E REVIEW ARTICLE

Tissue Oximetry and Clinical Outcomes

Philip Bickler, MD, PhD, John Feiner, MD, Mark Rollins, MD, PhD, and Lingzhong Meng, MD

A number of different technologies have been developed to measure tissue oxygenation, with
the goal of identifying tissue hypoxia and guiding therapy to prevent patient harm. In specific
cases, tissue oximetry may provide clear indications of decreases in tissue oxygenation such as
that occurring during acute brain ischemia. However, the causation between tissue hemoglobin-
oxygen desaturation in one organ (e.g., brain or muscle) and global outcomes such as mortality,
intensive care unit length of stay, and remote organ dysfunction remains more speculative. In
this review, we describe the current state of evidence for predicting clinical outcomes from tis-
sue oximetry and identify several issues that need to be addressed to clarify the link between
tissue oxygenation and outcomes. We focus primarily on the expanding use of near-infrared
spectroscopy to assess a venous-weighted mixture of venous and arterial hemoglobin-oxygen
saturation deep in tissues such as brain and muscle. Our analysis finds that more work is
needed in several areas: establishing threshold prediction values for tissue desaturation–
related injury in specific organs, defining the types of interventions required to correct changes
in tissue oxygenation, and defining the effect of interventions on outcomes. Furthermore, well-
designed prospective studies that test the hypothesis that monitoring oxygenation status in
one organ predicts outcomes related to other organs need to be done. Finally, we call for more
work that defines regional variations in tissue oxygenation and improves technology for mea-
suring and even imaging oxygenation status in critical organs. Such studies will contribute to
establishing that monitoring and imaging of tissue oxygenation will become routine in the care
of high-risk patients because the monitors will provide outputs that direct therapy to improve
clinical outcomes.  (Anesth Analg 2016;XXX:00–00)

T
he oxygenation status of tissues can change deleteri-
ously during hypotension, hypovolemia, hemorrhage,
shock, or ischemia caused by embolism; compression;
or other factors—situations during which arterial hemoglo-
bin (Hb) oxygen saturation (Sao2) or even mixed venous
saturation (Svo2) may remain normal. A number of different
technologies have been developed to measure tissue oxy-
gen or a surrogate of tissue oxygen. One example is tissue
Hb-oxygen saturation. Recently, a number of observational
and randomized controlled trials have tested the hypoth-
esis that measurement of tissue Hb-oxygen saturation with
near-infrared spectroscopy (NIRS) is associated with impor-
tant clinical outcomes.1–4 The hypothesis that NIRS predicts
outcomes has been tested in 2 paradigms: (1) in studies that
test the hypothesis that desaturation events in the monitored
organ predict outcomes specific to that organ (e.g., cerebral
oximetry used to predict stroke or cognitive changes); and
(2) in studies that have tested whether desaturation in a
From the Department of Anesthesia and Perioperative Care, University of
California San Francisco, San Francisco, California.
Accepted for publication March 1, 2016.
Funding: Funded by the University of California San Francisco (UCSF) Hypoxia
Research Laboratory from funds derived from the validation and testing of pulse
oximeters. No sponsor directly funded the study or participated in study design.
None of the authors have financial interests in clinical monitoring companies.
Casmed Inc. has provided patient sensors for a tissue oximetry study at UCSF.
Conflict of Interest: See Disclosures at the end of the article.
This article was presented as part of the international symposium
“Innovations and Advances in Monitoring Perfusion, Oxygenation and
Ventilation” (IAMPOV) held at St. Luke’s International University, Tokyo,
Japan, October 3, 2015.
Reprints will not be available from the authors.
Address correspondence to Philip Bickler, MD, PhD, Department of Anes-
thesia and Perioperative Care, University of California San Francisco, 513
Parnassus Ave., Box 0542, San Francisco, CA 94143. Address e-mail to philip.
bickler@ucsf.edu.
Copyright © 2016 International Anesthesia Research Society
DOI: 10.1213/ANE.0000000000001348
specific tissue predicts remote or global outcomes such as
failure of remote organs, length of intensive care unit (ICU)
admission, or mortality. The main purpose of this article
is to review the current state of knowledge about whether
either of these hypotheses is supportable and to identify
the gaps in knowledge and technology that would help to
resolve the remaining unknowns.
HOW COMMON IS TISSUE HYPOXIA OR
TISSUE HbO2 DESATURATION DURING THE
PERIOPERATIVE PERIOD?
Both tissue hypoxia and arterial desaturation events in the
perioperative period may be more common than is recog-
nized. The incidence of these events is a critical issue, but
we have surprisingly little information concerning the
frequency and depth of perioperative tissue desaturation
events. Understanding the temporal patterns and ampli-
tude of specific tissue vasculature desaturations events in
hospitalized patients is essential for determining whether
desaturations are linked to outcomes.
Consider a parallel situation regarding the incidence
of arterial desaturation events detected by pulse oxim-
etry. Although pulse oximetry has been used clinically for
>30 years, it has not been until recently that we have learned
that perioperative arterial desaturation events are common.
Desaturations occur rarely in the operating room or postan-
esthesia care unit when patients are closely monitored, but
frequently during the postoperative period when patients
are less monitored and exposed to the respiratory-depres-
sant effects of opioid pain medications. The Outcomes
Research Group at the Cleveland Clinic recently reported
that noncardiac surgery patients monitored with blinded,
continuous pulse oximetry experienced a large number of
desaturation events during the postoperative period. These
events were both frequent and prolonged.5 We now know
that interventions to correct pulse oximeter–measured

XXX 2016 • Volume XXX • Number XXX www.anesthesia-analgesia.org

1
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
 E REVIEW ARTICLE
desaturations can avert major undesirable outcomes. For
example, in a study by Taenzer et al.,6 institution of moni-
tored pulse oximetry on a medical/surgical ward reduced
serious adverse outcomes, such as ICU admissions for pul-
monary complications. However, other studies of pulse
oximetry and clinical outcomes have been unsuccessful in
clearly proving clinical benefit.7–9
There is currently a limited amount of knowledge
concerning the relative numbers of tissue desaturations
associated with tissue hypoxia in the perioperative and
postoperative periods. It is clear, and not surprising, that
some of the events do occur in the postoperative period,10
and it is not known whether these postoperative events are
more important to outcomes than are intraoperative events.
With respect to the frequency of intraoperative desaturation
events, several studies have shown that desaturation events
are common in patients undergoing shoulder surgery in the
beach chair or sitting positions.11,12 These observations were
driven by the occurrence of severe neurologic complications
in shoulder surgery patients. In major spine surgery, tissue
desaturation events may be common. In 73 patients at the
University of California San Francisco, there were 141 cere-
bral and 129 peripheral tissue desaturation events (defined
as a decrease in saturation of ≥5% within 30 minutes), and
only 43 of the total 270 events were simultaneous in both tis-
sues (Lingzhong Meng, unpublished data, November 2015).
REVIEW OF NON-NIRS TECHNOLOGY THAT
MEASURES AND MONITORS LOCAL TISSUE
OXYGEN LEVELS
Tissue oxygenation is the result of a complex interaction
of blood perfusion, arterial oxygen tension, Hb level and
dissociation conditions, and local oxygen consumption.
Consequently, each oxygen-monitoring technology involves
specific assumptions and limitations. Although the primary
focus of this review is NIRS, we will also provide a brief
overview of other tissue oxygen measurement technologies,
so that the strengths and limitations of NIRS can be under-
stood in a larger context.
Direct Tissue Po2 Measurement
The polarographic or Clark-type oxygen electrode is still
considered the “gold standard” of tissue oxygen measure-
ment. It involves the placement of an electrode into tissue
and provides a tissue oxygen partial pressure (Ptio2). It is
invasive, has a limited tissue-sampling area, consumes
oxygen in the surrounding tissue, is unable to reveal the
heterogeneity of oxygen levels in the tissue, and requires
temperature correction and periodic recalibration.13,14
Phosphorescence quenching fiber-optic probes involve a
chemical sensor located at the tip of a probe that changes
is fluorescence properties as a function of oxygen levels.15
Pulsed light is carried fiber optics to the sensor tip. The
change in the wavelength and amplitude of the return-
ing light is used to determine the oxygen partial pressure.
Although these probes do not consume oxygen, they other-
wise have similar limitations as the polarographic type of
measurement.
