Skip to main content
Intensive Care Medicine Experimental
Download PDF
  • Research Articles
  • Open access
  • Published:

Cerebrovascular pressure reactivity and brain tissue oxygen monitoring provide complementary information regarding the lower and upper limits of cerebral blood flow control in traumatic brain injury: a CAnadian High Resolution-TBI (CAHR-TBI) cohort study

Intensive Care Medicine Experimental volume 10, Article number: 54 (2022) Cite this article

Abstract

Background

Brain tissue oxygen tension (PbtO2) and cerebrovascular pressure reactivity monitoring have emerged as potential modalities to individualize care in moderate and severe traumatic brain injury (TBI). The relationship between these modalities has had limited exploration. The aim of this study was to examine the relationship between PbtO2 and cerebral perfusion pressure (CPP) and how this relationship is modified by the state of cerebrovascular pressure reactivity.

Methods

A retrospective multi-institution cohort study utilizing prospectively collected high-resolution physiologic data from the CAnadian High Resolution-TBI (CAHR-TBI) Research Collaborative database collected between 2011 and 2021 was performed. Included in the study were critically ill TBI patients with intracranial pressure (ICP), arterial blood pressure (ABP), and PbtO2 monitoring treated in any one of three CAHR-TBI affiliated adult intensive care units (ICU). The outcome of interest was how PbtO2 and CPP are related over a cohort of TBI patients and how this relationship is modified by the state of cerebrovascular reactivity, as determined using the pressure reactivity index (PRx).

Results

A total of 77 patients met the study inclusion criteria with a total of 377,744 min of physiologic data available for the analysis. PbtO2 produced a triphasic curve when plotted against CPP like previous population-based plots of cerebral blood flow (CBF) versus CPP. The triphasic curve included a plateau region flanked by regions of relative ischemia (hypoxia) and hyperemia (hyperoxia). The plateau region shortened when cerebrovascular pressure reactivity was disrupted compared to when it was intact.

Conclusions

In this exploratory analysis of a multi-institution high-resolution physiology TBI database, PbtO2 seems to have a triphasic relationship with CPP, over the entire cohort. The CPP range over which the plateau exists is modified by the state of cerebrovascular reactivity. This indicates that in critically ill TBI patients admitted to ICU, PbtO2 may be reflective of CBF.

Background

Despite traumatic brain injury (TBI) being a leading cause of death and disability worldwide [1], current guideline-based management, focused on intracranial pressure (ICP) and cerebral perfusion pressure (CPP), has resulted in limited recent improvements in outcome following moderate and severe TBI [2, 3]. Interest has shifted towards personalized medicine-based strategies that leverage contemporary multimodal monitoring techniques [4, 5]. Two methods that have come to the forefront are brain tissue oxygen tension (PbtO2) and ICP-based cerebrovascular pressure reactivity monitoring [6, 7].

PbtO2 monitoring requires the placement of a probe into viable brain tissue to measure oxygen tension [8]. A phase 2 randomized control trial (RCT) as well as observational studies have pointed towards improved outcomes when utilized in the TBI population [7, 9]. Consensus-based guidelines exist that outline the management of critically ill TBI patients with both ICP and PbtO2 monitoring and at least three large phase 3 RCTs are currently underway comparing PbtO2 augmented management to ICP and CPP based management alone in the setting of TBI [10,11,12,13].

Continuous cerebrovascular pressure reactivity monitoring uses the existing in situ ICP monitor [14]. While numerous methods exist, the pressure-reactivity index (PRx), which is the continuously computed correlation between slow-wave fluctuations in arterial blood pressure (ABP) and ICP, is the most studied and validated. PRx values range from − 1 to + 1 and cerebrovascular pressure reactivity is disrupted when there is a strong positive (> + 0.25) correlation between ICP and ABP [6]. Furthermore, a PRx above + 0.25 has been associated with worse neurologic outcomes following TBI [15, 16]. PRx is not directly modifiable, but the existence of a parabolic relationship between PRx and CPP in individual patients has led to the concept of personalized ‘optimal’ CPP management where individual PRx functions best [17, 18]. A full description of the computational methods is found elsewhere [17,18,19,20,21]. The output of this real-time analysis is a CPP at which PRx is at a minimum (CPPopt). The upper limit of reactivity (ULR) and lower limit of reactivity (LLR), are computed as the lower and upper limits of CPP values where PRx rises above + 0.25 [19, 22,23,24,25]. A management algorithm aiming to maintain CPP as close as possible to CPPopt has proven to be feasible and safe in a phase 2 RCT [21]. However, currently no Phase 3 data exist to support its efficacy in improving outcomes and no clinical guidelines are available for its use in TBI.

There are few studies examining the relationship between CPP, PRx and PbtO2 and none have leveraged contemporary data science techniques [26]. This work describes an exploratory multi-institutional retrospective cohort study that examines the interaction between these variables by leveraging prospectively collected high-resolution physiologic data from the CAnadian High Resolution-TBI (CAHR-TBI) Research Collaborative [27]. The primary aim of this study is to examine the relationship between PbtO2 and CPP in critically ill TBI patients. Secondary aims include: (1) examining how the relationship between PbtO2 and CPP is modified by the state of cerebrovascular pressure reactivity; (2) examining the relationship between PbtO2 and deviation from CPPopt, and (3) the independent prognostic utility of PbtO2 and CPPopt-based monitoring.

