguish between impairments related to brain injury and those that occur as a consequence of critical care. The general principles for the prevention and management of PICS and multidomain impairments in patients with moderate and severe neurological injuries are similar including the ICU liberation bundle, multidisciplinary team-based care throughout the continuum of care, and increasing awareness regarding the challenges of critical care survivorship among patients, families, and multidisciplinary team members. An extension of this concept, PICS-Family (PICS-F) refers to the mental health consequences of the intensive care experience for families and loved ones of ICU survivors. A dyadic approach to ICU survivorship with an emphasis on recognizing families and caregivers that may be at risk of developing PICS-F after neurocritical care illness can help improve outcomes for ICU survivors. In this review, we will summarize our current understanding of PICS and PICS-F, emerging literature on PICS in severe acute brain injury, strategies for preventing and treating PICS, and share our recommendations for future directions.

A Neurosurgical Readmissions Reduction Program in an Academic Hospital Leveraging Machine Learning, Workflow Analysis, and Simulation.

BACKGROUND: Predicting 30-day hospital readmissions is crucial for improving patient outcomes, optimizing resource allocation, and achieving financial savings. Existing studies reporting the development of machine learning (ML) models predictive of neurosurgical readmissions do not report factors related to clinical implementation. OBJECTIVES: Train individual predictive models with good performance (area under the receiver operating characteristic curve or AUROC > 0.8), identify potential interventions through semi-structured interviews, and demonstrate estimated clinical and financial impact of these models. METHODS: Electronic health records were utilized with five ML methodologies: gradient boosting, decision tree, random forest, ridge logistic regression, and linear support vector machine. Variables of interest were determined by domain experts and literature. The dataset was split divided 80% for training and validation and 20% for testing randomly. Clinical workflow analysis was conducted using semi-structured interviews to identify possible intervention points. Calibrated agent-based models (ABMs), based on a previous study with interventions, were applied to simulate reductions of the 30-day readmission rate and financial costs. RESULTS: The dataset covered 12,334 neurosurgical intensive care unit (NSICU) admissions (11,029 patients); 1,903 spine surgery admissions (1,641 patients), and 2,208 traumatic brain injury (TBI) admissions (2,185 patients), with readmission rate of 13.13, 13.93, and 23.73%, respectively. The random forest model for NSICU achieved best performance with an AUROC score of 0.89, capturing potential patients effectively. Six interventions were identified through 12 semi-structured interviews targeting preoperative, inpatient stay, discharge phases, and follow-up phases. Calibrated ABMs simulated median readmission reduction rates and resulted in 13.13 to 10.12% (NSICU), 13.90 to 10.98% (spine surgery), and 23.64 to 21.20% (TBI). Approximately $1,300,614.28 in saving resulted from potential interventions. CONCLUSION: This study reports the successful development and simulation of an ML-based approach for predicting and reducing 30-day hospital readmissions in neurosurgery. The intervention shows feasibility in improving patient outcomes and reducing financial losses.

Trends in management of patients with new-onset refractory status epilepticus (NORSE) from 2016 to 2023: An interim analysis.

In response to the evolving treatment landscape for new-onset refractory status epilepticus (NORSE) and the publication of consensus recommendations in 2022, we conducted a comparative analysis of NORSE management over time. Seventy-seven patients were enrolled by 32 centers, from July 2016 to August 2023, in the NORSE/FIRES biorepository at Yale. Immunotherapy was administered to 88% of patients after a median of 3 days, with 52% receiving second-line immunotherapy after a median of 12 days (anakinra 29%, rituximab 25%, and tocilizumab 19%). There was an increase in the use of second-line immunotherapies (odds ratio [OR] = 1.4, 95% CI = 1.1-1.8) and ketogenic diet (OR = 1.8, 95% CI = 1.3-2.6) over time. Specifically, patients from 2022 to 2023 more frequently received second-line immunotherapy (69% vs 40%; OR = 3.3; 95% CI = 1.3-8.9)-particularly anakinra (50% vs 13%; OR = 6.5; 95% CI = 2.3-21.0), and the ketogenic diet (OR = 6.8; 95% CI = 2.5-20.1)-than those before 2022. Among the 27 patients who received anakinra and/or tocilizumab, earlier administration after status epilepticus onset correlated with a shorter duration of status epilepticus (ρ = .519, p = .005). Our findings indicate an evolution in NORSE management, emphasizing the increasing use of second-line immunotherapies and the ketogenic diet. Future research will clarify the impact of these treatments and their timing on patient outcomes.

