Journal of Veterinary Emergency and Critical Care 00(0) 2018, pp 1–11

Clinical Practice Review doi: 10.1111/vec.12749

Venous oxygen saturation in critical illness

Rebecca A.L. Walton, DVM, DACVECC and Bernie D. Hansen, DVM, MS, DACVECC, DACVIM
Abstract

Objective – To review clinically relevant features of systemic oxygen delivery and consumption and the tech-
nique and use of venous oxygenation monitoring in human and veterinary medicine.
Data sources – Veterinary and human peer-reviewed medical literature including scientific reviews, clinical and
laboratory research articles, and authors’ clinical research experience.
Summary – Measurement of venous hemoglobin oxygen saturation (venous oxygenation) provides insight into
the balance between oxygen supply and tissue demand. In people, measurement of venous oxygen saturation
can reveal decompensation that is missed by physical examination and other routinely monitored parameters.
Therefore, measurement of mixed or central venous oxygenation measurement may help guide therapy and
predict outcome of critically ill patients. In dogs, low central venous oxygen saturation has been associated with
impaired cardiopulmonary function and poor outcome in several small studies of experimental shock or severe
clinical illness, suggesting that monitoring this variable may assist the treatment of severe illness in this species
as well.
Conclusion – Venous oxygenation reflects systemic oxygenation status and can be used to guide treatment
and estimate prognosis in critically ill patients. Measurement of venous oxygenation in veterinary patients is
feasible and is a potentially valuable tool in the management of patients with severe disease. This review is
intended to increase the understanding and awareness of the potential role of venous oxygen measurement in
veterinary patients.

(J Vet Emerg Crit Care 2018; 00(0): 1–11) doi: 10.1111/vec.12749

Keywords: early goal directed therapy, hemodynamic monitoring, oxygen consumption, shock, venous
oxygenation

Introduction
Abbreviations
Matching oxygen delivery to metabolic demand is a fun-
CaO2 arterial oxygen content
damental requirement of the cardiopulmonary system,
CO cardiac output
and failure to meet tissue oxygenation requirements is
CPR cardiopulmonary resuscitation
characteristic of circulatory shock. In fact, shock is of-
CvO2 venous oxygen content
ten defined as a consequence of regional or global im-
DO2 oxygen delivery
pairment of oxygen delivery, tissue oxygen deficiency
EGDT early goal-directed therapy
(hypoxia), or impaired utilization of oxygen (dysoxia).1
MODS multi-organ dysfunction syndrome
Abnormalities of tissue oxygenation are responsible for
O2 ER oxygen extraction ratio
many of the physical abnormalities observed in shock,
PAC pulmonary artery catheter
and if uncorrected may ultimately result in the develop-
ScvO2 central venous hemoglobin saturation with
ment of multi-organ failure and death. Although clini-
oxygen
cians rely on interpretation of physical signs to evaluate
SO2 hemoglobin saturation with oxygen
patients and direct therapy for shock, those findings are
SvO2 saturation of mixed venous hemoglobin with
often imprecise and typically reflect downstream conse-
oxygen
quences of the underlying problem of insufficient tissue
V̇O2 oxygen consumption
oxygenation. Consequently, clinicians have long sought
methods to more directly assess the adequacy of oxy-
gen delivery and consumption in shock. One such ap-
From the Department of Clinical Sciences, North Carolina State University,
College of Veterinary Medicine, Raleigh, NC, 27607.
proach is to directly measure oxygen delivery and con-
sumption, and then calibrate treatment to ensure that
The authors declare no conflict of interests.
consumption is not limited by inadequate supply. Al-
Address correspondence and reprints to
Dr. Bernie Hansen, Department of Clinical Sciences, North Carolina State
though this may be accomplished directly by analysis of
University, College of Veterinary Medicine, 1052 William Moore Drive, respiratory uptake of oxygen and exhalation of carbon
Raleigh, NC 27607, USA.
Email: bdhansen@ncsu.edu
Submitted August 11, 2016; Accepted January 03, 2017.