Methods

Study design

A retrospective multicenter cohort study utilizing a prospectively collected database of critically ill TBI patients was performed. The data originated from the CAnadian High Resolution-TBI (CAHR-TBI) Research Collaborative [27, 28]. Patients were admitted to one of three university-affiliated hospitals: Vancouver General Hospital (University of British Colombia), Foothills Medical Centre (University of Calgary), and Health Sciences Centre Winnipeg (University of Manitoba). Local research ethics approval at the University of Manitoba has been obtained for all aspects of this database (H2017:181 and H2017:188). Similarly, ethics approval was obtained for retrospective access to the database and for anonymous data transfer with each center for this project (H2020:118, H20-03,759 and REB20-0482).

Patient population

The CAHR-TBI database includes TBI patients treated in an adult intensive care unit (ICU) with invasive ICP and ABP monitoring at one of the affiliated hospitals. All patients were cared for using contemporary management strategies based on Brain Trauma Foundation (BTF) guidelines [2]. However, CPP values greater than 70 mmHg were generally not therapeutically lowered. PbtO2 and cerebrovascular pressure reactivity was managed based on local practice norms and varied from aggressive management to purely observation. Granular patient-specific and center-specific differences in management were not well captured and were therefore unavailable for incorporation into this study. Typically, based on common practice patterns at participating institutions, similar vasopressor, sedative and hyperosmolar/hypertonic agents were utilized between all sites. Patients with mild TBI or without invasive ICP monitoring were excluded [27]. Patient data were entered into the database from 2011 to 2021.

Included in this study were all patients in the CAHR-TBI database that had concurrent invasive PbtO2 monitoring. Those without PbtO2 monitoring were excluded. Age, biologic sex, admission Glasgow Coma Score (GCS) motor score, admission pupil exam, and 6- to 12-month Glasgow Outcome Score (GOS) (based on individual site outcome assessment periods), were extracted when available. Given the exploratory nature of this study, sample size calculations were not possible and therefore not performed.

High-resolution physiologic data collection

Three high-resolution physiologic data streams were utilized; ABP, ICP, and PbtO2. ABP was measured utilizing radial arterial lines. ICP was monitored using intra-parenchymal strain gauge probes (Codman ICP MicroSensor; Codman & Shurtlef Inc., Raynham, MA) placed in the frontal lobe or using external ventricular drains (Medtronic, Minneapolis, MN). PbtO2 was measured using intra-parenchymal brain tissue oxygenation probes (Licox Brain Tissue Oxygen Monitoring System; Integra LifeSciences Corp., Plainsboro, New Jersey) placed in viable frontal lobe tissue.

Data streams were recorded in digital high-frequency time series (≥ 100 Hz for ABP and ICP, 1 Hz for PbtO2) using analogue-to-digital signal converters (Data Translations, DT9804 or DT9826) where applicable. This digitized data was linked and stored in time series using Intensive Care Monitoring (ICM +) software (Cambridge Enterprise Ltd, Cambridge, UK).

Physiologic data cleaning and processing

For all high-resolution physiologic data, artifact clearing was performed utilizing manual removal by a qualified clinician utilizing ICM + software. All data were cleaned without knowledge of patient demographic or outcome information.

For each patient dataset, ABP and ICP waveforms were decimated using a 10-s moving average filter to eliminate higher frequency oscillations in these signals using ICM + functions. The resulting signals were used to derive a minute-by-minute updating value of PRx using a continuously updating Pearson correlation between ABP and ICP [17]. CPP was also derived as the difference between ABP and ICP. CPPopt, ULR, and LLR were all computed as continuously updated time series variables, where possible, utilizing a weighted multi-window technique examining the parabolic relationship between CPP and PRx as described recently [19, 21, 23,24,25]. A PRx value of + 0.25 was utilized as the threshold to determine both the ULR and LLR based on previous literature [16]. No attempt was made to interpolate missing data as entirely continuous data streams were not required for the subsequent analysis. Finally, all recorded and derived physiologic data (ABP, ICP, PbtO2, CPP, PRx, CPPopt, ULR and LLR) were exported as minute-by-minute comma-separated value (CSV) files for use in data analysis.

Physiologic data analysis and statistical methods

The data analysis was performed using R statistical software (R Foundation for Statistical Computing, Vienna, Austria). The data were isolated for values of PbtO2 greater than 0 mmHg, ABP greater than 20 mmHg, and CPP values between 20 and 150 mmHg as data outside of these parameters were likely artifactual. Next, a delta-CPPopt (ΔCPPopt) was calculated as the difference between CPP and CPPopt for every available datapoint. Finally, the following parameters were calculated for each subject in the study:

  1. 1.