Recovery Potential in Patients Who Died After Withdrawal of Life-Sustaining Treatment: A TRACK-TBI Propensity Score Analysis.

Among patients with severe traumatic brain injury (TBI), there is high prognostic uncertainty but growing evidence that recovery of independence is possible. Nevertheless, families are often asked to make decisions about withdrawal of life-sustaining treatment (WLST) within days of injury. The range of potential outcomes for patients who died after WLST (WLST+) is unknown, posing a challenge for prognostic modeling and clinical counseling. We investigated the potential for survival and recovery of independence after acute TBI in patients who died after WLST. We used Transforming Research and Clinical Knowledge in TBI (TRACK-TBI) data and propensity score matching to pair participants with WLST+ to those with a similar probability of WLST (based on demographic and clinical characteristics), but for whom life-sustaining treatment was not withdrawn (WLST-). To optimize matching, we divided the WLST- cohort into tiers (Tier 1 = 0-11%, Tier 2 = 11-27%, Tier 3 = 27-70% WLST propensity). We estimated the level of recovery that could be expected in WLST+ participants by evaluating 3-, 6-, and 12-month Glasgow Outcome Scale-Extended (GOSE) and Disability Rating Scale outcomes in matched WLST- participants. Of 90 WLST+ participants (80% male, mean [standard deviation; SD] age = 59.2 [17.9] years, median [IQR] days to WLST = 5.4 [2.2, 11.7]), 80 could be matched to WLST- participants. Of 56 WLST- participants who were followed at 6 months, 31 (55%) died. Among survivors in the overall sample and survivors in Tiers 1 and 2, more than 30% recovered at least partial independence (GOSE ≥4). In Tier 3, recovery to GOSE ≥4 occurred at 12 months, but not 6 months, post-injury. These results suggest a substantial proportion of patients with TBI and WLST may have survived and achieved at least partial independence. However, death or severe disability is a common outcome when the probability of WLST is high. While further validation is needed, our findings support a more cautious clinical approach to WLST and more complete reporting on WLST in TBI studies.

Comparative analysis of patients with new onset refractory status epilepticus preceded by fever (febrile infection-related epilepsy syndrome) versus without prior fever: An interim analysis.

Febrile infection-related epilepsy syndrome (FIRES) is a subset of new onset refractory status epilepticus (NORSE) that involves a febrile infection prior to the onset of the refractory status epilepticus. It is unclear whether FIRES and non-FIRES NORSE are distinct conditions. Here, we compare 34 patients with FIRES to 30 patients with non-FIRES NORSE for demographics, clinical features, neuroimaging, and outcomes. Because patients with FIRES were younger than patients with non-FIRES NORSE (median = 28 vs. 48 years old, p = .048) and more likely cryptogenic (odds ratio = 6.89), we next ran a regression analysis using age or etiology as a covariate. Respiratory and gastrointestinal prodromes occurred more frequently in FIRES patients, but no difference was found for non-infection-related prodromes. Status epilepticus subtype, cerebrospinal fluid (CSF) and magnetic resonance imaging findings, and outcomes were similar. However, FIRES cases were more frequently cryptogenic; had higher CSF interleukin 6, CSF macrophage inflammatory protein-1 alpha (MIP-1a), and serum chemokine ligand 2 (CCL2) levels; and received more antiseizure medications and immunotherapy. After controlling for age or etiology, no differences were observed in presenting symptoms and signs or inflammatory biomarkers, suggesting that FIRES and non-FIRES NORSE are very similar conditions.

Discovery of Generalizable TBI Phenotypes Using Multivariate Time-Series Clustering.

Traumatic Brain Injury (TBI) presents a broad spectrum of clinical presentations and outcomes due to its inherent heterogeneity, leading to diverse recovery trajectories and varied therapeutic responses. While many studies have delved into TBI phenotyping for distinct patient populations, identifying TBI phenotypes that consistently generalize across various settings and populations remains a critical research gap. Our research addresses this by employing multivariate time-series clustering to unveil TBI's dynamic intricates. Utilizing a self-supervised learning-based approach to clustering multivariate time-Series data with missing values (SLAC-Time), we analyzed both the research-centric TRACK-TBI and the real-world MIMIC-IV datasets. Remarkably, the optimal hyperparameters of SLAC-Time and the ideal number of clusters remained consistent across these datasets, underscoring SLAC-Time's stability across heterogeneous datasets. Our analysis revealed three generalizable TBI phenotypes (α, ß, and γ), each exhibiting distinct non-temporal features during emergency department visits, and temporal feature profiles throughout ICU stays. Specifically, phenotype α represents mild TBI with a remarkably consistent clinical presentation. In contrast, phenotype ß signifies severe TBI with diverse clinical manifestations, and phenotype γ represents a moderate TBI profile in terms of severity and clinical diversity. Age is a significant determinant of TBI outcomes, with older cohorts recording higher mortality rates. Importantly, while certain features varied by age, the core characteristics of TBI manifestations tied to each phenotype remain consistent across diverse populations.

