Improving Prediction of Favourable Outcome After 6 Months in Patients with Severe Traumatic Brain Injury Using Physiological Cerebral Parameters in a Multivariable Logistic Regression Model

Multivariate modeling and prediction of cerebral physiology in acute traumatic neural injury: A scoping review

Traumatic brain injury (TBI) poses a significant global public health challenge necessitating a profound understanding of cerebral physiology. The dynamic nature of TBI demands sophisticated methodologies for modeling and predicting cerebral signals to unravel intricate pathophysiology and predict secondary injury mechanisms prior to their occurrence. In this comprehensive scoping review, we focus specifically on multivariate cerebral physiologic signal analysis in the context of multi-modal monitoring (MMM) in TBI, exploring a range of techniques including multivariate statistical time-series models and machine learning algorithms. Conducting a comprehensive search across databases yielded 7 studies for evaluation, encompassing diverse cerebral physiologic signals and parameters from TBI patients. Among these, five studies concentrated on modeling cerebral physiologic signals using statistical time-series models, while the remaining two studies primarily delved into intracranial pressure (ICP) prediction through machine learning models. Autoregressive models were predominantly utilized in the modeling studies. In the context of prediction studies, logistic regression and Gaussian processes (GP) emerged as the predominant choice in both research endeavors, with their performance being evaluated against each other in one study and other models such as random forest, and decision tree in the other study. Notably among these models, random forest model, an ensemble learning approach, demonstrated superior performance across various metrics. Additionally, a notable gap was identified concerning the absence of studies focusing on prediction for multivariate outcomes. This review addresses existing knowledge gaps and sets the stage for future research in advancing cerebral physiologic signal analysis for neurocritical care improvement.

Intracranial Pressure-Derived Cerebrovascular Reactivity Indices and Their Critical Thresholds: A Canadian High Resolution-Traumatic Brain Injury Validation Study

Current neurointensive care guidelines recommend intracranial pressure (ICP) and cerebral perfusion pressure (CPP) centered management for moderate-severe traumatic brain injury (TBI) because of their demonstrated associations with patient outcome. Cerebrovascular reactivity metrics, such as the pressure reactivity index (PRx), pulse amplitude index (PAx), and RAC index, have also demonstrated significant prognostic capabilities with regard to outcome. However, critical thresholds for cerebrovascular reactivity indices have only been identified in two studies conducted at the same center. In this study, we aim to determine the critical thresholds of these metrics by leveraging a unique multi-center database. The study included a total of 354 patients from the CAnadian High-Resolution TBI (CAHR-TBI) Research Collaborative. Based on 6-month Glasgow Outcome Scores, patients were dichotomized into alive versus dead and favorable versus unfavorable. Chi-square values were then computed for incrementally increasing values of each physiological parameter of interest against outcome. The values that generated the greatest chi-squares for each parameter were considered to be the thresholds with the greatest outcome discriminatory capacity. To confirm that the identified thresholds provide prognostic utility, univariate and multivariable logistical regression analyses were performed adjusting for the International Mission for Prognosis and Analysis of Clinical Trials (IMPACT) variables. Through the chi-square analysis, a lower limit CPP threshold of 60 mm Hg and ICP thresholds of 18 mm Hg and 22 mm Hg were identified for both survival and favorable outcome predictions. For the cerebrovascular reactivity metrics, different thresholds were identified for the two outcome dichotomizations. For survival prediction, thresholds of 0.35, 0.25, and 0 were identified for PRx, PAx, and RAC, respectively. For favorable outcome prediction, thresholds of 0.325, 0.20, and 0.05 were found. Univariate logistical regression analysis demonstrated that the time spent above/below thresholds were associated with outcome. Further, multivariable logistical regression analysis found that percent time above/below the identified thresholds added additional variance to the IMPACT core model for predicting both survival and favorable outcome. In this study, we were able to validate the results of the previous two works as well as to reaffirm the ICP and CPP guidelines from the Brain Trauma Foundation (BTF) and the Seattle International Severe Traumatic Brain Injury Consensus Conference (SIBICC).

The added value of temporal data and the best way to handle it: A use-case for atrial fibrillation using general practitioner data

INTRODUCTION

Temporal data has numerous challenges for deep learning such as irregularity of sampling. New algorithms are being developed that can handle these temporal challenges better. However, it is unclear how the performance ranges from classical non-temporal models to newly developed algorithms. Therefore, this study compares different non-temporal and temporal algorithms for a relevant use case, the prediction of atrial fibrillation (AF) using general practitioner (GP) data.

METHODS

Three datasets with a 365-day observation window and prediction windows of 14, 180 and 360 days were used. Data consisted of medication, lab, symptom, and chronic diseases codings registered by the GP. The benchmark discarded temporality and used logistic regression, XGBoost models and neural networks on the presence of codings over the whole year. Pattern data extracted common patterns of GP codings and tested using the same algorithms. LSTM and CKConv models were trained as models incorporating temporality.

RESULTS

Algorithms which incorporated temporality (LSTM and CKConv, (max AUC 0.734 at 360 days prediction window) outperformed both benchmark and pattern algorithms (max AUC 0.723, with a significant improvement using the 360 days prediction window (p = 0.04). The difference between the benchmark and the LSTM or CKConv algorithm decreased with smaller prediction windows, indicating temporal importance for longer prediction windows. The CKConv and LSTM algorithm performed similarly, possibly due to limited sequence length.

CONCLUSION

Temporal models outperformed non-temporal models for the prediction of AF. For temporal models, CKConv is a promising algorithm to handle temporal data using GP data as it can handle irregular data.

Critical ICP thresholds in relation to outcome: Is 22 mmHg really the answer?