    Percent time with CPP less than 60 mmHg and percent time with CPP greater than 70 mmHg as these are defined targets in BTF guidelines [2].

  2. 2.

    Percent time with PbtO2 less than 20 mmHg as this is a defined threshold in the literature [9,10,11,12].

  3. 3.

    Percent time with CPP less than the computed LLR and percent time with CPP above the computed ULR.

Plots were created using a generalized additive model smoothing function. This included visualization of the relationship between PbtO2 and CPP as well as the relationship between PbtO2 and ΔCPPopt over the entire cohort to identify global relationships. The effect of cerebrovascular pressure reactivity function, defined with a PRx threshold of 0.25, was also examined as a modifier to these relationships.

Finally, multivariable logistic regression modeling was performed to examine the utility of various parameters and exposures for predicting dichotomized 6- to 12-month functional outcomes (Favorable: GOS 4–5 vs Unfavorable: GOS 1–3 and Alive vs Dead). This was performed to gain insight into the possible impact these exposures may have on clinical outcomes following TBI. Models all included the International Mission for Prognosis and Analysis of Clinical Trials in TBI (TBI-IMPACT) variables of age, admission GCS motor score, admission pupil exam, and Marshall CT classification [29]. In total, six models were examined:

  1. 1.

    TBI-IMPACT.

  2. 2.

    TBI-IMPACT + Percent time with CPP less than 60 mmHg + Percent time with CPP greater than 70 mmHg.

  3. 3.

    TBI-IMPACT + Percent time with CPP less than 60 mmHg + Percent time with CPP greater than 70 mmHg + Percent time with PbtO2 less than 20 mmHg.

  4. 4.

    TBI-IMPACT + Percent time with CPP less than the computed LLR.

  5. 5.

    TBI-IMPACT + Percent time with CPP above the computed ULR.

  6. 6.

    TBI-IMPACT + Percent time with CPP less than the computed LLR + Percent time with CPP above the computed ULR.

Model quality was assessed utilizing area under the curve-receiver operator characteristic (AUC-ROC) analysis. Additionally, the quality of each model was further examined through Akaike Information Criteria (AIC) and Nagelkerke pseudo-R2 analysis. The alpha was set at 0.05 and due to the exploratory nature of the analysis, correction for multiple comparisons was not performed.

Results

Cohort demographics

In total, 77 patients from the CAHR-TBI database met the inclusion criteria for this study (57 from the University of British Colombia, 12 from the University of Calgary, and 8 from the University of Manitoba). Demographic data for the cohort are summarized in Table 1.

Table 1 Patient demographics for the cohort
Full size table

Physiologic data

Following artifact clearing and data filtering, a total of 377,744 min of unique physiologic data recordings were available for analysis in this cohort with a median of 4882 min (IQR: 3219–7884 min) of physiologic data for each patient. The yield of the CPPopt calculations were acceptable with a median percent yield of 70% (IQR: 55–81%) for each patient [19, 21].

Qualitative visual analysis

The relationship between PbtO2 and CPP over the cohort can be seen in Fig. 1A. The plot resembles the classical, population-based, triphasic Lassen curve with a plateau between CPP values of approximately 60 mmHg to 110 mmHg [30]. Below this plateau region, PbtO2 sharply falls and above it PbtO2 sharply rises. Interestingly the plateau region has associated PbtO2 values between 25 and 30 mmHg which is notably higher than the current guideline-based treatment threshold of 20 mmHg [7, 9,10,11,12]. Unlike the classic population-based Lassen curve there are two distinct humps that occur within the plateau region.

Fig. 1
figure 1

The relationship of brain tissue oxygenation (PbtO2) versus cerebral perfusion pressure (CPP) is seen in (A). In B, this relationship is examined when cerebrovascular pressure reactivity is intact (green) as compared to when it is disrupted (red). Note that 95% confidence intervals for all tracings are shown in grey

Full size image

In Fig. 1B, the recorded physiologic data are divided based on if cerebrovascular pressure reactivity was intact (green) or disrupted (red) using a PRx threshold of + 0.25. In both tracings a variation of the classic triphasic, population-based, Lassen curve is seen [30], however, the plateau region is substantially expanded when PRx is less than + 0.25 than when it is greater than + 0.25. Again, the plateau region is associated with PbtO2 values between 25 and 30 mmHg. When cerebrovascular pressure reactivity is intact, two distinct humps occur within the plateau region. When cerebrovascular pressure reactivity is disrupted the hump at higher CPP values is interrupted, shortening the plateau region. Individual patient plots, found in Additional file 1: Appendix SA, did not consistently demonstrate this pattern.

The relationship between PbtO2 and ΔCPPopt over the cohort is presented in Fig. 2A. Additionally, the median ΔCPPopt associated with the LLR and ULR in the cohort are shown in red and green, respectively. Once again, a triphasic curve is observed with a plateau region containing two humps. The CPPopt, and the median ΔCPPopt associated with the LLR and ULR appear to be primarily centered around the hump at higher CPP values. In Fig. 2B, this relationship is examined when cerebrovascular pressure reactivity is intact (green) versus when it is disrupted (red). As with Fig. 1b while both curves contain a plateau region, that region spans a much larger range when cerebrovascular pressure reactivity is intact. Again, the plateau is found to be associated with a PbtO2 between 25 and 30 mmHg. Individual patient plots, found in Additional file 1: Appendix SB, did not consistently demonstrate this pattern.