Direct Ptio2 measurement has been used in clinical
research applications, including wound healing,16 trauma
2   
www.anesthesia-analgesia.org
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
and resuscitation,17 cancer tumor research and therapy opti-
mization,18 and neurologic injury. The available evidence on
the relationship of multimodality measurements to clinical
neurologic outcomes has been reviewed.19 Of interest, the
relationship of intracranial pressure monitoring to clinical
outcomes remains uncertain, as revealed by a recent meta-
analysis.20 However, the situation may be more positive
for brain oxygenation and outcomes. Among the clinical
settings in which Ptio2 monitoring tracks clinical events is
treatment of arterial vasospasm in subarachnoid hemor-
rhage.21 For an excellent recent review of brain Ptio2 mea-
surements and neurocritical care clinical outcomes, see the
study reported by De Georgia.22
Transcutaneous Tissue Monitoring
This noninvasive device uses both a polarographic oxy-
gen electrode and carbon dioxide electrode affixed to the
skin with a contact gel to promote gas diffusion. The skin
is warmed to 42 to 44°C to promote diffusion of oxygen
through the stratum corneum. Limitations of oxygen mea-
surement are similar to the Clark electrode. Transcutaneous
oxygen measurement is used clinically in wound healing
and peripheral vascular disease requiring amputation16 and
has also been used as a measure of resuscitation.23
Electron Paramagnetic Resonance
This oxygen measurement method requires placement of
a paramagnetic material, such as lithium phthalocyanine
crystals or India ink, in the tissue of interest. Electron para-
magnetic resonance oximetry is based on the paramagnetic
characteristics of molecular oxygen, which in its ground
state has 2 unpaired electrons and undergoes spin-exchange
interaction with the paramagnetic material. The relaxation
rate of the material (spin probe) increases proportionately
with the surrounding partial pressure of oxygen, shorten-
ing both the spin-spin and spin-lattice relaxation times and
broadening the electron paramagnetic resonance spectral
line width, which becomes a linear function of Po2.24 This
technique is limited by the need to implant a sensing mate-
rial, potential foreign body reaction over time, limited
sampling area, and lack of available spectroscopic instru-
mentation at most medical locations. Although human
research protocols are currently being used in wound heal-
ing, cancer, and peripheral vascular disease, no large out-
come data have been completed.24
Microdialysis
This methodology tracks the levels of metabolites in various
tissues of interest such as muscle or brain using microdialy-
sis. A catheter composed of a fine double-lumen catheter is
placed in the tissue of measurement and slowly perfused
with an isotonic fluid collected in a chamber for analysis.
The slow constant movement of fluid allows molecules of
interest to diffuse along a concentration gradient across the
semipermeable membrane.25 Compounds commonly mea-
sured include levels of lactate and pyruvate, glutamate, and
glucose. The ratio of lactate/pyruvate correlates with tissue
ischemia and anaerobic metabolism and has been used to
aid in management of hemorrhagic shock.26 Cerebral micro-
dialysis has noted an elevated lactate, and lactate-pyruvate
anesthesia & analgesia
 Tissue Oximetry and Clinical Outcomes
ratio is associated with poor neurologic outcomes in
subarachnoid hemorrhage and traumatic brain injury.27
Limitations are the indirect measure of tissue oxygen, that
is invasive, technically demanding, and may be influenced
by other pathologic processes separate from relative oxygen
levels.
Spectroscopy to Measure the Oxidation-
Reduction State of Intracellular Cytochromes
The earliest efforts to use NIRS for clinical application
involved attempts to assess the light absorption or fluores-
cence of cytochromes and of compounds such as reduced
nicotinamide adenine dinucleotide (NADH) which are a
function of Po2 in the cytosol and mitochondrion.28,29
Assessment of the redox state of cytochrome oxidase
in complex tissues in vivo is challenging. Cytochrome oxi-
dase contains 2 hemoproteins and 2 copper proteins with a
wide absorption spectrum in the infrared region. This over-
laps significantly with the near-infrared absorbance of Hb,
which is also an order of magnitude greater in amplitude
compared with the cytochrome spectrum in most tissue
regions of interest. In addition, the absorption coefficient of
cytochrome oxidase may not be constant.30 Validation of the
signals can be based on measurements in experimental ani-
mals involving injection of cyanide to alter the cytochrome
state independent of Hb-oxygen saturation.31,32
Even with these approaches, it is not surprising that NIRS
of cytochromes is subject to errors because of interference
from oxy- and reduced Hb in tissue and other factors.33,34
Promising efforts continue to measure the cytochrome
redox state using new advanced multispectral approaches
employing broadband light sources, spectrophotometer
arrays capable of measuring the intensity of multiple to
hundreds of wavelengths of light, and advanced computa-
tional algorithms.35,36 Currently, it is unknown whether the
cytochrome oxidation state could predict clinical outcomes.
Magnetic Resonance Imaging Techniques
19
F Magnetic Resonance Imaging
This method uses direct tissue injection of microliters of per-
fluorinated compounds such as hexafluorobenzene at mul-
tiple locations. Tissue oxygen concentrations surrounding
each droplet can be quantified with high specificity because
of the linear relationship between Ptio2 and the spin-lattice
relaxation (R1).37 This technique provides quantitative, spe-
cific, spatially resolved, temperature-independent Ptio2,
with a linear relationship for Ptio2 between 0 and >500 mm
Hg. The technique has shown value primarily in animal
research studies evaluating oxygen levels in various organs
but is limited in clinical studies by the need to administer a
perfluorinated substance and periodic access to a magnetic
resonance imaging (MRI) scanner capable of multinuclear
imaging.38
Blood Oxygen Level-Dependent–Based MRI and Dynamic
Susceptibility Contrast
The paramagnetic nature of deoxygenated Hb provides
an endogenous contrast agent for the blood oxygen level–
dependent (BOLD) technique. Increased amounts of deoxy-
genated Hb accelerate the spin-spin relaxation time (T2) and
weighted signal relaxation (T2*w). These parameters are
XXX 2016 • Volume XXX • Number XXX www.anesthesia-analgesia.org
3
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
used to determine the transverse relaxation rate of water in
blood and surrounding tissue (R2). Tissues with increased
perfusion have an increased R2 value. These changes in
R2 are not a direct measure of tissue oxygen levels and can
be influenced by other variables not related to tissue oxy-
genation. However, the changes in R2 are taken to imply
changes in tissue oxygenation.39 Despite these limitations, a
correlation between direct tissue oxygen measurement and
BOLD imaging noted a significant correlation in a human
study of prostate tumors.39
Dynamic susceptibility contrast MRI requires adminis-
tration of a contrast agent and permits calculation of cere-
bral blood volume and mean transit time. This information
combined with data from BOLD allows calculation of oxy-
gen extraction rations and the potential to identify compro-
mised tissue with mapping of cerebral metabolic oxygen
consumption rates.40,41 Although these techniques have
demonstrated promising results in patients with ischemic
brain injury, they are still in the preclinical setting and need
further validation.42,43
NIRS-BASED OXYHEMOGLOBIN SATURATION
MEASUREMENTS IN TISSUE
Overview of Technology
HbO2 tissue oximetry with near-infrared light is a nonin-
vasive, optical technology that integrates blood Hb-oxygen
saturation in a region of interest (Fig. 1). By the nature of the
technique, measurement is limited to the intravascular com-
partment and combines measures of arterial, venous, and
capillary blood and depending on the wavelengths of light
and tissue, oxy- and deoxymyoglobin. Unlike pulse oxim-
etry, NIRS does not involve detection of a pulsatile tissue
component but relies entirely on the Beer-Lambert law that
relates the concentration of a substance to its light absorp-
tion. The ability to discriminate various types of chromo-
phores varies with the number of wavelengths; for example,
a 2-wavelength NIRS device cannot discriminate between
the levels of reduced Hb, oxyhemoglobin, reduced myoglo-
bin, and oxymyoglobin. This can be a significant issue when
comparing the readings of an NIRS device on the fore-
head and thenar muscles, with obvious differences in the
amount of myoglobin present in the 2 tissues. For cerebral
NIRS devices, the assumption is that myoglobin is a trivial
contributor to the signal and that the vascular Hb-oxygen
component predominates. Cerebral oximeters are calibrated
by the manufacturer assuming a constant 70% or 75%
weighting toward venous blood saturation. A 2012 review
of the development of this technology is by Ferrari and
Quaresima.44 Oxygen consumption with insufficient oxygen
delivery will lead to an increase in reduced Hb, and since
Hb chromophores dominate the near-infrared absorption
spectrum, the estimation of Hb-oxygen saturation is argu-
ably less susceptible to errors and signal-to-noise problems
than are NIRS-based assessment of cytochrome redox sta-
tus. An in vivo calibration can be accomplished with direct
measurement of cerebral venous and arterial saturation in
blood samples.45 This calibration, incorporated into the cal-
culation algorithm that determines the displayed value of
Sco2, should theoretically reduce the interfering effects of
different Hb concentrations and variations in tissue light
transmission. However, this calibration approach assumes
 E REVIEW ARTICLE

Figure 1. Scheme of multiwavelength near-infrared spectroscopic measurement of cerebral tissue hemoglobin-oxygen saturation used by com-
mercially available cerebral oximeters. The technology typically involves 2–4 wavelengths of near-infrared incident light, laser generated and
carried to the scalp sensor/light source probe by fiber-optic cable, or from specific-wavelength light emitting diodes (LED) in the probe itself.