C Veterinary Emergency and Critical Care Society 2018 1
R.A.L. Walton & B.D. Hansen

dioxide, the technical challenges associated with this rogate for SvO2 .4,10,11 Central venous hemoglobin satura-
method prompted interest in alternative methods of tion with oxygen is measured in the right atrium or a cen-
assessment.2 A major advance in the ability to measure tral vein, typically the superior vena cava (cranial vena
systemic oxygenation was accomplished with the intro- cava in quadrupeds). As previously reviewed by Rivers
duction of bedside pulmonary artery catheterization as et al, ScvO2 monitoring has been used as a prognostic in-
described by Swan and Ganz in 1970.3 Standard use of dicator in a variety of conditions in people with cardiac
the pulmonary artery catheter (PAC) requires insertion disease, trauma, and sepsis.12 Additionally, ScvO2 has
into the subclavian vein (or the external jugular vein in been used as a therapeutic endpoint and target of resus-
dogs and cats) then sequential advancement through the citation in early goal-directed therapy for severe sepsis
superior vena cava, right heart, and pulmonary artery and septic shock.13 The goal of this review is to present
where the tip provides access to pulmonary arterial the physiologic rationale for the measurement of venous
blood.3 oxygenation and describe its application in critical illness
Oxygen saturation of hemoglobin (SaO2 ) is the ratio in human and veterinary medicine.
of fully oxygenated hemoglobin to the total hemoglobin
in the blood capable of binding oxygen, measured via
oximetry or calculated from partial pressure of oxygen Oxygen Delivery, Oxygen Consumption, and Ve-
(pO2 ). The oxygen saturation of pulmonary arterial (or nous Oxygenation
mixed venous) hemoglobin is commonly referred to as Cellular energy is obtained from the Krebs cycle, electron
SvO2 , and is a function of the balance between oxygen transport, and oxidative phosphorylation, all of which
delivered to and consumed by all tissues of the body. require oxygen. Under physiologic conditions oxygen
By measuring cardiac output (CO) and the oxygen con- availability greatly exceeds oxygen consumption and its
tent of both mixed venous blood (CvO2 ) and arterial delivery to cells is not a rate-limiting process.12,14 Oxy-
blood (CaO2 ), one can calculate oxygen delivery (DO2 gen delivery is a product of the oxygen content of ar-
= CaO2 × CO) and also oxygen consumption (V̇O2 ; terial blood and cardiac output as represented by the
via rearrangement of the Fick equation V̇O2 = CO × equation:
[CaO2 – CvO2 ]).2,4 Although this approach provides a
quantitative accounting of oxygen supply and demand, DO2 = CO × [1.34 × Hgb × SaO2 ] + [0.003 × PaO2 ],
it is also possible to draw conclusions about the ade-
quacy of systemic oxygen delivery by evaluating SvO2 where DO2 = total oxygen delivery, CO = cardiac out-
even without knowing the absolute values for CO, DO2 , put, 1.34 = the Hüfner constant for human hemoglobin
and V̇O2 . oxygen binding capacity in mL/g, Hgb = hemoglobin
Because blood oxygen content is largely a product of concentration in g/L, SaO2 is the oxygen saturation of
hemoglobin concentration and saturation, SvO2 gained hemoglobin, and PaO2 is the partial pressure of oxygen
status as a monitored variable for critically ill people in arterial blood. The value for Hüfner’s constant varies
by the 1990s.5 Despite the relative simplicity of SvO2 with species and carboxyhemoglobin concentration and
measurement compared with determining CO and DO2 , is between 1.34 and 1.39 in dogs.15
obtaining this measurement still requires a PAC. By the Oxygen consumption varies between tissues and is af-
time widespread interest in SvO2 as a monitored vari- fected by metabolic rate, non-nutrient blood flow with
able developed, PACs were already routinely used to low metabolic exchange, and basal oxygen extraction.16
monitor other aspects of cardiopulmonary function in Absolute oxygen consumption by the resting human
critically ill humans and the incorporation of SvO2 mon- brain averages 52 mL/min, the heart 34 mL/min,
itoring was a logical and convenient addition that was splanchnic bed 83 mL/min, kidney 19 ml/min, skele-
met with some enthusiasm.6 However, the use of PACs tal muscle 57 mL/min, and skin 12 mL/min.17,18 Al-
declined following the publication of several studies in though, in general, organ blood flow will be regulated
the first decade of this century that failed to demonstrate to meet demand, there is not an absolute relationship
a beneficial effect of this invasive technique on patient between the two and consumption can vary widely at
outcome in several settings and also identified a rela- any given flow rate. For each organ the balance of sup-
tively high frequency of adverse events associated with ply and demand determines the amount of oxygen ex-
their use.7–9 tracted from the blood and the SO2 of blood leaving the
Because obtaining central venous blood is technically organ. Venous SO2 values for resting humans have been
simpler, less expensive, and less often associated with se- reported as 69% for the brain, 37% for the heart, 66%
rious complications than obtaining mixed venous blood for the liver/splanchnic bed, 92% for the kidneys, 71%
via PACs, measurement of central venous hemoglobin for skeletal muscle, and 88% for skin, again highlighting
saturation (ScvO2 ) has been evaluated as a potential sur- the differences in metabolic demand and blood flow rate