PURPOSE

Intensive care for patients with traumatic brain injury (TBI) aims, among other tasks, at avoiding high intracranial pressure (ICP), which is perceived to worsen motor and cognitive deficits and increase mortality. International recommendations for threshold values for ICP were increased from 20 to 22 mmHg in 2016 following the findings in a study by Sorrentino et al., which were based on an observational study of patients with TBI of averaged ICP values. We aimed to reproduce their approach and validate the findings in a separate cohort.

METHODS

Three hundred thirty-one patients with TBI were included and categorised according to survival/death and favourable/unfavourable outcome at 6 months (based on Glasgow Outcome Score-Extended of 6-8 and 1-5, respectively). Repeated chi-square tests of survival and death (or favourable and unfavourable outcome) vs. high and low ICP were conducted with discrimination between high and low ICP sets at increasing values (integers) between 10 and 35 mmHg, using the average ICP for the entire monitoring period. The ICP limit returning the highest chi-square score was assumed to be the threshold with best discriminative ability. This approach was repeated after stratification by sex, age, and initial Glasgow Coma Score (GCS).

RESULTS

An ICP limit of 18 mmHg was found for both mortality and unfavourable outcome for the entire cohort. The female and the low GCS subgroups both had threshold values of 18 mmHg; for all other subgroups, the threshold varied between 16 and 30 mmHg. According to a multiple logistic regression analysis, age, initial GCS, and average ICP are independently associated with mortality and outcome.

CONCLUSIONS

Using identical methods and closely comparable cohorts, the critical thresholds for ICP found in the study by Sorrentino et al. could not be reproduced.

Predictors of outcomes 3 to 12 months after traumatic brain injury: a systematic review and meta-analysis

The exact factors predicting outcomes following traumatic brain injury (TBI) remain elusive. In this systematic review and meta-analysis, we examined factors influencing outcomes in adult patients with TBI, from 3 months to 1 year after injury. A search of four electronic databases-PubMed, Scopus, Web of Science, and ScienceDirect-yielded 29 studies for review and 16 for meta-analysis, in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. In patients with TBI of any severity, mean differences were observed in age (8.72 years; 95% confidence interval [CI], 4.77-12.66 years), lymphocyte count (-0.15 109/L; 95% CI, -0.18 to -0.11), glucose levels (1.20 mmol/L; 95% CI, 0.73-1.68), and haemoglobin levels (-0.91 g/dL; 95% CI, -1.49 to -0.33) between those with favourable and unfavourable outcomes. The prevalence rates of unfavourable outcomes were as follows: abnormal cisterns, 65.7%; intracranial pressure above 20 mmHg, 52.9%; midline shift of 5 mm or more, 63%; hypotension, 71%; hypoxia, 86.8%; blood transfusion, 70.3%; and mechanical ventilation, 90%. Several predictors were strongly associated with outcome. Specifically, age, lymphocyte count, glucose level, haemoglobin level, severity of TBI, pupillary reaction, and type of injury were identified as potential predictors of long-term outcomes.

The impact of sedative and vasopressor agents on cerebrovascular reactivity in severe traumatic brain injury

BACKGROUND

The aim of this study is to evaluate the impact of commonly administered sedatives (Propofol, Alfentanil, Fentanyl, and Midazolam) and vasopressor (Dobutamine, Ephedrine, Noradrenaline and Vasopressin) agents on cerebrovascular reactivity in moderate/severe TBI patients. Cerebrovascular reactivity, as a surrogate for cerebral autoregulation was assessed using the long pressure reactivity index (LPRx). We evaluated the data in two phases, first we assessed the minute-by-minute data relationships between different dosing amounts of continuous infusion agents and physiological variables using boxplots, multiple linear regression and ANOVA. Next, we assessed the relationship between continuous/bolus infusion agents and physiological variables, assessing pre-/post- dose of medication change in physiology using a Wilcoxon signed-ranked test. Finally, we evaluated sub-groups of data for each individual dose change per medication, focusing on key physiological thresholds and demographics.

RESULTS

Of the 475 patients with an average stay of 10 days resulting in over 3000 days of recorded information 367 (77.3%) were male with a median Glasgow coma score of 7 (4-9). The results of this retrospective observational study confirmed that the infusion of most administered agents do not impact cerebrovascular reactivity, which is confirmed by the multiple linear regression components having p value > 0.05. Incremental dose changes or bolus doses in these medications in general do not lead to significant changes in cerebrovascular reactivity (confirm by Wilcoxon signed-ranked p value > 0.05 for nearly all assessed relationships). Within the sub-group analysis that separated the data based on LPRx pre-dose, a significance between pre-/post-drug change in LPRx was seen, however this may be more of a result from patient state than drug impact.

CONCLUSIONS

Overall, this study indicates that commonly administered agents with incremental dosing changes have no clinically significant influence on cerebrovascular reactivity in TBI (nor do they impair cerebrovascular reactivity). Though further investigation in a larger and more diverse TBI patient population is required.

Regional disparity in continuously measured time-domain cerebrovascular reactivity indices: a scoping review of human literature

Objective: Cerebral blood vessels maintaining relatively constant cerebral blood flow (CBF) over wide range of systemic arterial blood pressure (ABP) is referred to as cerebral autoregulation (CA). Impairments in CA expose the brain to pressure-passive flow states leading to hypoperfusion and hyperperfusion. Cerebrovascular reactivity (CVR) metrics refer to surrogate metrics of pressure-based CA that evaluate the relationship between slow vasogenic fluctuations in cerebral perfusion pressure/ABP and a surrogate for pulsatile CBF/cerebral blood volume.Approach: We performed a systematically conducted scoping review of all available human literature examining the association between continuous CVR between more than one brain region/channel using the same CVR index.Main Results: In all the included 22 articles, only handful of transcranial doppler (TCD) and near-infrared spectroscopy (NIRS) based metrics were calculated for only two brain regions/channels. These metrics found no difference between left and right sides in healthy volunteer, cardiac surgery, and intracranial hemorrhage patient studies. In contrast, significant differences were reported in endarterectomy, and subarachnoid hemorrhage studies, while varying results were found regarding regional disparity in stroke, traumatic brain injury, and multiple population studies.Significance: Further research is required to evaluate regional disparity using NIRS-based indices and to understand if NIRS-based indices provide better regional disparity information than TCD-based indices.