This basic configuration can be used for cerebral oximetry, muscle oximetry, or other organs. In the case of cerebral oximetry, 2 probes are
typically mounted on the patient’s forehead to measure hemoglobin-oxygen saturation in the right and left frontal cortex. Unlike pulse oxim-
etry, the analysis of the signals from the photodetectors does not utilize the pulsatile component of the photodetector output for calculation
of the saturation. As in the probes depicted, 2 sets of photodetectors can be employed to capture light signals from different tissue depths,
enabling a partial correction for signals from blood in the scalp, galeal tissues, and bone sinuses. Calibration of cerebral oximeters involves
fitting of an algorithm to measurements of arterial and cerebral mixed venous blood, assuming a fixed mixture of these blood components,
usually 75:25 or 70:30 venous:arterial. No calibration by the user is possible. The format of the display of saturation signals varies with the
instrument, with most have a trending plot capability and features such as user-adjustable alarm settings.

that all individuals have the same constant ratio of cere-
bral venous and arterial blood within the tissue of where
saturation is measured. As discussed in the next section,
this assumption is likely incorrect. Clinical interest in the
devices is substantial, in part, because of the hope that brain
oxygenation state assessed by cerebral oximeters might pre-
dict important long-term outcomes and complications.
How Accurate Is NIRS for Measuring Brain
Tissue Oxyhemoglobin Saturation?
Based on a study evaluating the accuracy of 5 cerebral
oximeters by Bickler et al.,45 currently manufactured cere-
bral oximeters do not provide an “absolute” measurement
of oxyhemoglobin saturation in the tissue region of inter-
est, despite the theory that spatially resolved spectroscopy
can determine a scaled tissue Hb concentration and there-
fore the relative concentrations of oxy- and deoxyhemoglo-
bin. The between-subject variability and dynamic error of
readings revealed in the study by Bickler et al. also make
it difficult to determine the absolute threshold for predict-
ing tissue injury. Furthermore, the measured ratio of venous
and arterial blood in brain tissue is about 50:5046 and vari-
able, whereas instrument manufacturers calibrate to fixed
venous:arterial volume ratios of 70:30 or 75:25. Changes in
cerebral venous and arterial blood volumes during changes
in carbon dioxide or oxygen will violate assumptions about
constancy of arterial to venous blood volumes.45
Contamination of signals from blood in extracranial tis-
sue within the sampled volume of interest also complicates
the clinical utility of this technology,47,48 as depicted in
Figure  1. Administration of peripherally acting vasocon-
strictors (e.g., phenylephrine or norepinephrine) is one
example of an influence that can change the contribution
of extracranial blood and alter readings.49 In addition, sym-
pathetically mediated vasoconstriction from pain, hypo-
thermia, and hypovolemia significantly alters peripheral
tissue oxygenation and may also influence the extracranial
component of the reading. Therefore, the measurements are
at best relative assessments of tissue oxygenation and are
useful as trend monitors, not threshold of injury monitors.
HOW WIDESPREAD IS THE USE OF TISSUE
OXIMETRY?
Given the diversity of technologies available to assess tis-
sue well-being, it is of interest to know something about
current clinical practice patterns. For example, in a 2015
survey of all 31 neurocritical care units in the United
Kingdom, Wijayatilake et al.50 found that intracranial
pressure monitoring was used in all the 31 institutions.
Cerebral perfusion pressure was used in 30 of the 31 units,
and a cerebral perfusion pressure target of 60 to 70 mm
Hg was the most widely used target (25 of 31 units).
Transcranial Doppler was used in 12 units (39%); brain
tissue oxygen was used in 8 (26%); cerebral microdialysis
was used in 4 (13%); jugular bulb oximetry in 1 unit; and
NIRS was not used in any unit. The authors called for goal-
directed therapy targeting readings of these instruments
to improve outcomes, as has been the case for goal-driven

4   
www.anesthesia-analgesia.org anesthesia & analgesia
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
 Tissue Oximetry and Clinical Outcomes
therapy in sepsis.51 To our knowledge, a similar survey of
clinical tissue-monitoring technology has not been done in
the United States or other countries.
ORGAN-SPECIFIC OUTCOMES AND NIRS
The brain is the organ most frequently directly assessed
with NIRS. Although monitoring cerebral oxygenation to
prevent the adverse consequences of acute cerebral hypo-
perfusion and hypoxia is supportable with current knowl-
edge, the issue of predicting longer-term outcomes remains
less clear. Focusing first on NIRS-based cerebral oximetry,
we review a variety of central nervous system–specific out-
comes that have been assessed in both cardiac and noncar-
diac surgery.
Brain Ischemia Detection in Cardiopulmonary
Bypass/Cardiac Surgery
Harilall et al.52 examined the effects of interventions used to
correct cerebral desaturation events during cardiac surgery
on a biochemical marker of brain injury, S100B protein lev-
els in blood. Harilall et al. used a prioritized intraoperative
management protocol to maintain cerebral oximetry values
>75% of the baseline during cardiopulmonary bypass. The
mean desaturation time for the control group was 64 and
25  minutes in the intervention group. A significant reduc-
tion in S100B levels was found in the intervention group,
but it was not established if neurologic outcomes were
affected by the interventions to reduce the brain exposure
to lower cerebral saturations. Nonneurologic outcomes in
cardiac surgery are reviewed below.
Brain Ischemia Detection in Carotid
Endarterectomy
The use of cerebral oximetry in carotid endarterectomy
to diagnose cerebral hypoperfusion and determine which
patients received selective shunting has been compared
with electroencephalograph monitoring and transcranial
Doppler. However, it remains unclear whether cerebral
oximetry serves as a reliable clinical monitor in carotid
endarterectomy.53 Studies have not been sufficiently pow-
ered to determine the relationship between cerebral NIRS
and stroke, and the use of shunting is left to the surgeon’s
discretion. Transcranial Doppler cannot always be mea-
sured because of the lack of an intracranial window, and it
requires significant expertise. If cerebral saturation can be
shown to have similar predictive value as other modalities,
the simplicity and ease of measurement would be advanta-
geous. Among the studies relating tissue oximetry to spe-
cific end points (if not outcomes such as stroke) are a series
of studies using cerebral oximetry during carotid surgery
by Pennekamp et al.53–56
Cognitive Function and Postoperative Cognitive
Decline
Postoperative cognitive function is of concern after cardiac
surgery and after any surgery in the elderly. Studies have
used a variety of cognitive tests and different measures of
postoperative cognitive decline (POCD). Among the out-
comes predicted by low cerebral oximeter readings include
early postoperative neuropsychiatric impairment.
XXX 2016 • Volume XXX • Number XXX www.anesthesia-analgesia.org
5
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
A number of studies have found an association between
cerebral desaturation events and POCD in cardiac surgery
patients. In an observational study of 101 cardiac surgery
patients, Yao et al.57 reported that a multivariate analysis
revealed that time spent at saturations <40% predicted cog-
nitive deficits. Similarly, de Tournay-Jetté et al.58 found that
cerebral desaturation correlated with both early and late
POCD. Slater et al.4 performed a prospective randomized
trial in 240 cardiac surgery patients where interventions
were initiated for a >20% decrease in cerebral saturation,
compared with a control group blinded to the saturation.
No difference was found between groups. The authors sug-
gest that poor adherence to protocol may have contributed
to the lack of a treatment effect because prolonged cerebral
desaturation correlated to early POCD. In addition, 2 obser-
vational studies found no relationship between cerebral
saturation and POCD in cardiac surgery patients.59,60 In con-
clusion, it is too early to tell whether cerebral desaturation
events actually have significant predictive power for POCD,
and a multicenter interventional trial may be needed to
resolve the issue.
In elderly patients undergoing major abdominal surgery,
Casati et al.61 randomly assigned patients to an interven-
tion group receiving treatment for low cerebral saturation
or a control group where physicians were blinded to the
cerebral saturation. In the study group, an intervention
was initiated when cerebral saturation decreased <75% of
baseline. The intervention group had significantly higher
measures of saturation during surgery. Cognitive function,
assessed with a mini mental status score (MMSE), did not
differ between groups. However, in patients experiencing
intraoperative cerebral desaturation, MMSE was lower at
day 7 in the control group. In this study and an additional
observation study in a similar patient population, Casati et
al.62 found that cerebral saturation correlated with MMSE
and POCD. These findings are similar to those in a group of
arthroplasty patients studied by Lin et al.63
Adult Resuscitation and Care of the Premature
Newborn
Cerebral oximetry has been proposed as a monitor of the
adequacy of resuscitation during cardiac arrest.64,65 This
interesting concept needs further study.
