Special article

Journal of the Intensive Care Society

2020, Vol. 0(0), 1–12
Multimodal brain monitoring following ! The Intensive Care Society 2020
Article reuse guidelines:
traumatic brain injury: A primer for sagepub.com/journals-permissions
DOI: 10.1177/1751143720980273
intensive care practitioners journals.sagepub.com/home/jics

Colin Casault1 , Philippe Couillard1,2, Julie Kromm1,2,

Eric Rosenthal3, Andreas Kramer1,2 and Peter Brindley4

Abstract
Traumatic brain injury (TBI) is common and potentially devastating. Traditional examination-based patient monitoring
following TBI may be inadequate for frontline clinicians to reduce secondary brain injury through individualized therapy.
Multimodal neurologic monitoring (MMM) offers great potential for detecting early injury and improving outcomes. By
assessing cerebral oxygenation, autoregulation and metabolism, clinicians may be able to understand neurophysiology
during acute brain injury, and offer therapies better suited to each patient and each stage of injury.
Hence, we offer this primer on brain tissue oxygen monitoring, pressure reactivity index monitoring and cerebral
microdialysis. This narrative review serves as an introductory guide to the latest clinically-relevant evidence regarding
key neuromonitoring techniques.

Keywords
Brain tissue oxygen monitoring, pressure reactivity index, cerebral microdialysis, multimodality monitoring, traumatic
brain injury, neurocritical care

Introduction
function. For this review, we will provide a practical
With 27 million new cases and 8.1 million years lived primer with a focus on techniques for monitoring
with disability in 2016, traumatic brain injury(TBI) cerebral oxygenation, autoregulation, and
represents the leading cause of global disability.1–3 metabolism.
In the intensive care unit (ICU), TBI treatment focus-
es on preventing and treating secondary brain injury.
This has traditionally meant monitoring and mitigat-
Cerebral oxygenation monitoring
ing physiologic derangements in blood pressure, For at least three decades, clinicians have had the
oxygen saturation, the partial pressure of carbon technology to interrogate cerebral oxygen delivery
dioxide (CO2), and intracranial pressure (ICP). and utilization. Available tools include jugular bulb
In isolation, the clinical examination may not be mixed venous oximetry, near infrared spectroscopy,
sufficient to detect events that have been associated oxygen-15 positron emission tomography (PET) and
with worse outcomes, such as elevations in ICP, brain tissue oxygen tension monitoring (PbTO2). This
reduced cerebral oxygen availability, or subclinical review focuses on PbTO2 monitoring, as it is extensive-
seizures. Moreover, many of the treatments used in ly researched thus far and is being further evaluated
TBI management may further confound the clinical in ongoing randomized controlled trials (RCTs).
examination, such as analgesics, sedation, and phar-
macologic paralysis. Advanced cerebral monitoring 1
Department of Critical Care Medicine, University of Calgary, Calgary,
tools may be used to detect evolving secondary Canada
injury before, rather than during or after, overt neu- 2
Department of Clinical Neurosciences, University of Calgary, Calgary,
rologic deterioration (Figure 1).4,5 Collectively, these Canada
3
tools are referred to as multimodality monitoring Department of Critical Care Medicine, University of Alberta,
(MMM). Broadly, MMM involves the use of multiple Edmonton, Canada
4
Department of Neurology, Harvard University, Boston, MA, USA
techniques to measure and integrate information
Corresponding author:
regarding cerebral blood flow (CBF), oxygenation Colin Casault, Department of Critical Care Medicine, University of
(i.e., measuring the balance between delivery and Calgary, 2500 University Dr NW Calgary, AB T2N-1N4, Canada.
use), ICP, autoregulation, metabolism, and cortical Email: [email protected]
2 Journal of the Intensive Care Society 0(0)

Figure 1. Hypothetical Model for the implementation of MMM in traumatic brain injury.

PbTO2 monitors use a Clark electrode (Licox considers both 15 or 20 mmHg as acceptable, while
Integra Monitor, Integra Neurosciences, Plainsboro, the Brain Trauma Foundation (BTF) does not pro-
NJ, USA), or an oxygen quenching method vide specific guidance in this area.10 In on-going clin-
(Raumedic Neurovent-PTO, Raumedic, Münchberg, ical trials such as BOOST III and BONANZA, the
Germany), to determine the partial pressure of PbTO2 interventional threshold is 20 mmHg.11,12
oxygen in brain parenchyma. The Clark electrode Accordingly, many experts recommend a PbTO2
consists of a cathode and anode, surrounded by a target >20 mmHg following TBI, however regional
thin membrane, through which oxygen diffuses from variation may be observed.13
a 15 mm radius. Dissolved oxygen is reduced at the Following TBI, cerebral hypoxemia may be more
cathode, thereby allowing the sensor to estimate the common than previously thought. Using fluorine
partial pressure in the surrounding brain parenchy- PET, TBI patients have demonstrated diffuse cerebral
ma. The 0.5 mm probe is usually placed in frontal ischemia distant from traumatized tissue in seemingly
lobe white matter and requires approximately 30 to healthy parenchyma.14 Using the threshold of a
60 minutes for calibration before producing reliable PbTO2 value <20 mmHg, hypoxemia occurred in, at
readings. Catheters also require recalibration daily least, half of patients (50–86%) in the first few days
via O2 and CO2 reactivity testing.6 post injury.15–17 Importantly, episodes occurred
When using a PbTO2 monitor, probe location mat- despite normal ICP or cerebral perfusion pressure
ters, as it affects interpretation. Although still debat- (CPP), suggesting cerebral hypoxemia would be
ed, monitors are usually placed in the unaffected missed with traditional neuromonitoring techni-
hemisphere. The justification is that this enables ques.18 In short, cerebral hypoxemia would have
detection and mitigation of cerebral hypoxemia in been missed without MMM.
tissue that is comparatively healthy and salvageable. The presence of cerebral hypoxemia is important
In contrast, proponents of placing the probe in (or as it may be impactful to patient outcomes. Low
near) damaged tissue argue that this could preserve PbTO2 values are associated with higher patient mor-
at risk or “penumbral” tissue. The concern is these tality,19–21 and the detrimental effect appears to be
tissues may already be less responsive to therapy.7 In “dose related”, namely worsened by duration and
the case of diffuse injury, the probe is usually placed degree.18 In cohort studies, those TBI patients who
in the non-dominant hemisphere for practical and spent more time with low PbTO2 had worse Glasgow
safety reasons to minimize the risk of harm in the Outcome Scores exams (GOSE) and lower functional
case of iatrogenic hemorrhage. status at 30 days and 6 months.16,20–22 Furthermore,
“Normal” PbTO2 values are usually >23 þ/ patients who are not responsive to PbTO2 directed
7 mmHg.6 However, there is no universally accepted therapy- i.e. patients whose cerebral hypoxemia is
threshold at which tissue damage occurs. PbTO2 irreversible- are less likely to survive. In a prospective
values of <15 mmHg are associated with a widening observational cohort study, survivors had a 71%
lactate/glucose ratio and increasing glycerol levels, response rate to PbTO2 directed treatment compared
both of which suggest cellular energetic failure.8,9 to 44% in non-survivors.23
Consistent international consensus guidelines provide While PbTO2 directed therapy shows promise for
inconsistent direction on the most optimal PbTO2 ameliorating cerebral hypoxemia, choosing which
target. Consequently, the neurocritical care society therapy to use is complex. There are four components
Casault et al. 3