Temporal relationship between vasopressor and sedative administration and cerebrovascular response in traumatic brain injury: a time-series analysis

BACKGROUND

Although vasopressor and sedative agents are commonly used within the intensive care unit to mediate systemic and cerebral physiology, the full impact such agents have on cerebrovascular reactivity remains unclear. Using a prospectively maintained database of high-resolution critical care and physiology, the time-series relationship between vasopressor/sedative administration, and cerebrovascular reactivity was interrogated. Cerebrovascular reactivity was assessed through intracranial pressure and near infrared spectroscopy measures. Using these derived measures, the relationship between hourly dose of medication and hourly index values could be evaluated. The individual medication dose change and their corresponding physiological response was compared. Given the high number of doses of propofol and norepinephrine, a latent profile analysis was used to identify any underlying demographic or variable relationships. Finally, using time-series methodologies of Granger causality and vector impulse response functions, the relationships between the cerebrovascular reactivity derived variables were compared.

RESULTS

From this retrospective observational study of 103 TBI patients, the evaluation between the changes in vasopressor or sedative agent dosing and the previously described cerebral physiologies was completed. The assessment of the physiology pre/post infusion agent change resulted in similar overall values (Wilcoxon signed-ranked p value > 0.05). Time series methodologies demonstrated that the basic physiological relationships were identical before and after an infusion agent was changed (Granger causality demonstrated the same directional impact in over 95% of the moments, with response function being graphically identical).

CONCLUSIONS

This study suggests that overall, there was a limited association between the changes in vasopressor or sedative agent dosing and the previously described cerebral physiologies including that of cerebrovascular reactivity. Thus, current regimens of administered sedative and vasopressor agents appear to have little to no impact on cerebrovascular reactivity in TBI.

Associations between intracranial pressure thresholds and multimodal monitoring in acute traumatic neural injury: a scoping review

BACKGROUND

Current moderate/severe traumatic brain injury (TBI) guidelines suggest the use of an intracranial pressure (ICP) treatment threshold of 20 mmHg or 22 mmHg. Over the past decade, the use of various cerebral physiology monitoring devices has been incorporated into neurocritical care practice and termed "multimodal monitoring." Such modalities include those that monitor systemic hemodynamics, systemic and brain oxygenation, cerebral blood flow (CBF), cerebral autoregulation, electrophysiology, and cerebral metabolism. Given that the relationship between ICP and outcomes is not yet entirely understood, a comprehensive review of the literature on the associations between ICP thresholds and multimodal monitoring is still needed.

METHODS

We conducted a scoping review of the literature for studies that present an objective statistical association between ICP above/below threshold and any multimodal monitoring variable. MEDLINE, BIOSIS, Cochrane library, EMBASE, Global Health, and SCOPUS were searched from inception to July 2022 for relevant articles. Full-length, peer-reviewed, original works with a sample size of ≥50 moderate-severe TBI patients were included in this study.

RESULTS

A total of 13 articles were deemed eligible for final inclusion. The included articles were significantly heterogenous in terms of their designs, demographics, and results, making it difficult to draw any definitive conclusions. No literature describing the association between guideline-based ICP thresholds and measures of brain electrophysiology, cerebral metabolism, or direct metrics of CBF was found.

CONCLUSION

There is currently little literature that presents objective statistical associations between ICP thresholds and multimodal monitoring physiology. However, overall, the literature indicates that having ICP above guideline based thresholds is associated with increased blood pressure, increased cardiac decoupling, reduced parenchymal brain oxygen tension, and impaired cerebral autoregulation, with no association with CBF velocity within the therapeutic range of ICP. There was insufficient literature to comment on other multimodal monitoring measures.

Immune system disorders in the early post-injury period in patients after severe brain injury from the perspective of the severity of the injury

BACKGROUND

Brain injuries are the most common cause of death in productive age. Besides the extent of the injury, other systemic factors can also affect the outcome. Patients suffering from severe brain injury often experience extracranial inflammatory complications during the early period of treatment. Here, we investigate the changes in immunity in patients with brain injury.

METHODS

121 patients and 92 healthy controls were included in the research. Blood samples were collected on admission and analyzed by flow cytometry and biochemical methods. Multiple clusters of differentiation (CD) and antibody levels were investigated. The results were compared between patients and controls. In addition, results of two classes of severity (Glasgow Coma Scale, GCS, of 3-5 vs. 6-8) were also compared.

RESULTS

Parameters of humoral immunity in patients immediately after admission were significantly lower than those from healthy donors, with the exception of IgE elevated as much as to resemble allergic reaction (p < 0.01). Of cellular parameters, only natural killer (NK) cluster CD56 + was elevated (p < 0.01). Extracranial infectious complications were more common in patients with GCS 3-5.

CONCLUSIONS

Strong immune system disorders were observed in patients after severe brain injury, which may contribute to the worse outcome in such patients.