In neonates, cerebral NIRS monitoring with an inter-
vention to correct cerebral tissue desaturations compared
with blinded monitoring reduced the incidence of cerebral
desaturation by 58% in a phase II clinical trial.66 Additional
studies are needed to determine whether cerebral oximetry
improves outcomes.
Traumatic Brain Injury
The hypothesis that noninvasive assessment of brain tissue
oxygenation predicts outcomes is based on studies that have
used invasive methods to measure brain oxygenation. For
example, in traumatic brain injury, an electrode-measured
tissue Po2 of <28 mm Hg is associated with mortality.67
NIRS cerebral oximetry in traumatic brain injury has
been explored as a prognostic index or to assess the impact
of goal-directed treatment. An example is goal-directed
treatment of cerebral perfusion pressure after head injury.68
 E REVIEW ARTICLE
In a study by Rosenthal et al.,69 a correlation between cere-
bral oximetry measurements and jugular bulb venous mea-
surements indicates that the NIRS technology may be able
to provide an estimation of cerebral oxygenation status non-
invasively. Whether trends in noninvasive oximetry predict
clinical outcomes remains untested.
PREDICTION OF GLOBAL OUTCOMES WITH NIRS
The possible association of tissue hypoxia or Hb desaturation
and modifiable global outcomes is one of the most impor-
tant factors driving both research and clinical application of
NIRS. A number of studies have tested the hypothesis that
oxygenation status measured in one tissue predicts dysfunc-
tion of remote tissues or determines global outcomes such as
mortality. To date, 2 tissues have been associated with remote
outcomes: (1) the frontal cortex of brain and (2) peripheral
muscle (thenar muscle, leg muscle, and masseter muscle).
A range of outcomes has been associated with desaturation
events in these 2 tissue beds. Before we discuss the specifics
of some of the key studies, it is important to note that muscle
and brain are distinct physiologically with respect to blood
flow autoregulation. During stresses impairing tissue oxy-
gen delivery, such as hypotension, autoregulation of blood
flow in the brain is substantially greater than in the skeletal
muscle. Cerebral autoregulation was recently reviewed by
Meng et al.70 Because of its relatively small autoregulatory
capacity, muscle hypoxia/desaturation might be a “leading
indicator” of impaired oxygen delivery. Muscle at rest has
a low metabolic rate, and the degree to which this leading
indicator’s role is true during different types of impaired
oxygen delivery is not certain.
In contrast to muscle, hypoperfusion in the brain (absent
a thrombus) is probably reflective of low cardiac output or
shock, conditions with a poor prognosis. Because of auto-
regulation of blood flow in the cerebral cortex, decreased
global brain oxygenation is likely reflective of a systemic
problem with either low cardiac output or low vascular
resistance, as in sepsis. The brain is therefore a “trailing indi-
cator” of tissue hypoperfusion because of autoregulation.
The opposite situation occurs in skeletal muscle because it
experiences a relatively large reduction in blood flow dur-
ing shock; muscle is therefore a leading indicator of global
hypoperfusion compared with the brain. Figure 2 illustrates
the complexity of understanding how desaturations in one
tissue can relate to global outcomes and how interventions
to correct desaturation in one tissue have global effects.
There are a number of different possibilities for a mecha-
nistic link between stress in one tissue and remote effects
on other tissues or the whole organism. The concept of tis-
sue hypoxia as a biomarker for serious systemic illness has
been proposed.33 Probably, the most compelling possibility
is that tissue hypoxia or ischemia causes the release of pro-
inflammatory cytokines and other inflammatory mediators.
These mediators produce inflammatory reactions in remote
parts of the body. Chemical mediators (e.g., cytokines such
as interleukins and tumor necrosis factor) or activated neu-
trophils released into the circulation from these leading
indicator tissues could have global effects, including spe-
cific effects on different organs. Increases in circulating cyto-
kines is a well-documented consequence of global hypoxia
experienced during high-altitude exposure, after cerebral
6   
www.anesthesia-analgesia.org
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Figure 2. Hypothetical relationship of tissue desaturation/hypoxia
events and clinical outcomes in patients monitored with a cere-
bral oximeter and a peripheral pulse oximeter. Because of cerebral
autoregulation of blood flow and oxygen delivery,70 desaturation in
brain tissue is expected to occur later or only with more profound
deficits in oxygen delivery to less well-autoregulated tissue beds. In
the other tissues, HbO2 desaturation is unobserved with cerebral
oximetry or pulse oximetry monitors (red dashed arrows). Similarly,
treatments given to correct cerebral desaturations have unnoticed
benefits to other organs.
ischemia, myocardial infarction, sepsis, and a wide range of
infectious and inflammatory conditions.71 There is a large
body of literature on this topic, and we will not review it
here except to direct the reader to a review by Thompson
et al.72 that describes how hypoxia modulates neutrophil
function via the hypoxia-inducible factor signaling system.
Activation of lymphocytes or monocytes in peripheral tis-
sues and their subsequent migration into remote organs
such as brain is a second possibility. In this later case, it has
been proposed that activated monocytes and inflammation
mediate cognitive changes after major surgery, via mecha-
nisms involving nicotinic acetylcholine receptor signaling
and other factors.73–77 Tissue damaged by surgery or sub-
jected to ischemia or impaired oxygen delivery is the source
of these proinflammatory cells.
We will next review what is currently known about the
association between tissue oximetry and global outcomes.
Cerebral Saturation Predicts Cardiopulmonary,
Mortality, and Other Complications in Cardiac
Surgery Patients
Most of the studies to date concerning the relationship of tis-
sue oxygenation to remote outcomes have involved cardiac
surgery patients, probably because this population of patients
was among the first to be monitored with cerebral oximeters.
In a prospective randomized study of 200 elective coronary
artery bypass graft patients, Murkin et al.3 compared a blinded
control group and a group with interventions responding to
a decrease in cerebral saturation. Cerebral desaturation was
more prolonged in the control group, and treatment of satu-
rations resulted in a lower incidence of major organ morbid-
ity, mortality, and length of ICU length of stay. In aortic arch
surgery, Fischer et al.78 found significant associations between
cerebral saturation and major complications.
anesthesia & analgesia
 Tissue Oximetry and Clinical Outcomes
In cardiac surgery, low baseline brain oxygenation in
1200 patients undergoing cardiopulmonary bypass was
associated with mortality and other adverse outcomes such
as cardiopulmonary dysfunction.79 Similarly, a study from
Sun et al.80 examined the association between mortality and
cerebral saturation values <60% in approximately 2100 car-
diac surgery patients. Sun et al. found increased mortality
in the approximately 600 patients in the study group with
low cerebral saturation. Although in both studies, low cere-
bral oxygen saturation was associated with more significant
comorbidities, a multivariate analysis still showed a link
between low saturations and mortality.
Outcomes in Noncardiac Surgery
The earliest group of studies concerning cerebral satura-
tion and outcomes established only an association between
tissue oxygenation and outcomes and were limited by the
difficulty of establishing if other (possibly unmeasured
and possibly not modifiable) factors were actually driv-
ing outcomes. In the cardiac surgery studies cited above,
low cardiac output may be a nonmodifiable cause of low
cerebral saturation.63 More recent studies have begun to
explore whether tissue oxygenation is actually causing the
outcomes examined and whether interventions to correct
the desaturations/tissue hypoxia can avert the undesirable
outcomes. In such an interventional study, Casati et al.61 fol-
lowed elderly surgical patients undergoing major abdomi-
nal surgery. In these patients, low cerebral saturation (“area
under the curve” for saturations <75% of baseline) was
associated with longer postanesthesia care unit stays and
hospital admissions.61 The effect of interventions used to
treat the desaturations observed in the study in 2005 by
Casati et al. (blood pressure support or propofol to decrease
cerebral metabolic rate) had significant effects on outcomes.
Hospital length of stay was reduced in the intervention
group, with an average of 10 days (7–23 days) in a group
in which cerebral saturation was treated with an interven-
tion and 24 days (7–53 days) in the control (no interven-
tion) group. A similar prospective observational study by
Casati et al.62 in 2007 confirmed that low cerebral satura-
tion (in this case <50%) and a reduced mental status score
predicted longer hospital stay in elderly abdominal surgery
patients. Larger prospective randomized studies are neces-
sary. Identifying patients at risk may be useful, given the
simplicity of cerebral oximetry measurements.
Muscle Oxygenation and Prediction of Outcomes
in Sepsis or Trauma
This is an active area of current research, with >40 publi-
cations in the past 10 years reporting associations between
muscle tissue NIRS and a range of outcomes, including vol-
ume responsiveness and prognosis in trauma, sepsis, and
critical illness. These efforts have been based on the goal of
identifying markers of tissue oxygen deficits during early
management to guide therapy and establish prognosis for
conditions that continue to have a high mortality rate.81,82
Before advances in measuring tissue oxygenation with
NIRS, it was established that low values of central venous
saturation and acid-base disturbances correlated with poor
clinical outcomes.83,84 However, the changes in venous
XXX 2016 • Volume XXX • Number XXX www.anesthesia-analgesia.org
7
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
saturation and other variables were sometimes delayed and
thus not useful for guiding early treatment or establish-
ing early prognosis. Furthermore, overlap in the ranges of
venous saturation in controls and sick patients makes these
variables insensitive predictors of outcomes for individual
patients.