2 
C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749
 Venous SO2 monitoring

nutrient blood flow (increasing the SO2 in the inferior

[caudal] vena cava). The kidney receives 25% of the car-
diac output but accounts for only 7–8% of total oxygen
consumption.28,29 Therefore, under normal physiologic
conditions the inferior (caudal) vena cava receives ve-
nous blood that contains more oxygen than the superior
(cranial) vena cava.29

Oxygen Extraction and Oxygen Extraction Ratio

When tissue needs increase, oxygen delivery is increased
by local regulation of arterial blood flow. In response to
enhanced regional blood flow CO increases to maintain
systemic blood pressure and minute ventilation is in-
creased to maintain the SO2 , pCO2 , and pH of arterial
blood. Simultaneously, increased oxygen extraction re-
moves more oxygen from delivered blood, allowing for
greater O2 consumption at a given blood flow rate. These
2 mechanisms allow for a large range of oxygen utiliza-
tion based on need, and whole-body oxygen consump-
tion ranges from an average of 3.5 mL/kg/min at rest
to a maximal consumption as high as 81.6 mL/kg/min
during heavy exercise in elite athletes.30 The physiologi-
cal reserve for individual muscle groups is even greater;
for example, in one experiment the range of oxygen con-
Figure 1: Representative values for oxygen saturation (SO2 ) in
sumption in the human leg ranged from 18 mL/min at
arterial (dotted lines) and venous (solid black lines) blood from
rest to 845 mL/min at maximal exercise (leg extensions
various organs and locations in the dog.
requiring mainly the quadriceps muscles), a nearly 47-
Table 1: Summary of venous SO2 of individual organs from fold increase made possible by increased oxygen extrac-
Figure 1 tion and a dramatic increase in supply that was not taxed
by demands of other muscles as occurs in whole-body
Organ Venous SO2 exercise.31
Kidney 88% In addition to providing a mechanism to support in-
Skin 84% creased muscle activity and changes in organ homeo-
Posterior vena cava 80% static metabolic needs, increased oxygen extraction is an
Anterior vena cava 75%
essential mechanism of compensation in disease states
Jugular vein 72%
Muscle 67%
complicated by exhaustion of cardiopulmonary reserve.
Liver 63% Oxygen extraction can be increased globally by redis-
Gastrointestinal tract 55% tribution of blood flow based on regional metabolic
Heart 25% demand and at a tissue level by increasing functional
capillary density.32,33 These responses may be character-
ized by observing changes in the oxygen extraction ratio
between different tissues.18 Similar data are available for (O2 ER), expressed as V̇O2 divided by DO2 during a de-
dogs (Figure 1, Table 1).19–27 fined period of time (O2 ER = V̇O2 /DO2 ).18
Because mixed venous blood represents blood return- Conditions that increase O2 ER and reduce regional or
ing from all organs in the body it is the ideal sample to mixed venous oxygenation include increased demand
assess venous hemoglobin saturation for evaluation of (eg, exercise, seizures, inflammation, hyperthermia), re-
the overall oxygenation status of a patient.28 Typical val- duced supply (eg, relative or absolute hypovolemia,
ues for mixed venous oxygenation include an SO2 of 75% cardiac dysfunction, anemia, lung disease), or both. Sys-
and pO2 of 40 mm Hg. The SO2 of the superior (cranial) temic delivery of oxygen is often reduced during de-
vena cava is normally lower (70–75%) than the inferior velopment of shock and as DO2 falls, increases in O2 ER
(caudal) vena cava (75–80%) due to a combination of a occur through local feedback mechanisms.14 Conversely,
high rate of cerebral oxygen extraction (reducing the SO2 venous oxygenation may increase when tissue require-
in the superior [cranial] vena cava) and high renal non- ment for oxygen is reduced or when cellular capacity to