High spatial and temporal resolution cerebrovascular reactivity for humans and large mammals: A technological description of integrated fNIRS and niABP mapping system

Introduction: The process of cerebral vessels maintaining cerebral blood flow (CBF) fairly constant over a wide range of arterial blood pressure is referred to as cerebral autoregulation (CA). Cerebrovascular reactivity is the mechanism behind this process, which maintains CBF through constriction and dilation of cerebral vessels. Traditionally CA has been assessed statistically, limited by large, immobile, and costly neuroimaging platforms. However, with recent technology advancement, dynamic autoregulation assessment is able to provide more detailed information on the evolution of CA over long periods of time with continuous assessment. Yet, to date, such continuous assessments have been hampered by low temporal and spatial resolution systems, that are typically reliant on invasive point estimations of pulsatile CBF or cerebral blood volume using commercially available technology. Methods: Using a combination of multi-channel functional near-infrared spectroscopy and non-invasive arterial blood pressure devices, we were able to create a system that visualizes CA metrics by converting them to heat maps drawn on a template of human brain. Results: The custom Python heat map module works in "offline" mode to visually portray the CA index per channel with the use of colourmap. The module was tested on two different mapping grids, 8 channel and 24 channel, using data from two separate recordings and the Python heat map module was able read the CA indices file and represent the data visually at a preselected rate of 10 s. Conclusion: The generation of the heat maps are entirely non-invasive, with high temporal and spatial resolution by leveraging the recent advances in NIRS technology along with niABP. The CA mapping system is in its initial stage and development plans are ready to transform it from "offline" to real-time heat map generation.

Brain Tissue Oxygen Levels as a Perspective Therapeutic Target in Traumatic Brain Injury. Retrospective Cohort Study

Introduction

Management of traumatic brain injury (TBI) requires a multidisciplinary approach and represents a significant challenge for both neurosurgeons and intensivists. The role of brain tissue oxygenation (PbtO2) monitoring and its impact on posttraumatic outcomes remains a controversial topic.

Aim of the study

Our study aimed to evaluate the impact of PbtO2 monitoring on mortality, 30 days and 6 months neurological outcomes in patients with severe TBI compared with those resulting from standard intracranial pressure (ICP) monitoring.

Material and methods

In this retrospective cohort study, we analysed the outcomes of 77 patients with severe TBI who met the inclusion criteria. These patients were divided into two groups, including 37 patients who were managed with ICP and PbtO2 monitoring protocols and 40 patients who were managed using ICP protocols alone.

Results

There were no significant differences in demographic data between the two groups. We found no statistically significant differences in mortality or Glasgow Outcome Scale (GOS) scores one month after TBI. However, our results revealed that GOS scores at 6 months had improved significantly among patients managed with PbtO2; this finding was particularly notable for Glasgow Outcome Scale (GOS) scores of 4-5. Close monitoring and management of reductions in PbtO2, particularly by increasing the fraction of inspired oxygen, was associated with higher partial pressures of oxygen in this group.

Conclusions

Monitoring of PbtO2 may facilitate the appropriate evaluation and treatment of low PbtO2 and represents a promising tool for the management of patients with severe TBI. Additional studies will be needed to confirm these findings.

Association between cerebrovascular reactivity in adult traumatic brain injury and improvement in patient outcome over time: an exploratory analysis

BACKGROUND

Impaired cerebrovascular reactivity following moderate/severe traumatic brain injury (TBI) has emerged as a key potential driver of morbidity and mortality. However, the major contributions to the literature so far have been solely focused on single point measures of long-term outcome. Therefore, it remains unknown whether cerebrovascular reactivity impairment, during the acute phase of TBI, is associated with failure to improve in outcome across time.

METHODS

Cerebrovascular reactivity was measured using three intracranial pressure-based surrogate metrics. For each patient, % time spent above various literature-defined thresholds was calculated. Patients were dichotomized based on outcome transition into Improved vs Not Improved between 1 and 3 months, 3 and 6 months, and 1 and 6 months, based on the Glasgow Outcome Scale-Extended (GOSE). Univariate and multivariable logistic regression analyses were performed, adjusting for the International Mission for Prognosis and Analysis of Clinical Trials (IMPACT) variables.

RESULTS

Seventy-eight patients from the Winnipeg Acute TBI Database were included in this study. On univariate logistic regression analysis, higher % time with cerebrovascular reactivity metrics above clinically defined thresholds was associated with a lack of clinical improvement between 1 and 3 months and 1 and 6 months post injury (p < 0.05). These relationships held true on multivariable logistic regression analysis.

CONCLUSION

Our study demonstrates that impaired cerebrovascular reactivity, during the acute phase of TBI, is associated with failure to improve clinically over time. These preliminary findings highlight the significance that cerebrovascular reactivity monitoring carries in outcome recovery association in moderate/severe TBI.

Status Forecasting Based on the Baseline Information Using Logistic Regression

In the status forecasting problem, classification models such as logistic regression with input variables such as physiological, diagnostic, and treatment variables are typical ways of modeling. However, the parameter value and model performance differ among individuals with different baseline information. To cope with these difficulties, a subgroup analysis is conducted, in which models' ANOVA and rpart are proposed to explore the influence of baseline information on the parameters and model performance. The results show that the logistic regression model achieves satisfactory performance, which is generally higher than 0.95 in AUC and around 0.9 in F1 and balanced accuracy. The subgroup analysis presents the prior parameter values for monitoring variables including SpO2, milrinone, non-opioid analgesics and dobutamine. The proposed method can be used to explore variables that are and are not medically related to the baseline variables.

Optimal bispectral index level of sedation and cerebral oximetry in traumatic brain injury: a non-invasive individualized approach in critical care?

Background

Impaired cerebral autoregulation has been linked with worse outcomes, with literature suggesting that current therapy guidelines fail to significantly impact cerebrovascular reactivity. The cerebral oximetry index (COx_a) is a surrogate measure of cerebrovascular reactivity which can in theory be obtained non-invasively using regional brain tissue oxygen saturation and arterial blood pressure. The goal of this study was to assess the relationship between objectively measured depth of sedation through BIS and autoregulatory capacity measured through COx_a.

Methods

In a prospectively maintained observational study, we collected continuous regional brain tissue oxygen saturation, intracranial pressure, arterial blood pressure and BIS in traumatic brain injury patients. COx_a was obtained using the Pearson’s correlation between regional brain tissue oxygen saturation and arterial blood pressure and ranges from − 1 to 1 with higher values indicating impairment of cerebrovascular reactivity. Using BIS values and COx_a, a curve-fitting method was applied to determine the minimum value for the COx_a. The associated BIS value with the minimum COx_a is called BISopt. This BISopt was both visually and algorithmically determined, which were compared and assessed over the whole dataset.