Measuring tissue oxygenation in the thenar muscle of the
hand as an indicator tissue for systemic stress is a relatively
more recent methodology than measuring brain oxygen-
ation. As is true with cerebral oxygenation, a large overlap
of tissue oxygen saturation (Sto2) values between healthy
subjects and septic shock patients has limited the utility
of this variable for clinical care. The combination of tissue
oxygenation and a vascular occlusion test (VOT) has been
proposed to increase the sensitivity and predictive power,
with slower reoxygenation after VOT occurring in trauma
patients compared with controls.85 The VOT involves inflat-
ing a cuff on the arm and measuring the changes in muscle
tissue oxygenation; the rate of change in Sto2 during this
test is then used as an index compared with various out-
comes. The rate of change in saturation in a VOT is greater
in patients with septic shock than in healthy volunteers
and can be a marker for death and organ system failure.86
The physiologic validity of measuring Sto2 at the skeletal
muscle of the thenar eminence as a marker for outcomes
is partly based on the physiology of this tissue bed, as well
as the convenience of being able to do VOTs on the arm.
Also, signal contribution from skin and fat is typically small
in the thenar area. With muscle oximetry involving NIRS,
one has to be aware that myoglobin absorbs light at similar
near-infrared wavelengths as Hb and although oxymyoglo-
bin and deoxymyoglobin have overlapping spectra,87 the
oxygen affinity (P50) of myoglobin is only approximately 2.4
mm Hg compared with approximately 26 mm Hg for Hb.
Davis and Barstow88 estimated that the contribution of myo-
globin to the absorbance of light at NIRS wavelengths may
be as much as 50% to 70%.
In trauma patients, initial tissue desaturation in periph-
eral muscle correlates with central venous oxygen, acid-
base disturbance, elevated lactic acid, and hemorrhagic
shock.89,90 Studies involving thenar muscle NIRS in trauma
patients have identified associations between low NIRS val-
ues and organ dysfunction.91–94 On the basis of the number
of published studies, a meta-analysis on the value of NIRS
readings in trauma to predict organ dysfunction could be
performed.
FUTURE DIRECTIONS AND SUGGESTIONS FOR THE
DESIGN OF CLINICAL TRIALS
Wider use of tissue oximetry should be driven not by the
uncritical attitude that “knowing something about tissue
oxygenation is better than knowing nothing”; rather, evi-
dence-based studies are needed to support the use of this
moderately expensive but noninvasive technology. We have
identified 4 areas where additional knowledge is needed:
Technical Knowledge Concerning Exactly What
NIRS Is Measuring
Evidence that NIRS tracks tissue oxygenation under all
relevant conditions of patient and physiologic variables,
 E REVIEW ARTICLE
including skin pigment, prevailing arterial Pco2, and other
variables is often missing. Also, more information is needed
about the sensitivity and specificity in detecting true altera-
tions in the oxygenation state of a tissue. In the case of
cerebral oximeters, this is essentially the accuracy of mea-
surement of the cerebral oximeter compared with a gold
standard reference. For cerebral oximeters, the gold stan-
dard reference is currently a prespecified mixture of cerebral
mixed venous blood and arterial oxygen saturation. This
reference standard probably needs to be adjusted to account
for changes in the arterial:venous blood volume assump-
tions during hypoxia and alterations in carbon dioxide. This
might be difficult in clinical situations because independent
assessments of carbon dioxide and oxygenation would need
to be incorporated into the calculation algorithm. Although
we have proposed that something along these lines might
reduce between-individual variation in cerebral oximeter
readings,45 whether this would do so in practice remains to
be determined.
Physiologic Knowledge, Including Definition
of Thresholds for Tissue Damage That Are
Associated with Adverse Outcomes
This is a critical issue that remains unresolved. Previous
studies have mostly used arbitrary cutoffs for analysis.
One difficulty is that injury thresholds may vary among
individuals because of differences in age, cerebral metabo-
lism, anesthetic and sedative regimen, body temperature,
and other factors. However, if the technology accounts for
changes in oxygenation because of changes in flow and
blood volume in the sensor field, then autoregulation of
cerebral blood flow should not be a confounder in estab-
lishing threshold values. The other issue is the “area under
the curve” problem: Is 3 hours at a 10% reduced cerebral
saturation equivalent in injury to 1 hour at 30%? Greater
knowledge in this area is critical for study design because
interventions need to be assigned to an appropriate thresh-
old and duration for potential tissue injury. Without appro-
priate thresholds, interventions will be initiated that are
either unneeded or too late to alter outcomes. Furthermore,
interventions have their own risks; for example, vasopres-
sors may decrease blood flow in unintended ways and
transfusions expose patients to a variety of risks.
A Deeper Understanding of the Physiology of the
Determinants of Tissue Oxygen Delivery
Knowing what interventions are most effective in resolv-
ing potentially clinically important desaturations is a key
issue. The following types of questions are relevant: Is a
vasoconstrictive medication (e.g., phenylephrine) sufficient
to restore cerebral oxygenation, or are increases in cardiac
output (norepinephrine) and increases in oxygen delivery
(e.g., transfusion to increase oxygen-carrying capacity) to
tissues more beneficial? Does banked blood (with reduced
P50) cause problems? Are crystalloids or colloids equally
effective?
Improved Clinical Study Design
It is very important that clinical trials test whether goal-
directed therapy based on tissue oxygenation measurements
8   
www.anesthesia-analgesia.org
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
improves outcomes. For these types of studies, a powerful
study design is to randomly assign patients to have con-
tinuously recorded but blinded or unblinded oximetry dis-
played and prespecified goal-directed therapy. An example
of goal-directed therapy would be to increase arterial blood
pressure if tissue saturation decreased by 10% from a base-
line; if the tissue saturation did not respond, then another
intervention would be specified. The control group would
have standard clinical care, but the tissue oximetry reading
would not be displayed to the clinician. Specified outcomes
would then be compared in the 2 groups. The knowledge
created by this type of study would be an important step
in evaluating the clinical value of tissue oximetry. Finally,
we need more studies concerning the frequency and depth
of desaturation events across the perioperative period. This
knowledge is essential because even if effective interven-
tions are used during intraoperative tissue desaturation
events, unobserved and untreated postoperative desatura-
tions may be driving outcomes.
CONCLUSIONS
We believe that tissue oximetry with NIRS has the potential
to be a clinical monitor that can predict both tissue-specific
and global outcomes and can guide therapy to avert poor
outcomes. Because tissue oxygenation is coupled to organ
function much more directly than is arterial oxygen satura-
tion, it is not surprising that this could be true. However,
establishing a linkage between NIRS oximetry and clinical
outcomes has been challenging because it depends on being
able to show that averting local tissue hypoxemia improves
outcomes. The design of studies to do that must involve
a nonintervention group in which the periods of tissue
hypoxemia are blinded to the caregiver, and an intervention
group in which a protocol is used to treat the event that is
observed. Very few studies to date have met that standard.
Other areas of improved understanding that are needed to
move the field forward include the causes and frequency
of decreased tissue oxygen levels in various organs and a
better understanding of what interventions (e.g., vasoac-
tive medications, fluids, or blood) best treat each organ or
regional hypoxemic event. E
DISCLOSURES
Name: Philip Bickler, MD, PhD.
Contribution: This author wrote all sections of the manuscript
except for the specific contributions noted below for co-authors
and read, edited, revised, and approved the entire manuscript.
Conflicts of Interest: Philip Bickler has received research fund-
ing from Masimo Inc., Nonin Medical Inc., Bluepoint Medical,
CAS Medical, and numerous other pulse oximeter manufac-
turers. These sources have funded basic science research and
human physiology experiments. No direct funding of this man-
uscript or research therein was derived from corporate sources.
Attestation: Philip Bickler has seen the original study data,
reviewed the analysis of the data, and approved the final
manuscript.
Name: John Feiner, MD.
Contribution: This author helped design the study, conduct the
study, analyze the data, and write the manuscript. He wrote the
section of the manuscript concerning tissue-specific outcomes
with tissue oximetry.
anesthesia & analgesia
 Tissue Oximetry and Clinical Outcomes
Conflicts of Interest: John Feiner received research funding
from Masimo Inc., Nonin Medical Inc., Bluepoint Medical,
Xhale, and CAS Medical. Various pulse oximeter manufactur-
ers funded our laboratory to perform human accuracy studies.
None of these sources funded anything to do with the concep-
tion or writing of this manuscript.
Attestation: John Feiner has seen the original study data,
reviewed the analysis of the data, and approved the final
manuscript.