Intracranial pressure-flow relationships in traumatic brain injury patients expose gaps in the tenets of models and pressure-oriented management.

Background: The protocols and therapeutic guidance established for treating traumatic brain injuries (TBI) in neurointensive care focus on managing cerebral blood flow (CBF) and brain tissue oxygenation based on pressure signals. The decision support process relies on assumed relationships between cerebral perfusion pressure (CPP) and blood flow, pressure-flow relationships (PFRs), and shares this framework of assumptions with mathematical intracranial hemodynamic models. These foundational assumptions are difficult to verify, and their violation can impact clinical decision-making and model validity. Method: A hypothesis- and model-driven method for verifying and understanding the foundational intracranial hemodynamic PFRs is developed and applied to a novel multi-modality monitoring dataset. Results: Model analysis of joint observations of CPP and CBF validates the standard PFR when autoregulatory processes are impaired as well as unmodelable cases dominated by autoregulation. However, it also identifies a dynamical regime -or behavior pattern- where the PFR assumptions are wrong in a precise, data-inferable way due to negative CPP-CBF coordination over long timescales. This regime is of both clinical and research interest: its dynamics are modelable under modified assumptions while its causal direction and mechanistic pathway remain unclear. Conclusions: Motivated by the understanding of mathematical physiology, the validity of the standard PFR can be assessed a) directly by analyzing pressure reactivity and mean flow indices (PRx and Mx) or b) indirectly through the relationship between CBF and other clinical observables. This approach could potentially help personalize TBI care by considering intracranial pressure and CPP in relation to other data, particularly CBF. The analysis suggests a threshold using clinical indices of autoregulation jointly generalizes independently set indicators to assess CA functionality. These results support the use of increasingly data-rich environments to develop more robust hybrid physiological-machine learning models.

Pharmacokinetics of Levetiracetam Seizure Prophylaxis in Severe Traumatic Brain Injury.

BACKGROUND: Drug pharmacokinetics (PK) are altered in neurocritically ill patients, and optimal levetiracetam dosing for seizure prophylaxis is unknown. OBJECTIVE: This study evaluates levetiracetam PK in critically ill patients with severe traumatic brain injury (sTBI) receiving intravenous levetiracetam 1000 mg every 8 (LEV8) to 12 (LEV12) hours for seizure prophylaxis. METHODS: This prospective, open-label study was conducted at a level 1 trauma, academic, quaternary care center. Patients with sTBI receiving seizure prophylaxis with LEV8 or LEV12 were eligible for enrollment. Five sequential, steady-state, postdose serum levetiracetam concentrations were obtained. Non-compartmental analysis (NCA) and compartmental approaches were employed for estimating pharmacokinetic parameters and projecting steady-state trough concentrations. Pharmacokinetic parameters were compared between LEV8 and LEV12 patients. Monte Carlo simulations (MCS) were performed to determine probability of target trough attainment (PTA) of 6 to 20 mg/L. A secondary analysis evaluated PTA for weight-tiered levetiracetam dosing. RESULTS: Ten male patients (5 LEV8; 5 LEV12) were included. The NCA-based systemic clearance and elimination half-life were 5.3 ± 1.2 L/h and 4.8 ± 0.64 hours. A one-compartment model provided a higher steady-state trough concentration for the LEV8 group compared with the LEV12 group (13.7 ± 4.3 mg/L vs 6.3 ± 1.7 mg/L; P = 0.008). Monte Carlo simulations predicted regimens of 500 mg every 6 hours, 1000 mg every 8 hours, and 2000 mg every 12 hours achieved therapeutic target attainment. Weight-tiered dosing regimens achieved therapeutic target attainment using a 75 kg breakpoint. CONCLUSION AND RELEVANCE: Neurocritically ill patients exhibit rapid levetiracetam clearance resulting in a short elimination half-life. Findings of this study suggest regimens of levetiracetam 500 mg every 6 hours, 1000 mg every 8 hours, or 2000 mg every 12 hours may be required for optimal therapeutic target attainment. Patient weight of 75 kg may serve as a breakpoint for weight-guided dosing to optimize levetiracetam therapeutic target attainment for seizure prophylaxis.