Results

Of the 42 patients, we observed that most had a parabolic relationship between BIS and COx_a. This suggests a potential “optimal” depth of sedation where COx_a is the most intact. Furthermore, when comparing the BISopt algorithm with visual inspection of BISopt, we obtained similar results. Finally, BISopt % yield (determined algorithmically) appeared to be independent from any individual sedative or vasopressor agent, and there was agreement between BISopt found with COx_a and the pressure reactivity index (another surrogate for cerebrovascular reactivity).

Conclusions

This study suggests that COx_a is capable of detecting disruption in cerebrovascular reactivity which occurs with over-/under-sedation, utilizing a non-invasive measure of determination and assessment. This technique may carry implications for tailoring sedation in patients, focusing on individualized neuroprotection.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40635-022-00460-9.

Cerebral Autoregulation Monitoring in Traumatic Brain Injury: An Overview of Recent Advances in Personalized Medicine

Impaired cerebral autoregulation (CA) in moderate/severe traumatic brain injury (TBI) has been identified as a strong associate with poor long-term outcomes, with recent data highlighting its dominance over cerebral physiologic dysfunction seen in the acute phase post injury. With advances in bedside continuous cerebral physiologic signal processing, continuously derived metrics of CA capacity have been described over the past two decades, leading to improvements in cerebral physiologic insult detection and development of novel personalized approaches to TBI care in the intensive care unit (ICU). This narrative review focuses on highlighting the concept of continuous CA monitoring and consequences of impairment in moderate/severe TBI. Further, we provide a comprehensive description and overview of the main personalized cerebral physiologic targets, based on CA monitoring, that are emerging as strong associates with patient outcomes. CA-based personalized targets, such as optimal cerebral perfusion pressure (CPPopt), lower/upper limit of regulation (LLR/ULR), and individualized intra-cranial pressure (iICP) are positioned to change the way we care for TBI patients in the ICU, moving away from the "one treatment fits all" paradigm of current guideline-based therapeutic approaches, towards a true personalized medicine approach tailored to the individual patient. Future perspectives regarding research needs in this field are also discussed.

Dynamic Temporal Relationship Between Autonomic Function and Cerebrovascular Reactivity in Moderate/Severe Traumatic Brain Injury

There has been little change in morbidity and mortality in traumatic brain injury (TBI) in the last 25 years. However, literature has emerged linking impaired cerebrovascular reactivity (a surrogate of cerebral autoregulation) with poor outcomes post-injury. Thus, cerebrovascular reactivity (derived through the pressure reactivity index; PRx) is emerging as an important continuous measure. Furthermore, recent literature indicates that autonomic dysfunction may drive impaired cerebrovascular reactivity in moderate/severe TBI. Thus, to improve our understanding of this association, we assessed the physiological relationship between PRx and the autonomic variables of heart rate variability (HRV), blood pressure variability (BPV), and baroreflex sensitivity (BRS) using time-series statistical methodologies. These methodologies include vector autoregressive integrative moving average (VARIMA) impulse response function analysis, Granger causality, and hierarchical clustering. Granger causality testing displayed inconclusive results, where PRx and the autonomic variables had varying bidirectional relationships. Evaluating the temporal profile of the impulse response function plots demonstrated that the autonomic variables of BRS, ratio of low/high frequency of HRV and very low frequency HRV all had a strong relation to PRx, indicating that the sympathetic autonomic response may be more closely linked to cerebrovascular reactivity, then other variables. Finally, BRS was consistently associated with PRx, possibly demonstrating a deeper relationship to PRx than other autonomic measures. Taken together, cerebrovascular reactivity and autonomic response are interlinked, with a bidirectional impact between cerebrovascular reactivity and circulatory autonomics. However, this work is exploratory and preliminary, with further study required to extract and confirm any underlying relationships.

Continuous Time-Domain Cerebrovascular Reactivity Metrics and Discriminate Capacity for the Upper and Lower Limits of Autoregulation: A Scoping Review of the Animal Literature

Over a wide range of systemic arterial pressures, cerebral blood flow (CBF) is regulated fairly constantly by the cerebral vessels in a process termed cerebral autoregulation (CA), which is depicted by the Lassen autoregulatory curve. After traumatic brain injury (TBI), CA can get impaired and these impairments manifest in changes of the Lassen autoregulatory curve. Continuous surrogate metrics of pressure-based CA, termed cerebrovascular reactivity (CVR) metrics, evaluate the relationship between slow vasogenic fluctuations in a driving pressure for cerebral blood flow, and the most commonly studied and utilized measures are based in the time domain and have been increasingly applied in bedside TBI care and have sparked the investigation of individualized cerebral perfusion pressure targets. However, not all CVR metrics have been validated as true measures of autoregulation in the pre-clinical setting. We reviewed all available pre-clinical animal literature that assessed the association between continuous time-domain metrics of CVR and some aspect of the Lassen autoregulatory curve. All 15 articles found associated the evaluated continuous metrics to the lower limit of autoregulation curve whereas none looked at the upper limit. Most of the evaluated metrics showed the ability to discriminate the lower limit of autoregulation with various methods of perturbation. Further work is required to evaluate the utility of such surrogate measures against the upper limit of autoregulation, while also providing validation to the existing literature supporting specific indices and their ability to discriminate the lower limit.