Fig. 2
figure 2

The relationship between brain tissue oxygenation (PbtO2) and difference from optimal cerebral perfusion pressure (ΔCPPopt) is seen in A. The median ΔCPPopt value associated with the lower limit of reactivity (red) and upper limit of reactivity (green) are also plotted. In B, this relationship is examined when cerebrovascular pressure reactivity is intact (green) as compared to when it is disrupted (red). Note that 95% confidence intervals for all tracings are shown in grey

Full size image

Functional outcome models

The results of the multivariable logistic regression, AUC, AIC, and Nagelkerke pseudo-R2 analysis can be seen in Table 2. Multivariable logistic regression showed that none of the individual factors were statistically significant. There was no clear trend for models incorporating percent time with PbtO2 less than 20 mmHg, percent time below the LLR, and/or percent time above the ULR to perform better than those that were based on TBI-IMPACT variables and constant CPP parameters alone. The superiority of the PbtO2-based model over CPPopt-based models, or vice versa, was also equivocal.

Table 2 Performance of various multivariable logistic regression models to predict 6-month post-injury outcomes
Full size table

Discussion

Qualitative visual analysis

Through visualizing the relationship between PbtO2 and CPP or ΔCPPopt several interesting observations can be made. The first is that PbtO2, like cerebral blood flow (CBF), produces a triphasic curve when plotted against CPP or ΔCPPopt with a plateau region flanked by regions of relative ischemia (hypoxia) and hyperemia (hyperoxia). This plateau in PbtO2 over a range of CPP or ΔCPPopt values supports work previously conducted by Jaeger and colleagues in a smaller single-institution cohort [26].

The plateau region redemonstrates features recently identified by Klein et al. in a model of hypo- and hypertension in pigs with healthy brains [31]. In that study, which involved the intermittent measurement of CBF and pial vessel diameter through direct florescence microscopy by cranial window, the plateau region was composed of two distinct regions. Klein et al.’s study showed that at lower CPP values, within the plateau, the stability of CBF was attributable to the constriction of both small (< 70 µm) and large arterioles (> 70 µm), producing an initial hump in the plot of CBF versus CPP. At higher CPP values, within the plateau, stability in CBF was attributable to the constriction of larger arterioles while smaller arterioles were passively dilated, producing a second hump in the plateau region [31]. In the present study, these features are seen in the plateau regions of all cohort plots. In this cohort of critically ill TBI patients admitted to ICU, where oxygenation and hemoglobin concentrations are relatively stable, it is conceivable that PbtO2 primarily reflects changes in CBF attributable to changes in CPP. This may represent the first ever in human evidence of this biphasic plateau region, potentially produced by the action of small and large arterioles distinctly.

The plateau is truncated when cerebrovascular pressure reactivity is disrupted. This reinforces the notion that PRx is a valid measure of cerebral autoregulation in the traumatically injured brain. These results also indicate that PRx and cerebrovascular pressure reactivity PRx-based ΔCPPopt, LLR and ULR seem primarily sensitive to reactivity in larger arterioles. This can be seen in Fig. 1B where the morphology of the plateau hump associated with the isolated action of large arterioles is notably different when cerebrovascular pressure reactivity is disrupted. Additionally, in Fig. 2A the CPPopt (i.e., where ΔCPPopt equals zero) and LLR and ULR seem to be centered around the plateau region attributable to the action of large arterioles. This is not surprising as PRx relies on ICP as a surrogate for cerebral blood volume, which is impacted substantially more by the dilation and constriction of large arterioles than small ones [14].

Finally, the plateau region is associated with PbtO2 values of 25 mmHg to 30 mmHg. Current guideline-based management of PbtO2 in critically ill TBI patients primarily targets a PbtO2 greater than 20 mmHg [9,10,11,12]. The findings of this study point to a physiologically stable region of PbtO2 higher than currently targeted. Over this cohort, when CPP was found to be between the LLR and ULR, PbtO2 values were largely found to be within this range indicating that CPPopt-based management may be a suitable alternative when PbtO2 monitoring is unavailable.

Functional outcome models

Neither PbtO2-based or CPPopt-based logistic regression models were found to be statistically superior to the TBI-IMPACT model at predicting functional outcomes. The significant proportion of missing outcome data in this small cohort means that this study was underpowered to identify a statistical difference. No clear trend was identified when comparing models that incorporated PbtO2 and those that incorporated CPPopt-based parameters. However, it should be noted that since PbtO2 was actively managed in some institutions, the prognostic effect of PbtO2 might have been artificially diminished.