Common Data Elements for Disorders of Consciousness: Recommendations from the Working Group on Physiology and Big Data.

BACKGROUND: The implementation of multimodality monitoring in the clinical management of patients with disorders of consciousness (DoC) results in physiological measurements that can be collected in a continuous and regular fashion or even at waveform resolution. Such data are considered part of the "Big Data" available in intensive care units and are potentially suitable for health care-focused artificial intelligence research. Despite the richness in content of the physiological measurements, and the clinical implications shown by derived metrics based on those measurements, they have been largely neglected from previous attempts in harmonizing data collection and standardizing reporting of results as part of common data elements (CDEs) efforts. CDEs aim to provide a framework for unifying data in clinical research and help in implementing a systematic approach that can facilitate reliable comparison of results from clinical studies in DoC as well in international research collaborations. METHODS: To address this need, the Neurocritical Care Society's Curing Coma Campaign convened a multidisciplinary panel of DoC "Physiology and Big Data" experts to propose CDEs for data collection and reporting in this field. RESULTS: We report the recommendations of this CDE development panel and disseminate CDEs to be used in physiologic and big data studies of patients with DoC. CONCLUSIONS: These CDEs will support progress in the field of DoC physiologic and big data and facilitate international collaboration.

Common Data Elements for Disorders of Consciousness: Recommendations from the Working Group on Hospital Course, Confounders, and Medications.

The convergence of an interdisciplinary team of neurocritical care specialists to organize the Curing Coma Campaign is the first effort of its kind to coordinate national and international research efforts aimed at a deeper understanding of disorders of consciousness (DoC). This process of understanding includes translational research from bench to bedside, descriptions of systems of care delivery, diagnosis, treatment, rehabilitation, and ethical frameworks. The description and measurement of varying confounding factors related to hospital care was thought to be critical in furthering meaningful research in patients with DoC. Interdisciplinary hospital care is inherently varied across geographical areas as well as community and academic medical centers. Access to monitoring technologies, specialist consultation (medical, nursing, pharmacy, respiratory, and rehabilitation), staffing resources, specialty intensive and acute care units, specialty medications and specific surgical, diagnostic and interventional procedures, and imaging is variable, and the impact on patient outcome in terms of DoC is largely unknown. The heterogeneity of causes in DoC is the source of some expected variability in care and treatment of patients, which necessitated the development of a common nomenclature and set of data elements for meaningful measurement across studies. Guideline adherence in hemorrhagic stroke and severe traumatic brain injury may also be variable due to moderate or low levels of evidence for many recommendations. This article outlines the process of the development of common data elements for hospital course, confounders, and medications to streamline definitions and variables to collect for clinical studies of DoC.

Practice Standards for the Use of Multimodality Neuromonitoring: A Delphi Consensus Process.

OBJECTIVES: To address areas in which there is no consensus for the technologies, effort, and training necessary to integrate and interpret information from multimodality neuromonitoring (MNM). DESIGN: A three-round Delphi consensus process. SETTING: Electronic surveys and virtual meeting. SUBJECTS: Participants with broad MNM expertise from adult and pediatric intensive care backgrounds. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Two rounds of surveys were completed followed by a virtual meeting to resolve areas without consensus and a final survey to conclude the Delphi process. With 35 participants consensus was achieved on 49% statements concerning MNM. Neurologic impairment and the potential for MNM to guide management were important clinical considerations. Experts reached consensus for the use of MNM-both invasive and noninvasive-for patients in coma with traumatic brain injury, aneurysmal subarachnoid hemorrhage, and intracranial hemorrhage. There was consensus that effort to integrate and interpret MNM requires time independent of daily clinical duties, along with specific skills and expertise. Consensus was reached that training and educational platforms are necessary to develop this expertise and to provide clinical correlation. CONCLUSIONS: We provide expert consensus in the clinical considerations, minimum necessary technologies, implementation, and training/education to provide practice standards for the use of MNM to individualize clinical care.

Management Strategies Based on Multi-Modality Neuromonitoring in Severe Traumatic Brain Injury.