C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749 3
R.A.L. Walton & B.D. Hansen

quantify the percentage of hemoglobin that is present as

oxyhemoglobin.40 The absorption of light across much
of the spectrum between 650 and 1000 nm is different
for oxyhemoglobin and deoxyhemoglobin, and oxime-
ters use this differential absorption at 2 or more spe-
cific wavelengths to calculate the functional hemoglobin
saturation.41 The two most important methods of mea-
suring venous oxygenation in humans are continuous
in vivo measurement with an indwelling fiber optic
catheter and ex vivo analysis of a blood sample via
co-oximetry.42 The gold standard method of analysis
is co-oximetry performed on an anaerobically collected
blood sample, where relative concentrations of oxy-
Figure 2: Principles of critical oxygen delivery and the critical hemoglobin, deoxyhemoglobin (as well as carboxyhe-
extraction ratio, based on whole-body oxygen and delivery and moglobin, methemoglobin, and sulfhemoglobin) are de-
consumption in an anesthetized dog subjected to stepwise re- termined from the absorption of light at 128 distinct
moval of blood. Each data point represents V̇O2 and DO2 as mea- wavelengths.41 The fraction of total hemoglobin that is
sured by analysis of cardiac output, expired gas analysis, CaO2 , oxygenated and the fraction of hemoglobin capable of
and CvO2 using the Fick equation 10–20 minutes after removal of being oxygenated that is oxygenated are expressed as
up to 100 mL of blood. Linear trend lines are superimposed on the percentages.43
6 data points during relatively well-maintained consumption (A)
Continuous in vivo monitors utilize 2 or 3 wave-
and on the 3 data points demonstrating nearly a 1:1 relationship
lengths of light provided by narrow-bandwidth LED
between the fall in DO2 and the fall in V̇O2 (B, region of supply
dependence). Region A is characterized by a progressive increase
light sources; this light is transmitted into the blood-
in ERO2 , which maintains DO2 while SvO2 falls. Once extraction stream via an optical fiber and reflected back to a second
is maximal (critical ERO2 ratio of 0.79, upright arrow), consump- fiber connected to a photodiode (Figure 3). The device
tion becomes dependent on delivery, and as delivery falls below measures the relative absorption of the reflected light
5.4 mL/kg/min (downward arrow), consumption falls in direct at 2 wavelengths and the venous oxygen saturation is
proportion to the fall in delivery. Modifed from Nelson et al.14 reported as SO2 . The measurement is collected continu-
ously and the percent saturation is presented as a run-
utilize oxygen is compromised, as during mitochondrial ning average from data acquired in the preceding few
dysfunction in sepsis.34,35 seconds.42
The ability to increase the O2 ER to compensate for
falling DO2 has a biological limit. The extraction ra-
tio at the point at which a falling DO2 fails to meet Mixed vs. Central Venous Oximetry
tissue V̇O2 needs is called the critical extraction ratio Although it may be the ideal venous blood to evalu-
(Figure 2).14,35,36 The critical extraction ratio varies with ate, a major disadvantage of SvO2 measurement is the
species and is affected by disease states; it also varies be- need to use a PAC to access to pulmonary arterial blood.
tween tissues based on capillary density, tissue energy In contrast to SvO2 monitoring via a PAC, ScvO2 mon-
stores and sensitivity to metabolic feedback.35 Organ- itoring uses a central venous catheter situated in the
specific critical O2 ERs have been calculated with data superior (cranial) vena cava to measure oxygen satu-
from dogs in studies conducted to evaluate compen- ration of venous blood from returning from only the up-
satory mechanisms invoked during shock. When oxygen per (cranial) body. Central venous catheterization is less
delivery falls below these critical values, a phenomenon expensive and is technically simpler than pulmonary
known as supply-dependent oxygen consumption oc- arterial catheterization, with less potential for compli-
curs and persistent hypoxia and hypoperfusion leads to cations. Although there are few reports of the relation-
the development of a significant tissue oxygen debt.37,38 ship between the 2 measurements in health, it appears
Development of this debt contributes to the development that under normal physiologic conditions ScvO2 pre-
of multi-organ dysfunction syndrome (MODS).39 dicts SvO2 reasonably well and is usually 2–5% less,
largely due to aforementioned effects of high cerebral
oxygen uptake and high non-nutrient renal blood flow
Venous Oximetry
(Figure 1).29,42,44,45 The relationship between SvO2 and
Measurement of hemoglobin oxygen saturation is typi- ScvO2 becomes more tenuous during shock states when
cally performed with optical oximetry, a technique that the circulation is redistributed to the upper (cranial)
uses the absorption of specific wavelengths of light to body at the expense of splanchnic and renal circulation.46

4 
C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749
 Venous SO2 monitoring

Figure 3: The Q2 Plus continuous saturation monitor (ICU Medical Inc, San Clemente, CA, USA). (A) User interface. (B) End-view of
the optical module that connects to the optical fiber; the infrared light is the bright dot visible within. (C) The optical module connected
to a optical fiber probe (Opticath central venous oximetry probe, ICU Medical Inc) inserted into a central venous catheter in a canine
patient.