Figure 2. Determinants of cerebral oxygenation.

to consider when treating a low PbTO2 value: oxygen (PbTO2 < 20 mmHg)26 compared to those in the ICP
delivery, diffusion, demand and utilization (Figure 2). only group. Moreover, there was a non-significant
This, in turn, means concomitantly addressing ICP, trend towards improved outcomes (although, impor-
CPP, systemic oxygenation, hemoglobin, tempera- tantly, the BOOST-2 trial was not powered for this
ture, sedation, shivering, the partial pressure of outcome).26 As a result, clinicians are eagerly await-
carbon dioxide and seizures (See Figure 3). Pascual ing results from three RCTs using combined PbTO2-
et al. found the majority of patients require only a ICP directed therapy in severe TBI.11,12,27
single intervention to normalize PbTO2 values.
However, no individual intervention will optimize
Monitoring cerebral autoregulation
all low PbTO2 values.24 The SIBICC guidelines pro-
vide a tiered approach to managing patients with ICP The first description of cerebrovascular reactivity is
and PbTO2 monitoring.13 This strategy divides credited to Neils Lassen, who, in 1959, established
patients into four groups based on dichotomized that CBF remains nearly constant across a wide
ICP and PbTO2 variables. The groups are then range of CPP.28 Cerebral vascular reactivity is depen-
labelled by type including Type A (normal ICP/ dent on multiple contributors, including myogenic
PbTO2), Type B (elevated ICP but normal PbTO2), (e.g. changes in transmural pressure), neurogenic
Type C (Normal ICP and low PbTO2) and Type D (e.g. acetylcholine, serotonin, etc.), endothelial (e.g.
(elevated ICP and low PbTO2).25 While large RCTs nitric oxide, thromboxane, etc.) and metabolic (e.g.
are on-going, the SIBICC guidelines provide level pH, CO2, etc.) factors.29 Changes in these variables
III evidence to guide and standardize PbTO2 therapy. causes cerebral arterial dilation or constriction which,
Overall, there is evolving evidence to support mon- when functional, maintains a constant CBF. The
itoring brain tissue oxygenation, however RCTs are myogenic component, referred from here on as cere-
still needed to demonstrate treatment of PbTO2 posi- bral autoregulation, refers to the contraction or dila-
tively impacts outcomes in TBI. BOOST-2 was a pilot tion of cerebral blood vessels in response to changes
RCT that demonstrated protocolized PbTO2 & ICP in transmural pressure.29 When cerebral autoregula-
directed therapy compared to an ICP directed treat- tion is intact, cerebral arterioles constrict or dilate in
ment alone was both feasible and safe. Specifically, response to high or low mean arterial pressure
those in the PbTO2 & ICP directed therapy arm spent (MAP), respectively, which ensures a near-constant
less time with cerebral hypoxemia CBF across a wide range of CPP (Figure 4). In the
4 Journal of the Intensive Care Society 0(0)

Figure 3. Algorithm for the management of cerebral hypoxemia.

setting of brain injury, cerebral autoregulation may autoregulatory status of the patient. When cerebro-
become impaired. Autoregulatory exhaustion may vascular autoregulation is intact, rapid changes in
predispose patients to periods of cerebral ischemia, transmural pressure from increased CPP cause cere-
secondary to inappropriately low CBF, or cerebral bral arterial constriction, which leads to a reduction
hyperemia, due to abnormally high CBF, even at in cerebral blood volume and is detected by a subse-
guideline-recommended MAP, ICP, and CPP tar- quent decrease in ICP. Mathematically, this relation-
gets.30,31 Therefore, monitoring cerebral autoregula- ship is represented as a negative PRx value.
tion allows clinicians to individualize CPP targets to Alternatively, if autoregulation is exhausted, increases
avoid further secondary injury from cerebral ischemia in CPP cause passive arterial dilation rather than con-
or hyperemia. striction as transmural pressure increases.
By measuring changes in ICP in relation to Consequently, there is an increase in cerebral blood
changes in CPP over time, a Pearson’s moving corre- volume and is detected as an increase in ICP.
lation coefficient can be used to mathematically quan- Accordingly, this relationship is demonstrated as a
tify this relationship, known as the Pressure positive PRx value. Autoregulatory failure is typically
Reactivity Index (PRx), which represents the described by a PRx value of >0.2–0.3; however,
Casault et al. 5