Secondary brain injury after neurotrauma is comprised of a host of distinct, potentially concurrent and interacting mechanisms that may exacerbate primary brain insult. Multimodality neuromonitoring is a method of measuring multiple aspects of the brain in order to understand the signatures of these different pathomechanisms and to detect, treat, or prevent potentially reversible secondary brain injuries. The most studied invasive parameters include intracranial pressure (ICP), cerebral perfusion pressure (CPP), autoregulatory indices, brain tissue partial oxygen tension, and tissue energy and metabolism measures such as the lactate pyruvate ratio. Understanding the local metabolic state of brain tissue in order to infer pathology and develop appropriate management strategies is an area of active investigation. Several clinical trials are underway to define the role of brain tissue oxygenation monitoring and electrocorticography in conjunction with other multimodal neuromonitoring information, including ICP and CPP monitoring. Identifying an optimal CPP to guide individualized management of blood pressure and ICP has been shown to be feasible, but definitive clinical trial evidence is still needed. Future work is still needed to define and clinically correlate patterns that emerge from integrated measurements of metabolism, pressure, flow, oxygenation, and electrophysiology. Pathophysiologic targets and precise critical care management strategies to address their underlying causes promise to mitigate secondary injuries and hold the potential to improve patient outcome. Advancements in clinical trial design are poised to establish new standards for the use of multimodality neuromonitoring to guide individualized clinical care.

WindowSHAP: An efficient framework for explaining time-series classifiers based on Shapley values.

Unpacking and comprehending how black-box machine learning algorithms (such as deep learning models) make decisions has been a persistent challenge for researchers and end-users. Explaining time-series predictive models is useful for clinical applications with high stakes to understand the behavior of prediction models, e.g., to determine how different variables and time points influence the clinical outcome. However, existing approaches to explain such models are frequently unique to architectures and data where the features do not have a time-varying component. In this paper, we introduce WindowSHAP, a model-agnostic framework for explaining time-series classifiers using Shapley values. We intend for WindowSHAP to mitigate the computational complexity of calculating Shapley values for long time-series data as well as improve the quality of explanations. WindowSHAP is based on partitioning a sequence into time windows. Under this framework, we present three distinct algorithms of Stationary, Sliding and Dynamic WindowSHAP, each evaluated against baseline approaches, KernelSHAP and TimeSHAP, using perturbation and sequence analyses metrics. We applied our framework to clinical time-series data from both a specialized clinical domain (Traumatic Brain Injury - TBI) as well as a broad clinical domain (critical care medicine). The experimental results demonstrate that, based on the two quantitative metrics, our framework is superior at explaining clinical time-series classifiers, while also reducing the complexity of computations. We show that for time-series data with 120 time steps (hours), merging 10 adjacent time points can reduce the CPU time of WindowSHAP by 80 % compared to KernelSHAP. We also show that our Dynamic WindowSHAP algorithm focuses more on the most important time steps and provides more understandable explanations. As a result, WindowSHAP not only accelerates the calculation of Shapley values for time-series data, but also delivers more understandable explanations with higher quality.

Integrating, Harmonizing, and Curating Studies With High-Frequency and Hourly Physiological Data: Proof of Concept from Seven Traumatic Brain Injury Data Sets.

Research in severe traumatic brain injury (TBI) has historically been limited by studies with relatively small sample sizes that result in low power to detect small, yet clinically meaningful outcomes. Data sharing and integration from existing sources hold promise to yield larger more robust sample sizes that improve the potential signal and generalizability of important research questions. However, curation and harmonization of data of different types and of disparate provenance is challenging. We report our approach and experience integrating multiple TBI data sets containing collected physiological data, including both expected and unexpected challenges encountered in the integration process. Our harmonized data set included data on 1536 patients from the Citicoline Brain Injury Treatment Trial (COBRIT), Effect of erythropoietin and transfusion threshold on neurological recovery after traumatic brain injury: a randomized clinical trial (EPO Severe TBI), BEST-TRIP, Progesterone for the Treatment of Traumatic Brain Injury III Clinical Trial (ProTECT III), Transforming Research and Clinical Knowledge in Traumatic brain Injury (TRACK-TBI), Brain Oxygen Optimization in Severe Traumatic Brain Injury Phase-II (BOOST-2), and Ben Taub General Hospital (BTGH) Research Database studies. We conclude with process recommendations for data acquisition for future prospective studies to aid integration of these data with existing studies. These recommendations include using common data elements whenever possible, a standardized recording system for labeling and timing of high-frequency physiological data, and secondary use of studies in systems such as Federal Interagency Traumatic Brain Injury Research Informatics System (FITBIR), to engage investigators who collected the original data.