Intracranial Pressure–Derived Cerebrovascular Reactivity Indices, Chronological Age, and Biological Sex in Traumatic Brain Injury: A Scoping Review

To date, there has been limited literature exploring the association between age and sex with cerebrovascular reactivity (CVR) in moderate/severe traumatic brain injury (TBI). Given the known link between age, sex, and cerebrovascular function, knowledge of the impacts on continuously assessed CVR is critical for the development of future therapeutics. We conducted a scoping review of the literature for studies that had a direct statistical interrogation of the relationship between age, sex, and continuous intracranial pressure (ICP)-based indices of CVR in moderate/severe TBI. The ICP-based indices researched included: pressure reactivity index (PRx), pulse amplitude index (PAx), and RAC. MEDLINE, BIOSIS, EMBASE, SCOPUS, Global Health, and the Cochrane library were searched from inception to June 2021 for relevant articles. A total of 10 original studies fulfilled our inclusion criteria. Nine of the articles documented a correlation between advanced age and worse CVR, with eight using PRx (2192 total patients), three using PAx (978 total patients), and one using RAC (358 total patients), p < 0.05; R ranging from 0.17 to 0.495 for all indices across all studies. Three articles (1256 total patients) displayed a correlation between biological sex and PRx, with females trending towards higher PRx values (p < 0.05) in the limited available literature. However, no literature exists comparing PAx or RAC with biological sex. Findings showed that aging was associated with impaired CVR. We observed a trend between female sex and worse PRx values, but the literature was limited and statistical significance was borderline. The identified studies were few in number, carried significant population heterogeneity, and utilized grand averaging of large epochs of physiology during statistical comparisons with age and biological sex. Because of the heterogeneous nature of TBI populations and limited focus on the effects of age and sex on outcomes in TBI, it is challenging to highlight the differences between the indices and patient age groups and sex. The largest study showing an association between PRx and age was done by Zeiler and colleagues, where 165 patients were studied noting that patients with a mean PRx value above zero had a mean age above 51.4 years versus a mean age of 41.4 years for those with a mean PRx value below zero (p = 0.0007). The largest study showing an association between PRx and sex was done by Czosnyka and colleagues, where 469 patients were studied noting that for patients <50 years of age, PRx was worse in females (0.11 ± 0.047) compared to males (0.044 ± 0.031), p < 0.05. The findings from these 10 studies provide preliminary data, but are insufficient to definitively characterize the impact of age and sex on CVR in moderate/severe TBI. Future work in the field should focus on the impact of age and sex on multi-modal cerebral physiological monitoring.

Association of Age and Sex With Multi-Modal Cerebral Physiology in Adult Moderate/Severe Traumatic Brain Injury: A Narrative Overview and Future Avenues for Personalized Approaches

The impact of age and biological sex on outcome in moderate/severe traumatic brain injury (TBI) has been documented in large cohort studies, with advanced age and male sex linked to worse long-term outcomes. However, the association between age/biological sex and high-frequency continuous multi-modal monitoring (MMM) cerebral physiology is unclear, with only sparing reference made in guidelines and major literature in moderate/severe TBI. In this narrative review, we summarize some of the largest studies associating various high-frequency MMM parameters with age and biological sex in moderate/severe TBI. To start, we present this by highlighting the representative available literature on high-frequency data from Intracranial Pressure (ICP), Cerebral Perfusion Pressure (CPP), Extracellular Brain Tissue Oxygenation (PbtO₂), Regional Cerebral Oxygen Saturations (rSO₂), Cerebral Blood Flow (CBF), Cerebral Blood Flow Velocity (CBFV), Cerebrovascular Reactivity (CVR), Cerebral Compensatory Reserve, common Cerebral Microdialysis (CMD) Analytes and their correlation to age and sex in moderate/severe TBI cohorts. Then we present current knowledge gaps in the literature, discuss biological implications of age and sex on cerebrovascular monitoring in TBI and some future avenues for bedside research into the cerebrovascular physiome after TBI.

A machine learning model for predicting favorable outcome in severe traumatic brain injury patients after 6 months

Traumatic brain injury (TBI), which occurs commonly worldwide, is among the more costly of health and socioeconomic problems. Accurate prediction of favorable outcome in severe TBI patients could assist with optimizing treatment procedures, predicting clinical outcomes, and result in substantial economic savings. In this study, we examined the capability of a machine learning-based model in predicting "favorable" or "unfavorable" outcome after 6 months in severe TBI patients using only parameters measured on admission. Three models were developed using logistic regression, random forest, and support vector machines trained on parameters recorded from 2,381 severe TBI patients admitted to the neuro-intensive care unit of Rajaee (Emtiaz) Hospital (Shiraz, Iran) between 2015 and 2017. Model performance was evaluated using three indices: sensitivity, specificity, accuracy, and area under the curve (AUC). Ten-fold cross-validation method was used to estimate these indices. Overall, the developed models showed excellent performance with AUC >0.81, sensitivity and specificity of > 0.78. The top-three factors important in predicting 6-month post-trauma survival status in TBI patients are "GCS motor response," "pupillary reactivity," and "age." Machine learning techniques might be used to predict the 6-month outcome in TBI patients using only the parameters measured on admission when the machine learning is trained using a large data set.

Impact of Age and Biological Sex on Cerebrovascular Reactivity in Adult Moderate/Severe Traumatic Brain Injury: An Exploratory Analysis