Figure 4. Relationships between CBF, CPP and ICP with intact and disturbed autoregulation.

failure is more likely a continuum between 0 and patient’s autoregulatory function across a spectrum
þ0.35 rather than a binary threshold.32 It is during of CPP values and the subsequent calculation of an
this state clinicians should be particularly concerned individualized “optimal” target CPP (CPPopt),
about worsening secondary injury from ischemia or namely the CPP range with the lowest PRx value.34
hyperemia as the brain may receive insufficient or Software systems, such as ICMþ(Cambridge, UK),
excessive CBF in response to an ordinarily physiolog- can graph average PRx against CPP in real-time.
ic CPP. Because PRx is determined from ICP & For those without specialized software, cerebral
MAP, it can be sampled with an extremely high fre- autoregulatory status can be determined by observing
quency, providing minute to minute monitoring the ICP response to a bedside blood pressure chal-
information and represents a continuous autoregula- lenge.35 Experts recommend increasing MAP by
tion monitoring method compared to a static method 10 mmHg, using vasopressors, while assessing the
like transcranial doppler which represents only a response in ICP for 10–20 minutes. Generally, one
single point estimate at the time of the study. should not target a CPP of less than 50 or greater
Continuous monitoring of cerebral autoregulation than 90 mmHg for concern of precipitating iatrogenic
using PRx provides a distinct advantage compared cerebral ischemia or hyperemia.35
to static methods as changes in pH, sedation level, When using a PRx curve to determine CPPopt, four
CO2, temperature and many other factors may patterns are commonly observed; a U-shaped,
impact cerebrovascular reactivity dynamically. ascending, descending, or absent curve (Figure 5).
Aries et al. collected PRx values over time, aver- An ascending or descending autoregulatory profile
aged over five seconds (or longer) and graphed the suggests the clinician should consider a lower or
PRx against the CPP for periods of four hours.33 This higher MAP, respectively (Figure 6). Using four-
method provides an estimate of the individual hour epochs, CPPopt can be reliably determined in
6 Journal of the Intensive Care Society 0(0)

Figure 5. Common phenotypes of PRx Derived autoregulatory curves in severe TBI. (a) U-shaped curve phenotype where CPPopt
is identified at 75-80 mmHg. (b) An Ascending curve demonstrating worsening autoregulatory status as CPP is increased, therefore
the CPPopt is found at the lowest CPP with reasonable sampling 65-70 mmHg. (c) Descending curve pattern where autoregulatory
function is optimal at a higher CPP than BTF guidelines (90-95 mmHg) (d) No clear relationship can be discerned by the CPP sample.
Figure reproduced with the expressed written consent of Springer Publishing Group.

just over half (55%) of the recording time.33 To more outcome. Those patients who spent time below
consistently determine CPPopt, a multi-windowing CPPopt or had a more significant discrepancy between
method has recently been validated. This generates CPPopt and CPPactual, had worse neurologic out-
a mean estimation of CPPopt based on data from mul- comes, particularly if they had disturbed autoregula-
tiple PRx calculation windows, using epochs of tion.39 Therefore, it is possible deviation from CPPopt
2–8 hours.36,37 may worsen secondary brain injury by the previously
Importantly, the validity of a PRx value at a target mentioned mechanisms. Work by Mathieu et al. sup-
CPP is dependent on how long the patient spent in ports this hypothesis. TBI patients who spent more
that particular CPP range. If a patient spent a small time with a PRx > 0.25 had worse cerebral edema on
proportion(e.g. <5%) of monitored time at a partic- imaging.40 Because patients with disturbed autoregu-
ular CPP, then clinicians may choose to downplay the lation appear to be at higher risk of secondary brain
usefulness of that PRx value, because of insufficient injury, PRx can help predict neurologic outcome fol-
sampling. The period of time required to collect a lowing TBI.41
“definitively valid” sample is unclear, however, it is Despite PRx’s promise, there are important limita-
obvious that longer sampling times provide a more tions. So far, the literature is largely from observa-
reliable understanding of a patient’s autoregulatory tional studies. Therefore, it is unclear whether PRx
status at that CPP. defined autoregulatory failure is modifiable.42
International guidelines have largely not acknowl- Autoregulatory failure could be triggered by the
edged the inter-individual differences in autoregula- acute injury and result in profound physiologic
tion or its importance at the time of this writing. The derangements that may be irreversible.43
BTF guidelines recommend a CPP of 60–70mmHg Fortunately, work by Zeiler et al. suggests otherwise.
after TBI, but acknowledge this may fail to account For example, pharmacologic sedation may affect the
for patient differences.38 Regardless, deviation from amount of time patients spend with a PRx > 0.44
individualized CPPopt appears to correlate with While insufficiently studied, it seems intuitive that
Casault et al. 7

Figure 6. Proposed model for the use of PRx derived CPPopt in severe TBI.