Name: Mark Rollins, MD, PhD.
Contribution: This author wrote the section of the manuscript
concerning the technology for assessing tissue oxygenation and
read, edited, and approved the entire manuscript.
Conflicts of Interest: The author declares no conflicts of interest.
Attestation: Mark Rollins has read and approved the final
manuscript.
Name: Lingzhong Meng, MD.
Contribution: This author wrote the section of the paper on the
incidence of desaturation events. He was the principal investi-
gator of a study in spine surgery patients. He read, edited, and
approved the entire manuscript.
Conflicts of Interest: Lingzhong Meng received research fund-
ing from CAS medical for an on-going study on tissue oximetry
and patient outcomes.
Attestation: Lingzhong Meng has read and approved the final
manuscript.
This manuscript was handled by: Maxime Cannesson, MD, PhD.
REFERENCES
1. Ballard C, Jones E, Gauge N, Aarsland D, Nilsen OB, Saxby BK,
Lowery D, Corbett A, Wesnes K, Katsaiti E, Arden J, Amoako D,
Amaoko D, Prophet N, Purushothaman B, Green D. Optimised
anaesthesia to reduce post operative cognitive decline (POCD)
in older patients undergoing elective surgery, a randomised
controlled trial. PLoS One 2012;7:e37410
2. Colak Z, Borojevic M, Bogovic A, Ivancan V, Biocina B, Majeric-
Kogler V. Influence of intraoperative cerebral oximetry moni-
toring on neurocognitive function after coronary artery bypass
surgery: a randomized, prospective study. Eur J Cardiothorac
Surg 2015;47:447–54
3. Murkin JM, Adams SJ, Novick RJ, Quantz M, Bainbridge D,
Iglesias I, Cleland A, Schaefer B, Irwin B, Fox S. Monitoring
brain oxygen saturation during coronary bypass surgery: a ran-
domized, prospective study. Anesth Analg 2007;104:51–8
4. Slater JP, Guarino T, Stack J, Vinod K, Bustami RT, Brown JM
III, Rodriguez AL, Magovern CJ, Zaubler T, Freundlich K, Parr
GV. Cerebral oxygen desaturation predicts cognitive decline
and longer hospital stay after cardiac surgery. Ann Thorac Surg
2009;87:36–44
5. Sun Z, Sessler DI, Dalton JE, Devereaux PJ, Shahinyan A, Naylor
AJ, Hutcherson MT, Finnegan PS, Tandon V, Darvish-Kazem S,
Chugh S, Alzayer H, Kurz A. Postoperative hypoxemia is com-
mon and persistent: a prospective blinded observational study.
Anesth Analg 2015;121:709–15
6. Taenzer AH, Pyke JB, McGrath SP, Blike GT. Impact of pulse
oximetry surveillance on rescue events and intensive care unit
transfers: a before-and-after concurrence study. Anesthesiology
2010;112:282–7
7. Moller JT, Johannessen NW, Espersen K, Ravlo O, Pedersen
BD, Jensen PF, Rasmussen NH, Rasmussen LS, Pedersen T,
Cooper JB. Randomized evaluation of pulse oximetry in 20,802
patients: II. Perioperative events and postoperative complica-
tions. Anesthesiology 1993;78:445–53
8. Pedersen T, Nicholson A, Hovhannisyan K, Møller AM, Smith
AF, Lewis SR. Pulse oximetry for perioperative monitoring.
Cochrane Database Syst Rev 2014;3:CD002013
9. Shah A, Shelley KH. Is pulse oximetry an essential tool or just
another distraction? The role of the pulse oximeter in modern
anesthesia care. J Clin Monit Comput 2013;27:235–42
XXX 2016 • Volume XXX • Number XXX www.anesthesia-analgesia.org
9
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
10. Greenberg SB, Murphy G, Alexander J, Fasanella R, Garcia A,
Vender J. Cerebral desaturation events in the intensive care unit
following cardiac surgery. J Crit Care 2013;28:270–6
11. Yadeau JT, Liu SS, Bang H, Shaw PM, Wilfred SE, Shetty T,
Gordon M. Cerebral oximetry desaturation during shoulder
surgery performed in a sitting position under regional anesthe-
sia. Can J Anaesth 2011;58:986–92
12. Murphy GS, Szokol JW, Marymont JH, Greenberg SB, Avram
MJ, Vender JS, Vaughn J, Nisman M. Cerebral oxygen desatu-
ration events assessed by near-infrared spectroscopy during
shoulder arthroscopy in the beach chair and lateral decubitus
positions. Anesth Analg 2010;111:496–505
13. Clark LC Jr, Clark EW. A personalized history of the Clark oxy-
gen electrode. Int Anesthesiol Clin 1987;25:1–29
14. Fatt I. Polarographic Oxygen Sensors: Its Theory of Operation
and Its Application in Biology, Medicine, and Technology.
Malabar, FL: Krieger, 1982
15. Opitz N, Lübbers DW. Theory and development of fluores-
cence-based optochemical oxygen sensors: oxygen optodes. Int
Anesthesiol Clin 1987;25:177–97
16. Hopf HW, Rollins MD. Wounds: an overview of the role of oxy-
gen. Antioxid Redox Signal 2007;9:1183–92
17. Ikossi DG, Knudson MM, Morabito DJ, Cohen MJ, Wan JJ,
Khaw L, Stewart CJ, Hemphill C, Manley GT. Continuous mus-
cle tissue oxygenation in critically injured patients: a prospec-
tive observational study. J Trauma 2006;61:780–8
18. Vaupel P, Höckel M, Mayer A. Detection and characteriza-
tion of tumor hypoxia using pO2 histography. Antioxid Redox
Signal 2007;9:1221–35
19. Lazaridis C, Andrews CM. Brain tissue oxygenation, lactate-
pyruvate ratio, and cerebrovascular pressure reactivity moni-
toring in severe traumatic brain injury: systematic review and
viewpoint. Neurocrit Care 2014;21:345–55
20. Yuan Q, Wu X, Sun Y, Yu J, Li Z, Du Z, Mao Y, Zhou L, Hu
J. Impact of intracranial pressure monitoring on mortality in
patients with traumatic brain injury: a systematic review and
meta-analysis. J Neurosurg 2015;122:574–87
21. Deshaies EM, Jacobsen W, Singla A, Li F, Gorji R. Brain tissue
oxygen monitoring to assess reperfusion after intra-arterial
treatment of aneurysmal subarachnoid hemorrhage-induced
cerebral vasospasm: a retrospective study. AJNR Am J
Neuroradiol 2012;33:1411–5
22. De Georgia MA. Brain tissue oxygen monitoring in neurocriti-
cal care. J Intensive Care Med 2015;30:473–83
23. Yu M, Chapital A, Ho HC, Wang J, Takanishi D Jr. A prospective
randomized trial comparing oxygen delivery versus transcuta-
neous pressure of oxygen values as resuscitative goals. Shock
2007;27:615–22
24. Swartz HM, Williams BB, Zaki BI, Hartford AC, Jarvis LA,
Chen EY, Comi RJ, Ernstoff MS, Hou H, Khan N, Swarts SG,
Flood AB, Kuppusamy P. Clinical EPR: unique opportunities
and some challenges. Acad Radiol 2014;21:197–206
25. Tisdall MM, Smith M. Cerebral microdialysis: research tech-
nique or clinical tool. Br J Anaesth 2006;97:18–25
26. Burša F, Pleva L, Máca J, Sklienka P, Ševčík P. Tissue ischemia
microdialysis assessments following severe traumatic haemor-
rhagic shock: lactate/pyruvate ratio as a new resuscitation end
point? BMC Anesthesiol 2014;14:118
27. Grinspan ZM, Pon S, Greenfield JP, Malhotra S, Kosofsky BE.
Multimodal monitoring in the pediatric intensive care unit:
new modalities and informatics challenges. Semin Pediatr
Neurol 2014;21:291–8
28. Chance B. Spectrophotometry of intracellular respiratory pig-
ments. Science 1954;120:767–75
29. Jöbsis FF. Noninvasive, infrared monitoring of cerebral and
myocardial oxygen sufficiency and circulatory parameters.