Limitations

While this study has significant strengths, such as its use of high-resolution physiologic data collected over multiple geographically distinct regions, there are notable limitations. As an observational study, confounding unmeasured physiologic parameters are possible such as heterogeneity in management strategies. Secondly, the validity of the generated functional outcome models is limited by missing data. Additionally, the yield of CPPopt, LLR, and ULR was lower than that found in previous studies, however this may just represent the real-world performance of these methods [19]. Third, given that the partial pressure of oxygen in the arterial blood (PaO2), oxygen saturation (SpO2), hemoglobin values, and various clinical management details were not available for all patients, changes in PbtO2 with changes in CPP cannot wholly be attributable to changes in CBF. This is particularly significant given that patient management likely varied between patients and between participating institutions and has not been able to be accounted for in the presented analysis. Finally, data were examined over the entire cohort, and so the validity of these relationships in individuals needs to be further examined prospectively.

Future directions

This study does provide meaningful guidance for future research. Given that PbtO2 seems to have a similar relationship to CPP as CBF in the critically ill TBI, it may serve as an appropriate surrogate for microvascular CBF in this population. This will need to be validated in prospective studies in both humans and large animal models that incorporate PaO2, SpO2, and hemoglobin measurements.

The observation that a physiologic region of stability exists with PbtO2 values between 25 and 30 mmHg also warrants further exploration. While RCTs are currently underway to determine the utility of PbtO2 monitoring, current consensus-based guidelines and study protocols have centered around a treatment threshold of 20 mmHg [9,10,11,12]. The findings of this study indicate that more work is needed to identify an optimal target range for PbtO2.

The cerebrovascular pressure reactivity findings also need to be explored. Large animal models may help better understand the degree to which PRx favors the vascular reactivity of large arterioles over small ones. Additionally, the role of CPPopt in prognostication and personalized targeted therapy must also be explored as in this study, when CPP is near CPPopt, values of PbtO2 are reassuring. One major hurdle that will need to be overcome is the suboptimal yield of CPPopt derivation as well as its related parameters. While the multi-window technique utilized in this study represents an ongoing evolution of these algorithms, further improvements may aid its adoption as a tool to guide clinical management.

Conclusion

In this exploratory analysis of a multi-institution high-resolution physiology TBI database, PbtO2 seems to have a triphasic relationship with CPP and ΔCPPopt that resembles a traditional, population-based, Lassen curve over the entire cohort [30]. The morphology of the plateau region is similar to that found in a recent large animal study evaluating the relationship between intermittent CBF and CPP, which may provide insights into the physiologic mechanisms that are involved [31]. In critically ill TBI patients admitted to ICU, PbtO2 may serve as a surrogate for CBF. The plateau region is associated with PbtO2 values of 25 mmHg to 30 mmHg which may help inform future PbtO2-based management. Notably, the CPP range over which the plateau exists is modified by the state of cerebrovascular pressure reactivity. Finally, there is some early evidence that CPPopt has potential as an alternative when PbtO2 monitoring is not available.

Availability of data and materials

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Abbreviations

ABP:

Arterial blood pressure

AIC:

Akaike information criterion

AUC-ROC:

Area under the curve of the receiver operator characteristic

BTF:

Brain trauma foundation

CAHR-TBI:

CAnadian High Resolution-TBI Research Collaborative

CBF:

Cerebral blood flow

CPP:

Cerebral perfusion pressure

CPPopt:

Optimal cerebral perfusion pressure determined by PRx

CSV:

Comma-separated value

CT:

Computerized tomography

ΔCPPopt:

Difference between CPP and CPPopt

GCS:

Glasgow Coma Score

GOS:

Glasgow Outcome Score

ICM + :

Intensive care monitoring software

ICP:

Intracranial pressure

ICU:

Intensive care units

IQR:

Interquartile range

LLR:

Lower limit of reactivity

N/A:

Not available

PaO2 :

Partial pressure of oxygen in the arterial blood

PbtO2 :

Brain tissue oxygen tension

PRx:

Pressure reactivity index

RCT:

Randomized control trial

SpO2 :

Oxygen saturation

TBI:

Traumatic brain injury

TBI-IMPACT:

International Mission for Prognosis and Analysis of Clinical Trials in TBI

ULR:

Upper limit of reactivity

References

  1. Maas AI, Stocchetti N, Bullock R (2008) Moderate and severe traumatic brain injury in adults. Lancet Neurol 7:728–741. https://doi.org/10.1016/S1474-4422(08)70164-9

    Article  Google Scholar 

  2. Carney N, Totten AM, O’Reilly C et al (2017) Guidelines for the management of severe traumatic brain injury, fourth edition. Neurosurgery 80:6–15. https://doi.org/10.1227/NEU.0000000000001432

    Article  Google Scholar 

  3. Donnelly J, Czosnyka M, Adams H et al (2019) Twenty-five years of intracranial pressure monitoring after severe traumatic brain injury: a retrospective, single-center analysis. Neurosurgery 85:E75–E82. https://doi.org/10.1093/neuros/nyy468