manipulating sedation could alter autoregulation by clinical use in 1989.46 CMD uses a semi-permeable
reducing cerebral metabolic rate and in turn, CBV. catheter, which is often combined with other cerebral
Another limitation of PRx is related to its acquisi- monitors such as PbTO2, and ICP. Placed in the brain
tion. Determining CPPopt comes from understanding parenchyma, perfusate is instilled through a 10 mm
how PRx changes across a CPP range and typically catheter at a rate of 0.3 uL/min. This facilitates
requires 4–6 hours of data sampling. This limitation hourly sampling of small water-soluble molecules
means real time MAP adjustment may not be feasible like lactate, pyruvate, glucose, glutamate or glycerol,
as reported values reflect autoregulation from the pre- as well as cytokines and larger proteins with the
vious 4–6 hours. To put it more simply, the moment appropriate catheter.47,48
to intervene may have already passed 4–6 hours ago Molecules equilibrate with the microdialysate fluid
during data acquisition. Another criticism is PRx and reflect the environment of the brain tissue sur-
curves require experts to interpret the data, and rounding the catheter. Microdialysate compounds are
even then, there can be interobserver variability.39 typically 20–100 kDa but equilibration depends upon
However, the use of the aforementioned multi- molecule solubility, membrane area, membrane pore
windowing method and computer identified PRx size, perfusate flow rate and perfusate choice. At a
curves may mitigate these shortcomings.37,39 Finally, rate of 0.3 uL/min, the measured molecular concen-
no RCTs have been conducted to provide an evidence trates are approximately 70%-reflective of the extra-
based CPP guided algorithm clearly improves out- cellular space.49 By reducing the dialysate flow rate,
comes. As such, feasibility studies such as recovery can be increased to 100%, but at the cost of
COGiTATE are underway.45 In the meantime, PRx sampling frequency.50 In this article, we will focus on
offers promise. the commonly studied molecules such as lactate,
pyruvate and glucose.
Glucose is the primary energy substrate of the
Cerebral microdialysis brain. During aerobic metabolism, glucose is taken
Cerebral microdialysis (CMD) monitors cerebral up by neural cells and metabolized to pyruvate,
metabolism during acute brain injury. Conceived in which, in turn, fuels the Kreb’s cycle.49,51 Following
the 1970s by Ungerstedt & Pycock, CMD entered brain injury insult, neural cells may shift towards
8 Journal of the Intensive Care Society 0(0)

anaerobic metabolism, whereby, pyruvate is con- epilepsy.78–82 Regardless, it is important to recognize

verted to lactate, which can, in turn, be measured at the evidence for the use of CMD as a therapeutic
the bedside using CMD. In isolation, absolute lactate monitoring tool remains relatively unsubstantiated.
concentration is less useful than the lactate/pyruvate At the time of this writing, ‘Neuro-glycemic based’
(L/P) ratio, as lactate may serve as an alternative neu- insulin protocols should not be advocated for outside
ronal energy source. However, a more detailed discus- of the research context. Regarding their safety profile,
sion of the lactate shuttle hypothesis is beyond the CMD catheters appear to have a comparable compli-
scope of this article.52,53 A widening L/P ratio cation risk to parenchymal monitors: approximately
(LPR) may signal an on-going cerebral energy crisis 1–3%.78,83 It would be rare for CMD to be used in
(i.e. ischemia or hypermetabolism) or physiologic dis- isolation, and they are typically used in concert with
tress.49,51,54 While it is unclear that adjusting physio- PbTO2 and ICP monitors, which may be more sensi-
logic variables to treat abnormal LPR values is tive compared to ICP monitors alone.5
consistently effective, CMD can be used in concert Outside of cerebral glycemic monitoring, CMD
with PbTO2, and ICP monitors to potentially provide may help monitor the effects of temperature on cere-
an understanding of the relationship between cerebral bral metabolism. Oddo et al. showed improvements
metabolism, cerebral oxygenation, ICP, and cerebral in cerebral metabolism following induced normother-
autoregulation.54,55 mia guided by CMD, but, notably, these gains were
Neuroglycopenia, defined by a glucose of lost when shivering occurred.75,76 It remains unclear
<0.8 mmol/L, may be caused by reduced glucose whether CMD guided cerebral temperature treatment
supply (i.e. from low cerebral blood flow, or low changes outcomes, however the dangers of cerebral
serum glucose), or increased demand (i.e. secondary hyperthermia are well recognized.84,85 CMD could
to seizures, hyperthermia or cerebral hypermetabo- also be useful in predicting ICP spikes as the LPR
lism).51,56 During TBI, the normal cerebral glycemic >25 may predict intracranial hypertension.
supply-demand relationships are disturbed. Accordingly, CMD monitoring could spur proactive,
Accordingly, neuroglycopenia has been observed in rather than reactive ICP management.86 CMD could
approximately 75% of patients with moderate- also help understand the role of brain cytokines, iden-
severe TBI at least once,57 to which ICP and CPP tify biomarkers, and help monitor intracerebral med-
monitoring were insensitive.57,58 ication levels.87–91 Ultimately, the use of CMD
Previously, clinicians strived for tight serum glu- remains largely investigative, however it does provide
cose control which was found to be harmful.59 In the potential to understand cerebral metabolism,
TBI, CMD studies highlight that, when insufficiently which confers different neurophysiologic data from
monitored, tight glycemic control can also exacerbate the other aforementioned techniques.
cerebral metabolic crisis.60,61 This is clinically impor-
tant because metabolic crisis, typically defined by a
LPR >25–40 mmol/L, has been associated with
Conclusion
increased mortality and worse neurologic outcomes MMM, specifically PbTO2, PRx and CMD, are a
at 3–6 months.54,62–67 The duration of metabolic promising group of techniques for understanding
crisis is also associated with an unfavorable neurolog- the complex neurophysiology which occurs during
ic outcome (GOSE < 6) at six months.57 Even with a severe TBI. PbTO2 and PRx have accumulated
normal serum glucose, insulin infusions can substan- enough evidence that experts advocate for goal direct-
tially deplete cerebral glucose and potentially contrib- ed therapeutic protocols targeting PbTO2 > 20 mmHg
ute to secondary injury.68 Neuroglycopenia, even in and CPPopt as part of clinical trials, while CMD
the absence of cellular distress, is associated with poor remains largely investigational. Similarly, internation-
neurologic outcomes at six months.61,69–73 Vespa al guidelines from the BTF reflect these sentiments,
et al. found in over 70% of cases of neuroglycopenia however remain less rigorously defined to reflect the
detected by CMD, there was no other physiologic unclear state of the literature base as a whole.
indicator present(i.e. no systemic hypoglycemia, In isolation, no monitoring tool is likely to change
high ICP, low CPP, PbTO2, low jugular venous outcomes, but when used as part of a goal-directed
oxygen saturation, or seizure).69 In contrast to neuro- therapeutic strategy, it is hypothetically possible to
glycopenia, systemic hyperglycemia is also associated influence outcomes.92 However, achieving this level
with poor outcome.74 Therefore, similar to CPP, ICP of evidence starts from an applied physiologic question
and PbTO2, CMD could, hypothetically, be used to and sufficient supportive cohort studies, prior to the
individualize serum glucose control to mitigate cere- orchestration of RCT pilot studies and finally large
bral hypo- or hyperglycemia. therapeutic trials. In the case of MMM, BOOST-2
CMD has been studied in TBI and SAH to guide was successful in demonstrating protocolized PbTO2
glucose control, CPP, hyperventilation, and tempera- & ICP treatment was feasible, safe and reduced the
ture control.61,73,75–78 It has also been applied, albeit duration of cerebral hypoxemia. Accordingly, RCTs
less commonly, following ischemic stroke, intracrani- including BOOST-III, BONANZA and COGiTATE
al hemorrhage, hepatic encephalopathy, and seek to substantiate the use of a protocolized treatment
Casault et al. 9