Science 1977;198:1264–7
30. Kakihana Y, Matsunaga A, Yasuda T, Imabayashi T, Kanmura Y,
Tamura M. Brain oxymetry in the operating room: current sta-
tus and future directions with particular regard to cytochrome
oxidase. J Biomed Opt 2008;13:033001
31. Cooper CE, Cope M, Springett R, Amess PN, Penrice J,
Tyszczuk L, Punwani S, Ordidge R, Wyatt J, Delpy DT. Use
of mitochondrial inhibitors to demonstrate that cytochrome
 E REVIEW ARTICLE
oxidase near-infrared spectroscopy can measure mitochondrial
dysfunction noninvasively in the brain. J Cereb Blood Flow
Metab 1999;19:27–38
32. Lee J, Armstrong J, Kreuter K, Tromberg BJ, Brenner M. Non-
invasive in vivo diffuse optical spectroscopy monitoring of cya-
nide poisoning in a rabbit model. Physiol Meas 2007;28:1057–66
33. Boas DA, Franceschini MA. Haemoglobin oxygen saturation as
a biomarker: the problem and a solution. Philos Trans A Math
Phys Eng Sci 2011;369:4407–24
34. Devor A, Sakadžić S, Srinivasan VJ, Yaseen MA, Nizar K,
Saisan PA, Tian P, Dale AM, Vinogradov SA, Franceschini MA,
Boas DA. Frontiers in optical imaging of cerebral blood flow
and metabolism. J Cereb Blood Flow Metab 2012;32:1259–76
35. Bainbridge A, Tachtsidis I, Faulkner SD, Price D, Zhu T, Baer E,
Broad KD, Thomas DL, Cady EB, Robertson NJ, Golay X. Brain
mitochondrial oxidative metabolism during and after cerebral
hypoxia-ischemia studied by simultaneous phosphorus mag-
netic-resonance and broadband near-infrared spectroscopy.
Neuroimage 2014;102 Pt 1:173–83
36. Kolyva C, Ghosh A, Tachtsidis I, Highton D, Cooper CE, Smith
M, Elwell CE. Cytochrome c oxidase response to changes in
cerebral oxygen delivery in the adult brain shows higher brain-
specificity than haemoglobin. Neuroimage 2014;85 Pt 1:234–44
37. Liu S, Shah SJ, Wilmes LJ, Feiner J, Kodibagkar VD, Wendland
MF, Mason RP, Hylton N, Hopf HW, Rollins MD. Quantitative
tissue oxygen measurement in multiple organs using 19F MRI in
a rat model. Magn Reson Med 2011;66:1722–30
38. Mason RP, Zhao D, Pacheco-Torres J, Cui W, Kodibagkar
VD, Gulaka PK, Hao G, Thorpe P, Hahn EW, Peschke P.
Multimodality imaging of hypoxia in preclinical settings. Q J
Nucl Med Mol Imaging 2010;54:259–80
39. Baudelet C, Gallez B. How does blood oxygen level-dependent
(BOLD) contrast correlate with oxygen partial pressure (pO2)
inside tumors? Magn Reson Med 2002;48:980–6
40. Kavec M, Usenius JP, Tuunanen PI, Rissanen A, Kauppinen RA.
Assessment of cerebral hemodynamics and oxygen extraction
using dynamic susceptibility contrast and spin echo blood oxy-
genation level-dependent magnetic resonance imaging: appli-
cations to carotid stenosis patients. Neuroimage 2004;22:258–67
41. Shiino A, Yamauchi H, Morikawa S, Inubushi T. Mapping
of cerebral metabolic rate of oxygen using DSC and BOLD
MR imaging: a preliminary study. Magn Reson Med Sci
2012;11:109–15
42. Gersing AS, Ankenbrank M, Schwaiger BJ, Toth V, Janssen I,
Kooijman H, Wunderlich S, Bauer JS, Zimmer C, Preibisch C.
Mapping of cerebral metabolic rate of oxygen using dynamic
susceptibility contrast and blood oxygen level dependent
MR imaging in acute ischemic stroke. Neuroradiology
2015;57:1253–61
43. Hirsch NM, Toth V, Förschler A, Kooijman H, Zimmer C,
Preibisch C. Technical considerations on the validity of blood
oxygenation level-dependent-based MR assessment of vascular
deoxygenation. NMR Biomed 2014;27:853–62
44. Ferrari M, Quaresima V. A brief review on the history of human
functional near-infrared spectroscopy (fNIRS) development
and fields of application. Neuroimage 2012;63:921–35
45. Bickler PE, Feiner JR, Rollins MD. Factors affecting the perfor-
mance of 5 cerebral oximeters during hypoxia in healthy volun-
teers. Anesth Analg 2013;117:813–23
46. Ito H, Ibaraki M, Kanno I, Fukuda H, Miura S. Changes in the
arterial fraction of human cerebral blood volume during hyper-
capnia and hypocapnia measured by positron emission tomog-
raphy. J Cereb Blood Flow Metab 2005;25:852–7
47. Davie SN, Grocott HP. Impact of extracranial contamination on
regional cerebral oxygen saturation: a comparison of three cere-
bral oximetry technologies. Anesthesiology 2012;116:834–40
48. Greenberg S, Murphy G, Shear T, Patel A, Simpson A, Szokol J,
Avram MJ, Vender J. Extracranial contamination in the INVOS
5100C versus the FORE-SIGHT ELITE cerebral oximeter: a
prospective observational crossover study in volunteers. Can J
Anaesth 2016;63:24–30
49. Sørensen H, Secher NH, Siebenmann C, Nielsen HB, Kohl-
Bareis M, Lundby C, Rasmussen P. Cutaneous vasoconstric-
tion affects near-infrared spectroscopy determined cerebral
10   
www.anesthesia-analgesia.org
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
oxygen saturation during administration of norepinephrine.
Anesthesiology 2012;117:263–70
50. Wijayatilake DS, Talati C, Panchatsharam S. The monitoring
and management of severe traumatic brain injury in the United
Kingdom: is there a consensus? A national survey. J Neurosurg
Anesthesiol 2015;27:241–5
51. Gupta RG, Hartigan SM, Kashiouris MG, Sessler CN, Bearman
GM. Early goal-directed resuscitation of patients with sep-
tic shock: current evidence and future directions. Crit Care
2015;19:286
52. Harilall Y, Adam JK, Biccard BM, Reddi A. The effect of opti-
mising cerebral tissue oxygen saturation on markers of neu-
rological injury during coronary artery bypass graft surgery.
Heart Lung Circ 2014;23:68–74
53. Pennekamp CW, Bots ML, Kappelle LJ, Moll FL, de Borst GJ.
The value of near-infrared spectroscopy measured cerebral
oximetry during carotid endarterectomy in perioperative stroke
prevention. A review. Eur J Vasc Endovasc Surg 2009;38:539–45
54. Pennekamp CW, Immink RV, den Ruijter HM, Kappelle LJ, Bots
ML, Buhre WF, Moll FL, de Borst GJ. Near-infrared spectros-
copy to indicate selective shunt use during carotid endarterec-
tomy. Eur J Vasc Endovasc Surg 2013;46:397–403
55. Pennekamp CW, Immink RV, den Ruijter HM, Kappelle LJ,
Ferrier CM, Bots ML, Buhre WF, Moll FL, de Borst GJ. Near-
infrared spectroscopy can predict the onset of cerebral hyper-
perfusion syndrome after carotid endarterectomy. Cerebrovasc
Dis 2012;34:314–21
56. Pennekamp CW, Moll FL, de Borst GJ. The potential benefits
and the role of cerebral monitoring in carotid endarterectomy.
Curr Opin Anaesthesiol 2011;24:693–7
57. Yao FS, Tseng CC, Ho CY, Levin SK, Illner P. Cerebral oxygen
desaturation is associated with early postoperative neuropsy-
chological dysfunction in patients undergoing cardiac surgery.
J Cardiothorac Vasc Anesth 2004;18:552–8
58. de Tournay-Jetté E, Dupuis G, Bherer L, Deschamps A, Cartier
R, Denault A. The relationship between cerebral oxygen satu-
ration changes and postoperative cognitive dysfunction in
elderly patients after coronary artery bypass graft surgery.
J Cardiothorac Vasc Anesth 2011;25:95–104
59. Hong SW, Shim JK, Choi YS, Kim DH, Chang BC, Kwak YL.
Prediction of cognitive dysfunction and patients’ outcome fol-
lowing valvular heart surgery and the role of cerebral oximetry.