Those conditions result in further desaturation of blood The often-large discrepancy between the 2 measure-
reaching the inferior (caudal) vena cava, reversing the re- ments in critically ill people has prompted some investi-
lationship found in health, and ScvO2 may exceed SvO2 gation into ways to improve their correlation; one report
by up to 20%.44,47 Nevertheless, in one study of dogs with suggests to advance the tip of the monitoring catheter
experimental hemorrhage ScvO2 correlated well with just inside the right atrium to allow sufficient mixing of
SvO2 , remaining 3–6% higher throughout hemorrhage blood from the cranial and caudal vena cava to bring
and resuscitation.47 Although a single absolute value for the ScvO2 measured there closer to the SvO2 .50,51 Be-
ScvO2 does not predict SvO2 very well in sick human cause placement of a catheter tip further into the right
patients, the 2 values tend to change in parallel with atrium is associated with right atrial thrombus forma-
clinically acceptable correlation, and changes in either tion or rupture and pericardial tamponade in people and
usually reflect clinically significant change in cardiopul- the authors have observed both of these complications in
monary status. dogs, the risk should be minimized by using radiographs
Because the relationship between the 2 measure- to confirm that the catheter is not advanced excessively
ments varies with underlying pathology, the suitabil- far into the chamber.50,51
ity of ScvO2 as a surrogate measure for SvO2 has been
widely debated. In particular, there may be significant
and unpredictable discrepancy between the 2 measure-
ments during septic shock.48 In contrast to states of hy- Interpretation of Venous Oxygen Saturation
povolemic shock, septic shock is characterized by mul- The 4 principle determinants of venous oxygen satura-
tiple pathophysiologic states resulting in alterations of tion values are arterial oxygen saturation, hemoglobin
oxygen-carrying capacity and mitochondrial oxygen uti- concentration, cardiac output, and tissue oxygen con-
lization. Although the correlation between the two mea- sumption (Figure 4).44 In healthy (or at least clini-
surements in studies of septic patients has been statis- cally stable) resting people and animals the SvO2 and
tically significant, wide confidence intervals around the ScvO2 are consistently in the range of 70–75% and 65–
relationship suggest that clinically important errors will 70%, respectively.52,53 In the presence of disease ve-
be made by assuming ScvO2 accurately predicts SvO2 in nous oxygen saturation values may become abnormally
individual patients.48,49 low or high. Low values, representing increased oxygen


C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749 5
R.A.L. Walton & B.D. Hansen

Figure 4: Potential causes of abnormal central or mixed venous oxygen saturation.

extraction, occur when oxygen delivery is insufficient by decreasing oxygen consumption via pathological im-
to meet metabolic demand. Pathological impairment of pairment of oxygen extraction from blood, as occurs with
oxygen delivery can be due to lung disease with im- mitochondrial and microcirculatory dysfunction in sep-
paired oxygenation of blood (resulting in reduced arte- sis. As reviewed by Singer, sepsis-induced mitochondrial
rial hemoglobin saturation), low hemoglobin concentra- dysfunction directly inhibits mitochondrial respiration,
tion (resulting in low arterial oxygen carrying capacity), reducing tissue capacity to utilize oxygen and contribut-
impaired circulation (due to decreased effective circu- ing to MODS.55 Microcirculatory alterations include
lating volume and other causes of reduced cardiac out- decreased functional capillary density, increased
put), or any combination of these. Therefore, animals microvascular permeability, and increased perfusion
with unexplained low ScvO2 or SvO2 values should be heterogeneity, which contribute to the defect in oxygen
evaluated for the presence of pulmonary disease, ane- extraction observed in septic patients.56,57 These micro-
mia and states of reduced cardiac output, including circulatory abnormalities result in altered vascular au-
hypovolemia, cardiac disease or arrhythmias. Because toregulatory mechanisms such as impaired vasomotor
ScvO2 or SvO2 may also be reduced by any cause of high tone, resulting in functional shunting of oxygen.58,59
tissue oxygen consumption relative to delivery, condi- Mitochondrial failure causes cells to be unable to uti-
tions such as shivering or muscle tremors should be lize the oxygen presented to them.60 As tissue oxy-
controlled before assuming that low saturations are due gen delivery decreases and alterations in tissue uti-
solely to pathological imbalance. lization of oxygen occur, a large imbalance between
Sepsis may be associated with tissue hypoxia and oxygen delivery and utilization develops resulting in
dysfunction secondary to macrocirculatory, microcir- tissue hypoxia. Under physiologic conditions mitochon-
culatory, and mitochondrial failure. Macrocirculatory drial metabolism accounts for approximately 98% of to-
failure is characterized by failure of oxygen delivery tal body oxygen consumption.55 In sepsis the mitochon-
secondary to decreased intravascular volume, cardiac dria can be affected in at least 2 ways: impaired perfusion
output, and pulmonary function; the absolute and rela- results in low mitochondria pO2 , and direct inhibition of
tive hypovolemia develops secondary to extracorporeal mitochondrial respiration secondary to the generation
fluid losses, tissue edema, and systemic vasodilation. of reactive oxygen species.55,60 This impaired respiration
These changes in macrocirculation may be assessed via despite normal local pO2 , called cytopathic hypoxia, rep-
monitoring parameters that include central venous pres- resents the inability of cells to utilize oxygen and results
sure, mean arterial pressure, cardiac index, and ScvO2 , in high venous SO2 .60,61 Mitochondrial failure ultimately
and will tend to cause reductions in any or all of these.54 results in insufficient energy production and the initia-
In contrast, venous oxygen saturation may be increased tion of apoptosis.