    Article  Google Scholar 

  4. Le Roux P, Menon DK, Citerio G et al (2014) Consensus summary statement of the international multidisciplinary consensus conference on multimodality monitoring in neurocritical care: a statement for healthcare professionals from the Neurocritical Care Society and the European Society of Intensive Care Medicine. Neurocrit Care 21:1–26. https://doi.org/10.1007/s12028-014-0041-5

    Article  Google Scholar 

  5. Casault C, Couillard P, Kromm J et al (2022) Multimodal brain monitoring following traumatic brain injury: a primer for intensive care practitioners. J Intensive Care Soc 23:191–202. https://doi.org/10.1177/1751143720980273

    Article  Google Scholar 

  6. Zeiler FA, Ercole A, Czosnyka M et al (2020) Continuous cerebrovascular reactivity monitoring in moderate/severe traumatic brain injury: a narrative review of advances in neurocritical care. Br J Anaesth 124:440–453. https://doi.org/10.1016/j.bja.2019.11.031

    Article  Google Scholar 

  7. Hays LMC, Udy A, Adamides AA et al (2022) Effects of brain tissue oxygen (PbtO2) guided management on patient outcomes following severe traumatic brain injury: a systematic review and meta-analysis. J Clin Neurosci 99:349–358. https://doi.org/10.1016/j.jocn.2022.03.017

    Article  CAS  Google Scholar 

  8. Stewart C, Haitsma I, Zador Z et al (2008) The new Licox combined brain tissue oxygen and brain temperature monitor: assessment of in vitro accuracy and clinical experience in severe traumatic brain injury. Neurosurgery 63:1159–1164. https://doi.org/10.1227/01.NEU.0000333265.19131.7C. (discussion 1164–1165)

    Article  Google Scholar 

  9. Okonkwo DO, Shutter LA, Moore C et al (2017) Brain oxygen optimization in severe traumatic brain injury phase-ii: a phase ii randomized trial. Crit Care Med 45:1907–1914. https://doi.org/10.1097/CCM.0000000000002619

    Article  Google Scholar 

  10. Chesnut R, Aguilera S, Buki A et al (2020) A management algorithm for adult patients with both brain oxygen and intracranial pressure monitoring: the Seattle International Severe Traumatic Brain Injury Consensus Conference (SIBICC). Intensive Care Med 46:919–929. https://doi.org/10.1007/s00134-019-05900-x

    Article  Google Scholar 

  11. Bernard F, Barsan W, Diaz-Arrastia R et al (2022) Brain oxygen optimization in severe traumatic brain injury (BOOST-3): a multicentre, randomised, blinded-endpoint, comparative effectiveness study of brain tissue oxygen and intracranial pressure monitoring versus intracranial pressure alone. BMJ Open 12:e060188. https://doi.org/10.1136/bmjopen-2021-060188

    Article  Google Scholar 

  12. ACTRN12619001328167 (2019) The BONANZA trial- a randomised controlled trial that is testing whether a management strategy guided by early brain tissue oxygen monitoring in patients in with severe traumatic brain injury improves long term neurological and functional outcomes. https://trialsearch.who.int/Trial2.aspx?TrialID=ACTRN12619001328167. Accessed 31 Oct 2022

  13. Payen J-F, Richard M, Francony G et al (2020) Comparison of strategies for monitoring and treating patients at the early phase of severe traumatic brain injury: the multicentre randomised controlled OXY-TC trial study protocol. BMJ Open 10:e040550. https://doi.org/10.1136/bmjopen-2020-040550

    Article  Google Scholar 

  14. Czosnyka M, Smielewski P, Kirkpatrick P et al (1997) Continuous assessment of the cerebral vasomotor reactivity in head injury. Neurosurgery 41:11–19. https://doi.org/10.1097/00006123-199707000-00005

    Article  CAS  Google Scholar 

  15. Zeiler FA, Ercole A, Cabeleira M et al (2019) Univariate comparison of performance of different cerebrovascular reactivity indices for outcome association in adult TBI: a CENTER-TBI study. Acta Neurochir 161:1217–1227. https://doi.org/10.1007/s00701-019-03844-1

    Article  Google Scholar 

  16. Sorrentino E, Diedler J, Kasprowicz M et al (2012) Critical thresholds for cerebrovascular reactivity after traumatic brain injury. Neurocrit Care 16:258–266. https://doi.org/10.1007/s12028-011-9630-8

    Article  CAS  Google Scholar 

  17. Steiner LA, Czosnyka M, Piechnik SK et al (2002) Continuous monitoring of cerebrovascular pressure reactivity allows determination of optimal cerebral perfusion pressure in patients with traumatic brain injury. Crit Care Med 30:733–738. https://doi.org/10.1097/00003246-200204000-00002

    Article  Google Scholar 

  18. Aries MJH, Czosnyka M, Budohoski KP et al (2012) Continuous determination of optimal cerebral perfusion pressure in traumatic brain injury. Crit Care Med 40:2456–2463. https://doi.org/10.1097/CCM.0b013e3182514eb6