Age and biological sex are two potential important modifiers of cerebrovascular reactivity post-traumatic brain injury (TBI) requiring close evaluation for potential subgroup responses. The goal of this study was to provide a preliminary exploratory analysis of the impact of age and biological sex on measures of cerebrovascular function in moderate/severe TBI. Forty-nine patients from the prospectively maintained TBI database at the University of Manitoba with archived high-frequency digital cerebral physiology were evaluated. Cerebrovascular reactivity indices were derived as follows: PRx (correlation between intracranial pressure [ICP] and mean arterial pressure [MAP]), PAx (correlation between pulse amplitude of ICP [AMP] and MAP), and RAC (correlation between AMP and cerebral perfusion pressure [CPP]). Time above clinically significant thresholds for each index was calculated over different periods of the acute intensive care unit stay. The association between PRx, PAx, and RAC measures with age was assessed using linear regression, and an age trichotomization scheme (<40, 40-60, >60) using Kruskal-Wallis testing. Similarly, association with biological sex was tested using Mann-Whitney U testing. Biological sex did not demonstrate an impact on any measures of cerebrovascular reactivity. Linear regression between age and PAx and RAC demonstrated a statistically significant positive linear relationship. Median PAx and RAC measures between trichotomized age categories demonstrated statistically significant increases with advancing age. The PRx failed to demonstrate any statistically significant relationship with age in this cohort, suggesting that in elderly patients with controlled ICP, PAx and RAC may be better metrics for detecting impaired cerebrovascular reactivity. Biological sex appears to not be associated with differences in cerebrovascular reactivity in this cohort. The PRx performed the worst in detecting impaired cerebrovascular reactivity in those with advanced age, where PAx and RAC appear to have excelled. Future work is required to validate these findings and explore the utility of different cerebrovascular reactivity indices.

Integrative Neuroinformatics for Precision Prognostication and Personalized Therapeutics in Moderate and Severe Traumatic Brain Injury

Despite changes in guideline-based management of moderate/severe traumatic brain injury (TBI) over the preceding decades, little impact on mortality and morbidity have been seen. This argues against the “one-treatment fits all” approach to such management strategies. With this, some preliminary advances in the area of personalized medicine in TBI care have displayed promising results. However, to continue transitioning toward individually-tailored care, we require integration of complex “-omics” data sets. The past few decades have seen dramatic increases in the volume of complex multi-modal data in moderate and severe TBI care. Such data includes serial high-fidelity multi-modal characterization of the cerebral physiome, serum/cerebrospinal fluid proteomics, admission genetic profiles, and serial advanced neuroimaging modalities. Integrating these complex and serially obtained data sets, with patient baseline demographics, treatment information and clinical outcomes over time, can be a daunting task for the treating clinician. Within this review, we highlight the current status of such multi-modal omics data sets in moderate/severe TBI, current limitations to the utilization of such data, and a potential path forward through employing integrative neuroinformatic approaches, which are applied in other neuropathologies. Such advances are positioned to facilitate the transition to precision prognostication and inform a top-down approach to the development of personalized therapeutics in moderate/severe TBI.

Sedation and cerebrovascular reactivity in traumatic brain injury: another potential avenue for personalized approaches in neurocritical care?

BACKGROUND

Impaired cerebrovascular reactivity appears to be linked to worse global outcome in adult traumatic brain injury (TBI). Literature suggests that current treatments administered in TBI care, in the intensive care unit (ICU), fail to greatly impact recorded cerebrovascular reactivity measures. In particular, the impact of sedation on cerebrovascular reactivity in traumatic brain injury (TBI) remains unclear in vivo. The goal of this study was to preliminarily assess the relationship between objectively measured depth of sedation and cerebrovascular reactivity in TBI.

METHODS

Within, we describe a case series of 5 adult TBI patients with TBI, during which objective high-frequency physiology for sedation depth, using bispectral index (BIS), and both intracranial pressure (ICP) and arterial blood pressure (ABP) were recorded. Pressure reactivity index (PRx) and RAP (a metric of cerebral compensatory reserve) were derived. Relationships between cerebrovascular reactivity and compensatory reserve monitoring with BIS metrics were explored using descriptive plots.

RESULTS

A total of 5 cases in our prospectively maintained database with high-frequency physiology for ICP, ABP, and BIS. Through error bar plotting, it can be seen that each patient displays a parabolic relationship between BIS and PRx. This suggests a potential "optimal" depth of sedation where cerebrovascular reactivity is the most intact.

CONCLUSIONS

This small series highlights the potential impact of depth of sedation on cerebrovascular reactivity in TBI. It suggests that there may be an individual optimal depth of sedation, so as to optimize cerebrovascular reactivity. Further study of objective depth of sedation and its impact on cerebrovascular physiology in TBI is required.

The Impact of Vasopressor and Sedative Agents on Cerebrovascular Reactivity and Compensatory Reserve in Traumatic Brain Injury: An Exploratory Analysis

The impact of vasopressor and sedative drugs on cerebrovascular reactivity in traumatic brain injury (TBI) remains unclear. The aim of this study was to evaluate the impact of changes of doses of commonly administered sedation (i.e., propofol, fentanyl, and ketamine) and vasopressor agents (i.e., norepinephrine [NE], phenylephrine [PE], and vasopressin[VSP]) on cerebrovascular reactivity and compensatory reserve in patients with moderate/severe TBI. Using the Winnipeg Acute TBI Database, we identified 38 patients with more than 1000 distinct changes of infusion rates and more than 500 h of paired drug infusion/physiology data. Cerebrovascular reactivity was assessed using pressure reactivity index (PRx) and cerebral compensatory reserve was assessed using RAP (the correlation [R] between pulse amplitude of intracranial pressure [ICP; A] and ICP [P]). We evaluated the data in two phases. First, we assessed the relationship between mean hourly dose of medication and its relation to both mean hourly index values, and time spent above a given index threshold. Second, we evaluated time-series data for each individual dose change per medication, assessing for a statistically significant change in PRx and RAP metrics. The results of the analysis confirmed that, overall, the mean hourly dose of sedative (propofol, fentanyl, and ketamine) and vasopressor (NE, PE, and VSP) agents does not impact hourly cerebrovascular reactivity or compensatory reserve measures. Similarly, incremental dose changes in these medications in general do not lead to significant changes in cerebrovascular reactivity or compensatory reserve. For propofol with incremental dose increases, in situations where PRx is intact (i.e., PRx <0 prior), a statistically significant increase in PRx was seen. However, this may not indicate deteriorating cerebrovascular reactivity as the final PRx (∼0.05) may still be considered to be intact cerebrovascular reactivity. As such, this finding with regards to propofol remains “weak.” This study indicates that commonly administered sedative and vasopressor agents with incremental dosing changes have no clinically significant influence on cerebrovascular reactivity or compensatory reserve in TBI. These results should be considered preliminary, requiring further investigation.