algorithms for PbTO2 & ICP directed care and CPPopt national burden of traumatic brain injury and spinal
respectively. In the author’s opinion, supportive ran- cord injury, 1990–2016: a systematic analysis for the
domized trials are of great importance. These studies Global Burden of Disease Study 2016. Lancet Neurol
seek to demonstrate, not only, the proof of concept, 2019; 18: 56–87.
4. Chen HI, Stiefel MF, Oddo M, et al. Detection of cereb
but also to recognize unintended complications sec-
global, regional, and national burden of traumatic
ondary to the therapies used (e.g. vasopressors, IV
brain injury and spinal cord injury, 1990–2016: a sys-
fluids, hyperoxia, etc).
tematic analysis for the global burden of disease study
Critical care practitioners should understand these 2016ral compromise with multimodality monitoring in
tools should be used in an integrative fashion, com- patients with subarachnoid hemorrhage. Neurosurgery
bining MMM data with the clinical examination, sys- 2011; 69: 53–63; discussion 63.
temic monitoring, neuroimaging, and additional 5. Bouzat P, Marques-Vidal P, Zerlauth JB, et al.
specialized monitoring tools, such as EEG, transcra- Accuracy of brain multimodal monitoring to detect
nial doppler, etc., to enhance and individualize cerebral hypoperfusion after traumatic brain injury*.
patient care. Clinicians do not merely want more Crit Care Med Feb 2015; 43: 445–452.
data; we want information with which to monitor 6. Pennings FA, Schuurman PR, van den Munckhof P,
and individualize therapy meaningfully. MMM has et al. Brain tissue oxygen pressure monitoring in
not yet been shown to improve patient outcomes awake patients during functional neurosurgery: the
assessment of normal values. J Neurotrauma 2008; 25:
definitively. We need randomized trial data to show
1173–1177.
how to optimally guide use of MMM and enable
7. Hawryluk GW, Phan N, Ferguson AR, et al. Brain
better informed, individualized decisions at the bed- tissue oxygen tension and its response to physiological
side. However, as should be clear from this review, manipulations: influence of distance from injury site in
there is enough potential that this topic warrants a swine model of traumatic brain injury. J Neurosurg
ongoing attention. While we wait for randomized 2016; 125: 1217–1228.
trial data, a case can be made for greater bedside 8. Clausen T, Alves OL, Reinert M, et al. Association
brain monitoring and guarded clinical optimism. between elevated brain tissue glycerol levels and poor
outcome following severe traumatic brain injury.
Authors’ contributions J Neurosurg 2005; 103: 233–238.
9. Valadka AB, Goodman JC, Gopinath SP, et al.
Acknowledgements: The authors would like to thank the
Comparison of brain tissue oxygen tension to
Departments of Critical Care & Neurology at the
microdialysis-based measures of cerebral ischemia in
University of Calgary, University of Alberta & Harvard
fatally head-injured humans. J Neurotrauma 1998; 15:
University. Moreover, we would like to thank the reviewers
and editorial team at Journal of the Intensive Care Society 509–519.
10. Oddo M and Bosel J. Monitoring of brain and systemic
for their hard work and consideration.
oxygenation in neurocritical care patients. Participants
in the International Multidisciplinary Consensus
Declaration of conflicting interests
Conference on Multimodality M. Neurocrit Care
The author(s) declared no potential conflicts of interest with 2014; 21 Suppl 2 : S103–20.
respect to the research, authorship, and/or publication of 11. Brain Oxygen Optimization in Severe TBI, Phase 3.
this article. https://ClinicalTrials.gov/show/NCT03754114
(accessed 2 December 2020).
Funding 12. BONANZA Trial, http://anzctr.org.au/Trial/
The author(s) received no financial support for the research, Registration/TrialReview.aspx?id=378178&showOrig
authorship, and/or publication of this article. inal=true&isReview=true (accessed 2 December
2020).
ORCID iDs 13. Chesnut R, Aguilera S, Buki A, et al. A management
algorithm for adult patients with both brain oxygen and
Colin Casault https://orcid.org/0000-0002-3032-1206
intracranial pressure monitoring: the Seattle interna-
Peter Brindley https://orcid.org/0000-0001-7585-3591
tional severe traumatic brain injury consensus confer-
ence (SIBICC). Intensive Care Med May 2020; 46:
References 919–929.
1. Schiller JS, Lucas JW, Ward BW, et al. Summary 14. Veenith TV, Carter EL, Geeraerts T, et al.
health statistics for U.S. adults: National Health Pathophysiologic mechanisms of cerebral ischemia
Interview Survey, 2010. Vital Health Stat 2012; 252: and diffusion hypoxia in traumatic brain injury.
1–207. JAMA Neurol 2016; 73: 542–550.
2. Davis T and Ings A; National Institute of Health and 15. van Santbrink H, Maas AI and Avezaat CJ.
Care Excellence. Head injury: triage, assessment, inves- Continuous monitoring of partial pressure of brain
tigation and early management of head injury in chil- tissue oxygen in patients with severe head injury.
dren, young people and adults (NICE guideline CG Neurosurgery 1996; 38: 21–31.
176). Arch Dis Child Educ Pract Ed 2015; 100: 97–100. 16. van den Brink WA, van Santbrink H, Steyerberg EW,
3. James SL, Theadom A, Ellenbogen RG, et al.; Injury et al. Brain oxygen tension in severe head injury.
GBDTB, Spinal Cord Injury C. Global, regional, and Neurosurgery 2000; 46: 868–876; discussion 876–878.
10 Journal of the Intensive Care Society 0(0)