    Article  Google Scholar 

  19. Liu X, Maurits NM, Aries MJH et al (2017) Monitoring of optimal cerebral perfusion pressure in traumatic brain injured patients using a multi-window weighting algorithm. J Neurotrauma 34:3081–3088. https://doi.org/10.1089/neu.2017.5003

    Article  Google Scholar 

  20. Needham E, McFadyen C, Newcombe V et al (2017) Cerebral perfusion pressure targets individualized to pressure-reactivity index in moderate to severe traumatic brain injury: a systematic review. J Neurotrauma 34:963–970. https://doi.org/10.1089/neu.2016.4450

    Article  Google Scholar 

  21. Tas J, Beqiri E, van Kaam RC et al (2021) Targeting autoregulation-guided cerebral perfusion pressure after traumatic brain injury (COGiTATE): a feasibility randomized controlled clinical trial. J Neurotrauma 38:2790–2800. https://doi.org/10.1089/neu.2021.0197

    Article  Google Scholar 

  22. Beqiri E, Brady KM, Lee JK et al (2021) Lower limit of reactivity assessed with PRx in an experimental setting. Acta Neurochir Suppl 131:275–278. https://doi.org/10.1007/978-3-030-59436-7_51

    Article  Google Scholar 

  23. Donnelly J, Güiza F, Depreitere B et al (2021) Visualising the pressure-time burden of elevated intracranial pressure after severe traumatic brain injury: a retrospective confirmatory study. Br J Anaesth 126:e15–e17. https://doi.org/10.1016/j.bja.2020.09.018

    Article  Google Scholar 

  24. Donnelly J, Zeiler FA, Güiza F et al (2021) Visualization of intracranial pressure insults after severe traumatic brain injury: influence of individualized limits of reactivity. Acta Neurochir Suppl 131:7–10. https://doi.org/10.1007/978-3-030-59436-7_2

    Article  Google Scholar 

  25. Donnelly J, Czosnyka M, Adams H et al (2017) Individualizing thresholds of cerebral perfusion pressure using estimated limits of autoregulation. Crit Care Med 45:1464–1471. https://doi.org/10.1097/CCM.0000000000002575

    Article  Google Scholar 

  26. Jaeger M, Dengl M, Meixensberger J, Schuhmann MU (2010) Effects of cerebrovascular pressure reactivity-guided optimization of cerebral perfusion pressure on brain tissue oxygenation after traumatic brain injury. Crit Care Med 38:1343–1347. https://doi.org/10.1097/CCM.0b013e3181d45530

    Article  Google Scholar 

  27. Bernard F, Gallagher C, Griesdale D et al (2020) The CAnadian high-resolution traumatic brain injury (CAHR-TBI) research collaborative. Can J Neurol Sci 47:551–556. https://doi.org/10.1017/cjn.2020.54

    Article  Google Scholar 

  28. Batson C, Froese L, Sekhon M et al (2022) Impact of chronological age and biological sex on cerebrovascular reactivity in moderate/severe traumatic brain injury: a CAnadian high-resolution TBI (CAHR-TBI) study. J Neurotrauma. https://doi.org/10.1089/neu.2022.0293

    Article  Google Scholar 

  29. Maas AIR, Marmarou A, Murray GD et al (2007) Prognosis and clinical trial design in traumatic brain injury: the IMPACT study. J Neurotrauma 24:232–238. https://doi.org/10.1089/neu.2006.0024

    Article  Google Scholar 

  30. Lassen NA (1959) Cerebral blood flow and oxygen consumption in man. Physiol Rev 39:183–238. https://doi.org/10.1152/physrev.1959.39.2.183

    Article  CAS  Google Scholar 

  31. Klein SP, De Sloovere V, Meyfroidt G, Depreitere B (2022) Differential hemodynamic response of pial arterioles contributes to a quadriphasic cerebral autoregulation physiology. JAHA 11:e022943. https://doi.org/10.1161/JAHA.121.022943

    Article  Google Scholar 

Download references

Acknowledgements

FAZ receives research support from NSERC, CIHR, the MPI Neuroscience Research Operating Fund, the Health Sciences Centre Foundation Winnipeg, the Canada Foundation for Innovation (CFI) (Project #: 38583), Research Manitoba (Grant #: 3906), the University of Manitoba VPRI Research Investment Fund (RIF), and the University of Manitoba MPI Professorship in Neuroscience. AG is supported through the University of Manitoba Clinician Investigator Program, the University of Manitoba Dean’s Fellowship, the Manitoba Medical Services Foundation Research and Education Fellowship, the R. Samuel McLaughlin Research Fellowship, and a Canadian Institutes of Health Research (CIHR) Fellowship (Grant #: 472286). LF is supported through the University of Manitoba-Biomedical Engineering (BME) Fellowship Grant, Research Manitoba–Health Sciences PhD Studentship, NSERC (ALLRP-576386-22) and the University of Manitoba Graduate Enhancement of Tri-Agency Stipend (GETS) program. EPT acknowledges funding support from Strategic Research Area Neuroscience (StratNeuro, Karolinska Institutet), The Erling-Persson Family Foundation and Region Stockholm Clinical Research Appointment (ALF Klinisk Forskare, FoUI-955376). RR acknowledges funding support from Svenska Kulturfonden, Medicinska Understödsföreningen Liv & Hälsa and Finska Läkaresällskapet.