The impact of hypertonic saline on cerebrovascular reactivity and compensatory reserve in traumatic brain injury: an exploratory analysis

Background

Intravenous hypertonic saline is utilized commonly in critical care for treatment of acute or refractory elevations of intracranial pressure (ICP) in traumatic brain injury (TBI) patients. Though there is a clear understanding of the general physiological effects of a hypertonic saline solution over long periods of time, smaller epoch effects of hypertonic saline (HTS) have not been thoroughly analyzed. The aim of this study was to perform a direct evaluation of the high-frequency response of HTS on the cerebrovascular physiological responses in TBI.

Methods

We retrospectively reviewed our prospectively maintained adult TBI database for those with archived high-frequency cerebral physiology and available HTS treatment information. We evaluated different epochs of physiology around HTS bolus dosing, comparing pre- with post-HTS. We assessed for changes in slow fluctuations in ICP, pulse amplitude of ICP (AMP), cerebral perfusion pressure (CPP), mean arterial pressure (MAP), cerebrovascular reactivity (as measured through pressure reactivity index (PRx)), and cerebral compensatory reserve (correlation (R) between AMP (A) and ICP (P)). Comparisons of mean measures and percentage time above clinically relevant thresholds for the physiological parameters were compared pre- and post-HTS using descriptive statistics and Mann-Whitney U testing. We assessed for subgroups of physiological responses using latent profile analysis (LPA).

Results

Fifteen patients underwent 69 distinct bolus infusions of hypertonic saline. Apart from the well-documented decrease in ICP, there was also a reduction in AMP. The analysis of cerebrovascular reactivity response to HTS solution had two main effects. For patients with grossly impaired cerebrovascular reactivity pre-HTS (PRx > + 0.30), HTS bolus led to improved reactivity. However, for those with intact cerebrovascular reactivity pre-HTS (PRx < 0), HTS bolus demonstrated a trend towards more impaired reactivity. This indicates that HTS has different impacts, dependent on pre-bolus cerebrovascular status. There was no significant change in metrics of cerebral compensatory reserve. LPA failed to demonstrate any subgroups of physiological responses to HTS administration.

Conclusions

The direct decrease in ICP and AMP confirms that a bolus dose of a HTS solution is an effective therapeutic agent for intracranial hypertension. However, in patients with intact autoregulation, hypertonic saline may impair cerebral hemodynamics. These findings regarding cerebrovascular reactivity remain preliminary and require further investigation.

Electronic supplementary material

The online version of this article (10.1007/s00701-020-04579-0) contains supplementary material, which is available to authorized users.

Cerebrovascular Response to Phenylephrine in Traumatic Brain Injury: A Scoping Systematic Review of the Human and Animal Literature

Intravenous phenylephrine (PE) is utilized commonly in critical care for cardiovascular support. Its impact on the cerebrovasculature is unclear and its use may have important implications during states of critical neurological illness. The aim of this study was to perform a scoping review of the literature on the cerebrovascular/cerebral blood flow (CBF) effects of PE in traumatic brain injury (TBI), evaluating both animal models and human studies. We searched MEDLINE, BIOSIS, EMBASE, Global Health, SCOPUS, and the Cochrane Library from inception to January 2020. We identified 12 studies with various animal models and 4 studies in humans with varying TBI pathology. There was a trend toward a consistent increase in mean arterial pressure (MAP) by the injection of PE systemically, and by proxy, an increase of the cerebral perfusion pressure (CPP). There was a consistent constriction of cerebral vessels by PE reported in the small number of studies documenting such a response. However, the heterogeneity of the literature on the CBF/cerebral blood volume (CBV) response makes the strength of the conclusions on PE limited. Studies were heterogeneous in design and had significant limitations, with most failing to adjust for confounding factors in cerebrovascular/CBF response. This review highlights the significant knowledge gap on the cerebrovascular/CBF effects of PE administration in TBI, calling for further study on the impact of PE on the cerebrovasculature both in vivo and in experimental settings.

The Limited Impact of Current Therapeutic Interventions on Cerebrovascular Reactivity in Traumatic Brain Injury: A Narrative Overview

Current intensive care unit (ICU) treatment strategies for traumatic brain injury (TBI) care focus on intracranial pressure (ICP)- and cerebral perfusion pressure (CPP)-directed therapeutics, dictated by guidelines. Impaired cerebrovascular reactivity in moderate/severe TBI is emerging as a major associate with poor outcome and appears to dominate the landscape of physiologic derangement over the course of a patient's ICU stay. Within this article, we review the literature on the known drivers of impaired cerebrovascular reactivity in adult TBI, highlight the current knowledge surrounding the impact of guideline treatment strategies on continuously monitored cerebrovascular reactivity, and discuss current treatment paradigms for impaired reactivity. Finally, we touch on the areas of future research, as we strive to develop specific therapeutics for impaired cerebrovascular reactivity in TBI. There exists limited literature to suggest advanced age, intracranial injury patterns of diffuse injury, and sustained ICP elevations may drive impaired cerebrovascular reactivity. To date, the literature suggests there is a limited impact of such ICP/CPP guideline-based therapies on cerebrovascular reactivity, with large portions of a given patients ICU period spent with impaired cerebrovascular reactivity. Emerging treatment paradigms focus on the targeting individualized CPP and ICP thresholds based on cerebrovascular reactivity, without directly targeting the pathways involved in its dysfunction. Further work involved in uncovering the molecular pathways involved in impaired cerebrovascular reactivity is required, so that we can develop therapeutics directed at its prevention and treatment.