17. Meixensberger J, Jaeger M, Vath A, et al. Brain tissue 33. Aries MJ, Czosnyka M, Budohoski KP, et al.
oxygen guided treatment supplementing ICP/CPP ther- Continuous determination of optimal cerebral perfu-
apy after traumatic brain injury. J Neurol Neurosurg sion pressure in traumatic brain injury. Crit Care Med
Psychiatry 2003; 74: 760–764. Aug 2012; 40: 2456–2463.
18. Longhi L, Pagan F, Valeriani V, et al. Monitoring brain 34. Czosnyka M, Smielewski P, Kirkpatrick P, et al.
tissue oxygen tension in brain-injured patients reveals Continuous assessment of the cerebral vasomotor reac-
hypoxic episodes in normal-appearing and in peri-focal tivity in head injury. Neurosurgery 1997; 41: 11–17. dis-
tissue. Intensive Care Med 2007; 33: 2136–2142. cussion 17–19.
19. Narotam PK, Morrison JF and Nathoo N. Brain tissue 35. Klein SP, Depreitere B and Meyfroidt G. How I mon-
oxygen monitoring in traumatic brain injury and major itor cerebral autoregulation. Crit Care 2019; 23: 160.
trauma: outcome analysis of a brain tissue oxygen- 36. Depreitere B, Guiza F, Van den Berghe G, et al.
directed therapy. JNS 2009; 111: 672–682. Pressure autoregulation monitoring and cerebral perfu-
20. Chang JJ, Youn TS, Benson D, et al. Physiologic and sion pressure target recommendation in patients with
functional outcome correlates of brain tissue hypoxia in severe traumatic brain injury based on minute-by-
traumatic brain injury. Crit Care Med 2009; 37: minute monitoring data. J Neurosurg 2014; 120:
283–290. 1451–1457.
21. Spiotta AM, Stiefel MF, Gracias VH, et al. Brain tissue 37. Donnelly J, Czosnyka M, Adams H, et al.
oxygen-directed management and outcome in patients Individualizing thresholds of cerebral perfusion pres-
with severe traumatic brain injury. JNS 2010; 113: sure using estimated limits of autoregulation. Crit
571–580. Care Med 2017; 45: 1464–1471.
22. Oddo M, Levine JM, Mackenzie L, et al. Brain hypoxia 38. Carney N, Totten AM, O’Reilly C, et al. Guidelines for
is associated with short-term outcome after severe trau- the management of severe traumatic brain injury,
matic brain injury independently of intracranial hyper- fourth edition. Neurosurgery 2017; 80: 6–15.
tension and low cerebral perfusion pressure. 39. Kramer AH, Couillard PL, Zygun DA, et al.
Neurosurgery 2011; 69: 1037–1045, Discussion 1045. Continuous assessment of “optimal” cerebral perfusion
23. Bohman LE, Heuer GG, Macyszyn L, et al. Medical pressure in traumatic brain injury: a cohort study of
management of compromised brain oxygen in patients feasibility, reliability, and relation to outcome.
with severe traumatic brain injury. Neurocrit Care 2011; Neurocrit Care 2019; 30: 51–61.
14: 361–369. 40. Mathieu F, Zeiler FA, Whitehouse DP, et al.
24. Pascual JL, Georgoff P, Maloney-Wilensky E, et al. Relationship between measures of cerebrovascular reac-
Reduced brain tissue oxygen in traumatic brain tivity and intracranial lesion progression in acute TBI
injury: are most commonly used interventions success- patients: an exploratory analysis. Neurocrit Care 2020;
ful? J Trauma 2011; 70: 535–546. 32: 373–382.
25. Chesnut R, Aguilera S, Buki A, et al. A management 41. Rivera-Lara L, Zorrilla-Vaca A, Geocadin R, et al.
algorithm for adult patients with both brain oxygen and Predictors of outcome with cerebral autoregulation
intracranial pressure monitoring: the Seattle interna- monitoring: a systematic review and meta-analysis.
tional severe traumatic brain injury consensus confer- Crit Care Med 2017; 45: 695–704.
ence (SIBICC). Intensive Care Med 2020; 46: 919–929. 42. Needham E, McFadyen C, Newcombe V, et al.
26. Okonkwo DO, Shutter LA, Moore C, et al. Brain Cerebral perfusion pressure targets individualized to
oxygen optimization in severe traumatic brain injury Pressure-Reactivity index in moderate to severe trau-
phase-II: a phase II randomized trial. Crit Care Med matic brain injury: a systematic review.
2017; 45: 1907–1914. J Neurotrauma 2017; 34: 963–970.
27. Impact of early optimization of brain oxygenation on 43. Zeiler FA, Donnelly J, Nourallah B, et al. Intracranial
neurological outcome after severe traumatic brain and extracranial injury burden as drivers of impaired
injury, https://ClinicalTrials.gov/show/NCT02754063 cerebrovascular reactivity in traumatic brain injury.
(accessed 2 December 2020). J Neurotrauma 2018; 35: 1569–1577.
28. Lassen NA. Cerebral blood flow and oxygen consump- 44. Zeiler FA, Ercole A, Beqiri E, et al. Cerebrovascular
tion in man. Physiol Rev 1959; 39: 183–238. reactivity is not associated with therapeutic intensity in
29. Rivera-Lara L, Zorrilla-Vaca A, Geocadin RG, et al. adult traumatic brain injury: a CENTER-TBI analysis.
Cerebral autoregulation-oriented therapy at the bed- Acta Neurochir 2019; 161: 1955–1964.
side: a comprehensive review. Anesthesiology 2017; 45. Beqiri E, Smielewski P, Robba C, et al. Feasibility of
126: 1187–1199. individualised severe traumatic brain injury manage-
30. Guiza F, Meyfroidt G, Piper I, et al. Cerebral perfusion ment using an automated assessment of optimal cere-
pressure insults and associations with outcome in adult bral perfusion pressure: the COGiTATE phase II study
traumatic brain injury. J Neurotrauma 2017; 34: protocol. BMJ Open 2019; 9: e030727.
2425–2431. 46. Ungerstedt U and Pycock C. Functional correlates of
31. Czosnyka M, Balestreri M, Steiner L, et al. Age, intra- dopamine neurotransmission. Bull Schweiz Akad Med
cranial pressure, autoregulation, and outcome after Wiss 1974; 30: 44–55.
brain trauma. J Neurosurg 2005; 102: 450–454. 47. Roberts DJ, Jenne CN, Leger C, et al. Association
32. Zeiler FA, Donnelly J, Smielewski P, et al. Critical between the cerebral inflammatory and matrix metal-
thresholds of intracranial pressure-derived continuous loproteinase responses after severe traumatic brain
cerebrovascular reactivity indices for outcome predic- injury in humans. J Neurotrauma 2013; 30: 1727–1736.
tion in noncraniectomized patients with traumatic 48. Carteron L, Bouzat P and Oddo M. Cerebral micro-
brain injury. J Neurotrauma 15 2018; 35: 1107–1115. dialysis monitoring to improve individualized
Casault et al. 11