Eur J Cardiothorac Surg 2008;33:560–5
60. Reents W, Muellges W, Franke D, Babin-Ebell J, Elert O.
Cerebral oxygen saturation assessed by near-infrared spectros-
copy during coronary artery bypass grafting and early postop-
erative cognitive function. Ann Thorac Surg 2002;74:109–14
61. Casati A, Fanelli G, Pietropaoli P, Proietti R, Tufano R, Danelli
G, Fierro G, Fierro G, De Cosmo G, Servillo G. Continuous
monitoring of cerebral oxygen saturation in elderly patients
undergoing major abdominal surgery minimizes brain expo-
sure to potential hypoxia. Anesth Analg 2005;101:740–7
62. Casati A, Fanelli G, Pietropaoli P, Proietti R, Tufano R, Montanini
S, Danelli G, Nuzzi M, Mentegazzi F, Torri G, Martani C,
Spreafico E, Fierro G, Pugliese F, De Cosmo G, Aceto P, Servillo
G, Monaco F; Collaborative Italian Study Group on Anaesthesia
in Elderly Patients. Monitoring cerebral oxygen saturation in
elderly patients undergoing general abdominal surgery: a pro-
spective cohort study. Eur J Anaesthesiol 2007;24:59–65
63. Lin R, Zhang F, Xue Q, Yu B. Accuracy of regional cerebral
oxygen saturation in predicting postoperative cognitive dys-
function after total hip arthroplasty: regional cerebral oxygen
saturation predicts POCD. J Arthroplasty 2013;28:494–7
64. Genbrugge C, Dens J, Meex I, Boer W, Eertmans W, Sabbe M,
Jans F, De Deyne C. Regional cerebral oximetry during car-
diopulmonary resuscitation: useful or useless? J Emerg Med
2016;50:198–207
65. Ibrahim AW, Trammell AR, Austin H, Barbour K, Onuorah E,
House D, Miller HL, Tutt C, Combs D, Phillips R, Dickert NW,
Zafari AM. Cerebral oximetry as a real-time monitoring tool to
assess quality of in-hospital cardiopulmonary resuscitation and
post cardiac arrest care. J Am Heart Assoc 2015;4:e001859
66. Hyttel-Sorensen S, Pellicer A, Alderliesten T, Austin T, van Bel
F, Benders M, Claris O, Dempsey E, Franz AR, Fumagalli M,
anesthesia & analgesia
 Tissue Oximetry and Clinical Outcomes
Gluud C, Grevstad B, Hagmann C, Lemmers P, van Oeveren W,
Pichler G, Plomgaard AM, Riera J, Sanchez L, Winkel P, Wolf
M, Greisen G. Cerebral near infrared spectroscopy oximetry in
extremely preterm infants: phase II randomised clinical trial.
BMJ 2015;350:g7635
67. Eriksson EA, Barletta JF, Figueroa BE, Bonnell BW, Sloffer CA,
Vanderkolk WE, McAllen KJ, Ott M. The first 72 hours of brain
tissue oxygenation predicts patient survival with traumatic
brain injury. J Trauma Acute Care Surg 2012;72:1345–9
68. Lewis S, Wong M, Myburgh J, Reilly P. Determining cerebral
perfusion pressure thresholds in severe head trauma. Acta
Neurochir Suppl 1998;71:174–6
69. Rosenthal G, Furmanov A, Itshayek E, Shoshan Y, Singh V.
Assessment of a noninvasive cerebral oxygenation monitor
in patients with severe traumatic brain injury. J Neurosurg
2014;120:901–7
70. Meng L, Hou W, Chui J, Han R, Gelb AW. Cardiac output and
cerebral blood flow: the integrated regulation of brain perfu-
sion in adult humans. Anesthesiology 2015;123:1198–208
71. Laveti D, Kumar M, Hemalatha R, Sistla R, Naidu VG, Talla
V, Verma V, Kaur N, Nagpal R. Anti-inflammatory treatments
for chronic diseases: a review. Inflamm Allergy Drug Targets
2013;12:349–61
72. Thompson AA, Binham J, Plant T, Whyte MK, Walmsley SR.
Hypoxia, the HIF pathway and neutrophilic inflammatory
responses. Biol Chem 2013;394:471–7
73. Degos V, Maze M, Vacas S, Hirsch J, Guo Y, Shen F, Jun K, van
Rooijen N, Gressens P, Young WL, Su H. Bone fracture exacerbates
murine ischemic cerebral injury. Anesthesiology 2013;118:1362–72
74. Degos V, Vacas S, Han Z, van Rooijen N, Gressens P, Su H,
Young WL, Maze M. Depletion of bone marrow-derived mac-
rophages perturbs the innate immune response to surgery and
reduces postoperative memory dysfunction. Anesthesiology
2013;118:527–36
75. Feng X, Degos V, Koch LG, Britton SL, Zhu Y, Vacas S, Terrando
N, Nelson J, Su X, Maze M. Surgery results in exaggerated and
persistent cognitive decline in a rat model of the Metabolic
Syndrome. Anesthesiology 2013;118:1098–105
76. Vacas S, Degos V, Feng X, Maze M. The neuroinflammatory
response of postoperative cognitive decline. Br Med Bull
2013;106:161–78
77. Vacas S, Degos V, Tracey KJ, Maze M. High-mobility group
box 1 protein initiates postoperative cognitive decline by
engaging bone marrow-derived macrophages. Anesthesiology
2014;120:1160–7
78. Fischer GW, Lin HM, Krol M, Galati MF, Di Luozzo G, Griepp
RB, Reich DL. Noninvasive cerebral oxygenation may predict
outcome in patients undergoing aortic arch surgery. J Thorac
Cardiovasc Surg 2011;141:815–21
79. Heringlake M, Garbers C, Käbler JH, Anderson I, Heinze
H, Schön J, Berger KU, Dibbelt L, Sievers HH, Hanke T.
Preoperative cerebral oxygen saturation and clinical outcomes
in cardiac surgery. Anesthesiology 2011;114:58–69
80. Sun X, Ellis J, Corso PJ, Hill PC, Lowery R, Chen F, Lindsay J.
Mortality predicted by preinduction cerebral oxygen saturation
after cardiac operation. Ann Thorac Surg 2014;98:91–6
XXX 2016 • Volume XXX • Number XXX www.anesthesia-analgesia.org
11
Copyright © 2016 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
81. Kauvar DS, Wade CE. The epidemiology and modern manage-
ment of traumatic hemorrhage: US and international perspec-
tives. Crit Care 2005;9(Suppl 5):S1–9
82. Stoller J, Halpin L, Weis M, Aplin B, Qu W, Georgescu C,
Nazzal M. Epidemiology of severe sepsis: 2008–2012. J Crit
Care 2016;31:58–62
83. Moore FA, McKinley BA, Moore EE, Nathens AB, West M,
Shapiro MB, Bankey P, Freeman B, Harbrecht BG, Johnson
JL, Minei JP, Maier RV; Inflammation and the Host Response
to Injury Collaborative Research Program. Inflammation and
the Host Response to Injury, a large-scale collaborative project:
patient-oriented research core—standard operating procedures
for clinical care. III. Guidelines for shock resuscitation. J Trauma
2006;61:82–9
84. Paladino L, Sinert R, Wallace D, Anderson T, Yadav K, Zehtabchi
S. The utility of base deficit and arterial lactate in differentiating
major from minor injury in trauma patients with normal vital
signs. Resuscitation 2008;77:363–8
85. Gómez H, Torres A, Polanco P, Kim HK, Zenker S, Puyana
JC, Pinsky MR. Use of non-invasive NIRS during a vascular
occlusion test to assess dynamic tissue O2 saturation response.
Intensive Care Med 2008;34:1600–7
86. Payen D, Luengo C, Heyer L, Resche-Rigon M, Kerever S,
Damoisel C, Losser MR. Is thenar tissue hemoglobin oxygen
saturation in septic shock related to macrohemodynamic vari-
ables and outcome? Crit Care 2009;13(suppl 5):S6
87. Schenkman KA, Marble DR, Burns DH, Feigl EO. Myoglobin
oxygen dissociation by multiwavelength spectroscopy. J Appl
Physiol (1985) 1997;82:86–92
88. Davis ML, Barstow TJ. Estimated contribution of hemoglobin
and myoglobin to near infrared spectroscopy. Respir Physiol
Neurobiol 2013;186:180–7
89. Crookes BA, Cohn SM, Bloch S, Amortegui J, Manning R, Li
P, Proctor MS, Hallal A, Blackbourne LH, Benjamin R, Soffer
D, Habib F, Schulman CI, Duncan R, Proctor KG. Can near-
infrared spectroscopy identify the severity of shock in trauma
patients? J Trauma 2005;58:806–13
90. McKinley BA, Marvin RG, Cocanour CS, Moore FA. Tissue
hemoglobin O2 saturation during resuscitation of traumatic
shock monitored using near infrared spectrometry. J Trauma
2000;48:637–42
91. Cohn SM. Potential benefit of vasopressin in resuscitation of
hemorrhagic shock. J Trauma 2007;62:S56–7
92. Duret J, Pottecher J, Bouzat P, Brun J, Harrois A, Payen JF,
Duranteau J. Skeletal muscle oxygenation in severe trauma
patients during haemorrhagic shock resuscitation. Crit Care
2015;19:141
93. Moore FA, Nelson T, McKinley BA, Moore EE, Nathens
AB, Rhee P, Puyana JC, Beilman GJ, Cohn SM; StO2 Study
Group. Massive transfusion in trauma patients: tissue hemo-
globin oxygen saturation predicts poor outcome. J Trauma
2008;64:1010–23
94. Nicks BA, Campons KM, Bozeman WP. Association of low non-
invasive near-infrared spectroscopic measurements during ini-
tial trauma resuscitation with future development of multiple
organ dysfunction. World J Emerg Med 2015;6:105–10