Funding

This work was directly supported through the Natural Sciences and Engineering Research Council of Canada (NSERC) (DGECR-2022-00260, RGPIN-2022-03621 and ALLRP-576386-22) and the Manitoba Public Insurance (MPI) Neuroscience Research Operating Fund.

Author information

Author notes

  1. Frederick A. Zeiler

    Present address: Section of Neurosurgery, Department of Surgery, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, MB, Canada

  2. Mypinder Sekhon

    Present address: Division of Critical Care, Department of Medicine, University of British Columbia, Vancouver, BC, Canada

  3. Andreas H. Kramer and Frederick A. Zeiler are co-senior authors on this manuscript.

Authors and Affiliations

  1. Department of Human Anatomy and Cell Science, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, Canada

    Alwyn Gomez & Frederick A. Zeiler

  2. Section of Neurosurgery, Department of Surgery, Rady Faculty of Health Sciences, University of Manitoba, Winnipeg, MB, Canada

    Alwyn Gomez

  3. Biomedical Engineering, Faculty of Engineering, University of Manitoba, Winnipeg, MB, Canada

    Logan Froese & Frederick A. Zeiler

  4. Department of Anesthesiology, Pharmacology & Therapeutics, University of British Columbia, Vancouver, BC, Canada

    Mypinder Sekhon, Donald Griesdale & Eleen Yang

  5. Department of Neurology, Karolinska University Hospital, Stockholm, Sweden

    Eric P. Thelin

  6. Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden

    Eric P. Thelin & Frederick A. Zeiler

  7. Department of Neurosurgery, University of Helsinki and Helsinki University Hospital, Helsinki, Finland

    Rahul Raj

  8. Department of Intensive Care, Maastricht University Medical Center, Maastricht, The Netherlands

    Marcel Aries

  9. Section of Neurosurgery, Department of Clinical Neurosciences, University of Calgary, Calgary, Canada

    Clare Gallagher

  10. Section of Critical Care, Department of Medicine, University of Montreal, Montreal, QC, Canada

    Francis Bernard

  11. Department of Critical Care Medicine, University of Calgary, Calgary, Canada

    Andreas H. Kramer

  12. Centre On Aging, University of Manitoba, Winnipeg, Canada

    Frederick A. Zeiler

  13. Division of Anaesthesia, Department of Medicine, Addenbrooke’s Hospital, University of Cambridge, Cambridge, UK

    Frederick A. Zeiler

  14. Department of Clinical Neurosciences, University of Calgary, Calgary, Canada

    Clare Gallagher & Andreas H. Kramer

  15. Hotchkiss Brain Institute, University of Calgary, Calgary, Canada

    Clare Gallagher & Andreas H. Kramer

  16. School of Mental Health and Neurosciences, University Maastricht, Maastricht, The Netherlands

    Marcel Aries

Authors
  1. Alwyn Gomez
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Mypinder Sekhon
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Donald Griesdale
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Logan Froese
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Eleen Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Eric P. Thelin
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Rahul Raj
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Marcel Aries
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Clare Gallagher
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Francis Bernard
    View author publications

    You can also search for this author in PubMed Google Scholar

  11. Andreas H. Kramer
    View author publications

    You can also search for this author in PubMed Google Scholar

  12. Frederick A. Zeiler
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

AG aided in study conception, data collection, preparation, and analysis as well as preparation and editing of the manuscript. MS aided in data collection and editing of the manuscript. DG aided in data collection and editing of the manuscript. LF aided in data collection and analysis as well as editing of the manuscript. EY aided in data collection and editing of the manuscript. EPT aided in editing of the manuscript. RR aided in editing of the manuscript. MA aided in editing of the manuscript. CG aided in data collection and editing of the manuscript. FB aided in editing of the manuscript. AHK aided in data collection and editing of the manuscript. FAZ aided in study conception, data collection, and editing of the manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Alwyn Gomez.

Ethics declarations

Ethics approval and consent to participate

Local research ethics approval at the University of Manitoba has been obtained for all aspects of this database (H2017:181 and H2017:188). Similarly, ethics approval was obtained for retrospective access to the database and for anonymous data transfer with each center for this project (H2020:118, H20-03759 and REB20-0482).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Appendix SA.

Plots of PbtO2 versus CPP for individual patients. Appendix SB. Plots of PbtO2 versus ΔCPPopt for individual patients.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gomez, A., Sekhon, M., Griesdale, D. et al. Cerebrovascular pressure reactivity and brain tissue oxygen monitoring provide complementary information regarding the lower and upper limits of cerebral blood flow control in traumatic brain injury: a CAnadian High Resolution-TBI (CAHR-TBI) cohort study. ICMx 10, 54 (2022). https://doi.org/10.1186/s40635-022-00482-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s40635-022-00482-3

Keywords