neurointensive care therapy: an update of recent clinical Bullock R, Marmarou A, Alessandri B, Watson J
data. Front Neurol 2017; 8: 601. (eds) Neuromonitoring in Brain Injury. Acta
49. Hutchinson PJ, O’Connell MT, Al-Rawi PG, et al. Neurochirurgica Supplements, Vol 75. Vienna:
Clinical cerebral microdialysis: a methodological Springer. 1999. DOI: https://doi.org/10.1007/978-3-
study. J Neurosurg 2000; 93: 37–43. 7091-6415-0_6
50. Langemann H, Alessandri B, Mendelowitsch A, et al. 64. Sanchez JJ, Bidot CJ, O’Phelan K, et al.
Extracellular levels of glucose and lactate measured by Neuromonitoring with microdialysis in severe traumat-
quantitative microdialysis in the human brain. Neurol ic brain injury patients. Acta Neurochir Suppl 2013; 118:
Res 2001; 23: 531–536. 223–227.
51. Hlatky R, Valadka AB, Goodman JC, et al. Patterns of 65. Sanchez-Porras R, Santos E, Czosnyka M, et al. Long’
energy substrates during ischemia measured in the brain pressure reactivity index (LPRx) as a measure of autor-
by microdialysis. J Neurotrauma 2004; 21: 894–906. egulation correlates with outcome in traumatic brain
doi:10.1089/0897715041526195 injury patients. Acta Neurochir 2012; 154: 1575–1881.
52. Sala N, Suys T, Zerlauth JB, et al. Cerebral extracellu- 66. Zauner A, Doppenberg E, Woodward J, et al.
lar lactate increase is predominantly nonischemic in Multiparametric continuous monitoring of brain
patients with severe traumatic brain injury. J Cereb metabolism and substrate delivery in neurosurgical
Blood Flow Metab 2013; 33: 1815–1822. patients. Neurol Res 1997; 19: 265–273.
53. Gallagher CN, Carpenter KL, Grice P, et al. The 67. Zauner A, Doppenberg E, Woodward J, et al.
human brain utilizes lactate via the tricarboxylic acid Continuous monitoring of cerebral substrate delivery
cycle: a 13C-labelled microdialysis and high-resolution and clearance: initial experience in 24 patients with
nuclear magnetic resonance study. Brain 2009; 132(Pt severe acute brain injuries. Neurosurgery 1997; 41:
10): 2839–2849. 1082–1093.
54. Timofeev I, Carpenter KL, Nortje J, et al. Cerebral 68. Schlenk F, Graetz D, Nagel A, et al. Insulin-related
extracellular chemistry and outcome following traumat- decrease in cerebral glucose despite normoglycemia in
ic brain injury: a microdialysis study of 223 patients. aneurysmal subarachnoid hemorrhage. Crit Care 2008;
Brain 2011; 134: 484–494. 12: R9.
55. Lazaridis C and Andrews CM. Brain tissue oxygena- 69. Vespa PM, McArthur D, O’Phelan K, et al. Persistently
tion, lactate-pyruvate ratio, and cerebrovascular pres- low extracellular glucose correlates with poor outcome
sure reactivity monitoring in severe traumatic brain 6 months after human traumatic brain injury despite a
injury: systematic review and viewpoint. Neurocrit lack of increased lactate: a microdialysis study. J Cereb
Care 2014; 21: 345–355. Blood Flow Metab 2003; 23: 865–877.
56. Verweij BH, Muizelaar JP, Vinas FC, Peterson PL, 70. Goodman JC, Valadka AB, Gopinath SP, et al.
et al. Impaired cerebral mitochondrial function after Extracellular lactate and glucose alterations in the
traumatic brain injury in humans. J Neurosurg 2000; brain after head injury measured by microdialysis.
93: 815–820. Crit Care Med 1999; 27: 1965–1973.
57. Stein NR, McArthur DL, Etchepare M, et al. Early 71. Gupta DK, Singla R, Kale SS, et al. Intracerebral
cerebral metabolic crisis after TBI influences outcome hypoglycemia and its clinical relevance as a prognostic
despite adequate hemodynamic resuscitation. Neurocrit indicator in severe traumatic brain injury: a cerebral
Care 2012; 17: 49–57. microdialysis study from India. Neurol India 2016; 64:
58. Nelson DW, Thornquist B, MacCallum RM, et al. 259–264.
Analyses of cerebral microdialysis in patients with trau- 72. Kurtz P, Claassen J, Schmidt JM, et al. Reduced brain/
matic brain injury: relations to intracranial pressure, serum glucose ratios predict cerebral metabolic distress
cerebral perfusion pressure and catheter placement. and mortality after severe brain injury. Neurocrit Care
BMC Med 2011; 9: 21. 2013; 19: 311–319.
59. Investigators N-SS, Finfer S, Chittock DR, et al. 73. Yokobori S, Watanabe A, Matsumoto G, et al. Time
Intensive versus conventional glucose control in critical- course of recovery from cerebral vulnerability after
ly ill patients. N Engl J Med 2009; 360: 1283–1297. severe traumatic brain injury: a microdialysis study.
60. Vespa P, Boonyaputthikul R, McArthur DL, et al. J Trauma 2011; 71: 1235–1240.
Intensive insulin therapy reduces microdialysis glucose 74. Jeremitsky E, Omert LA, Dunham CM, et al. The
values without altering glucose utilization or improving impact of hyperglycemia on patients with severe brain
the lactate/pyruvate ratio after traumatic brain injury. injury. J Trauma 2005; 58: 47–50.
Crit Care Med 2006; 34: 850–856. 75. Oddo M, Frangos S, Maloney-Wilensky E, et al. Effect
61. Oddo M, Schmidt JM, Carrera E, et al. Impact of tight of shivering on brain tissue oxygenation during induced
glycemic control on cerebral glucose metabolism after normothermia in patients with severe brain injury.
severe brain injury: a microdialysis study. Crit Care Neurocrit Care 2010; 12: 10–16.
Med 2008; 36: 3233–3238. 76. Oddo M, Frangos S, Milby A, et al. Induced normo-
62. Zeiler FA, Thelin EP, Helmy A, et al. A systematic thermia attenuates cerebral metabolic distress in
review of cerebral microdialysis and outcomes in TBI: patients with aneurysmal subarachnoid hemorrhage
relationships to patient functional outcome, neurophys- and refractory fever. Stroke 2009; 40: 1913–1916.
iologic measures, and tissue outcome. Acta Neurochir 77. Nordstrom CH, Reinstrup P, Xu W, et al.
(Wien) 2017; 159: 2245–2273. Assessment of the lower limit for cerebral perfusion
63. Alessandri B, Doppenberg E, Zauner A, et al. Evidence pressure in severe head injuries by bedside monitoring
for time-dependent glutamate-mediated glycolysis in of regional energy metabolism. Anesthesiology 2003; 98:
head-injured patients: A microdialysis study. In: 809–814.
12 Journal of the Intensive Care Society 0(0)

78. Hutchinson PJ, Jalloh I, Helmy A, et al. Consensus 85. Reaven NL, Lovett JE and Funk SE. Brain injury and
statement from the 2014 international microdialysis fever: hospital length of stay and cost outcomes.
forum. Intensive Care Med 2015; 41: 1517–1528. J Intensive Care Med 2009; 24: 131–139.
79. Oddo M, Milby A, Chen I, et al. Hemoglobin concen- 86. Belli A, Sen J, Petzold A, et al. Metabolic failure pre-
tration and cerebral metabolism in patients with aneu- cedes intracranial pressure rises in traumatic brain
rysmal subarachnoid hemorrhage. Stroke 2009; 40: injury: a microdialysis study. Acta Neurochir (Wien)
1275–1281. 2008; 150: 461–469; discussion 470.
80. Kurtz P, Schmidt JM, Claassen J, et al. Anemia is asso- 87. Tisdall M, Russo S, Sen J, et al. Free phenytoin con-
ciated with metabolic distress and brain tissue hypoxia centration measurement in brain extracellular fluid: a
after subarachnoid hemorrhage. Neurocrit Care 2010; pilot study. Br J Neurosurg 2006; 20: 285–289.
13: 10–16. 88. Fernández-Candil J, Gamb us PL, Troc
oniz IF, et al.
81. Çavuş I, Romanyshyn JC, Kennard JT, et al. Elevated Feasibility of microdialysis to determine interstitial
basal glutamate and unchanged glutamine and GABA rocuronium concentration in the muscle tissue of anes-
in refractory epilepsy: microdialysis study of 79 patients thetized neurosurgical patients. J Neurosurg Anesthesiol
at the Yale epilepsy surgery program. Ann Neurol 2016; 2010; 22: 163–169.
80: 35–45. 89. Sarrafzadeh A, Schlenk F, Gericke C, et al. Relevance
82. Hutchinson PJ, Gimson A, Al-Rawi PG, et al. of cerebral interleukin-6 after aneurysmal subarachnoid
Microdialysis in the management of hepatic encepha- hemorrhage. Neurocrit Care 2010; 13: 339–346.
lopathy. Neurocrit Care 2006; 5: 202–208. 90. Casault C, Al Sultan AS, Banoei M, et al. Cytokine
83. Bailey RL, Quattrone F, Curtin C, et al. The safety of responses in severe traumatic brain injury: where there
multimodality monitoring using a Triple-Lumen bolt in is smoke, is there fire? Neurocrit Care 2019; 30: 22–32.
severe acute brain injury. World Neurosurg 2019; 130: 91. Banoei MM, Casault C, Metwaly SM, et al.
e62–e67. Metabolomics and biomarker discovery in traumatic
84. Greer DM, Funk SE, Reaven NL, et al. Impact of fever brain injury. J Neurotrauma 2018; 35: 1831–1848.
on outcome in patients with stroke and neurologic 92. Rivers E, Nguyen B, Havstad S, et al. Early goal-
injury: a comprehensive meta-analysis. Stroke 2008; directed therapy in the treatment of severe sepsis and
39: 3029–3035. septic shock. N Engl J Med 2001; 345: 1368–1377.