6 
C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749
 Venous SO2 monitoring

Clinical Application of ScvO2 in Human Medicine hemodilution. Thus, ScvO2 provides additional informa-
tion regarding oxygen delivery and demand following
Because central venous oxygen saturation is compara-
resuscitation that is not reflected by hemodynamic mon-
tively easy to obtain and provides insight into systemic
itoring alone.64
oxygen delivery and consumption, its value as a moni-
tored parameter has been evaluated for several clinical
syndromes.
Sepsis
Sepsis is a leading cause of death in people in the United
States with contemporary mortality rates reaching up to
Trauma 50%.65 ScvO2 has been used as a prognostic indicator
Hemorrhage is the leading cause of morbidity and mor- and therapeutic target in people with sepsis, and stud-
tality in human trauma patients.62 Severe hemorrhage ies have identified an association between low ScvO2 (<
impairs oxygen delivery by reducing venous return and 65%) and increased illness severity, organ failure, and
cardiac output, and hemorrhage control and reestablish- mortality.66,67 Additionally, high ScvO2 values have also
ment of normal vital signs are key steps in resuscitation. been associated with increased mortality in patients with
Identification of shock states and its response to therapy sepsis.61 The practice of using normal ScvO2 as a thera-
have traditionally been achieved by evaluation of physi- peutic goal in the initial resuscitation of septic shock was
cal exam findings and heart rate. Heart rate is one of the widely adopted following the publication of the results
primary determinants of cardiac output, and tachycardia of this approach by Rivers et al in 2001.13 In this con-
in the face of hypovolemia may be essential to maintain text, early goal-directed therapy (EGDT) involves ma-
oxygen delivery and consumption. Typically, the heart nipulation of blood oxygen content and cardiac preload,
rate stabilizes with control of hemorrhage and restora- afterload, and contractility to balance oxygen delivery
tion of effective circulating volume. However, normal- with oxygen demand and reach a goal of a ScvO2 ࣙ
ization of heart rate and other physical signs of perfusion 70%.13 Compared to their standard-care patients, goal-
are not always associated with reversal of the shock state, directed therapy in the Rivers study reduced mortal-
especially in more severely injured patients, and in some ity from 46.5% to 30.5%. Similarly, in another clinical
of them occult shock persists despite normalization of trial of severe sepsis or fluid-refractory septic shock chil-
these parameters.63 When occult shock persists the pa- dren and adolescents were assigned to groups defined by
tient is at an increasing risk of complications including whether or not treatment was based on ScvO2 monitor-
MODS. ing and goal directed resuscitation.68 The ScvO2 (EGDT)
Central venous hemoglobin saturation with oxygen group had a substantially lower mortality rate (11.8%
has been used as an early indicator of patient circulatory versus 39.2% in the control group) and a lower incidence
status during hemorrhage and is a reliable and accurate of organ dysfunction, consistent with the notion that
method of detecting blood loss in trauma patients. As EGDT using an endpoint of ScvO2 ࣙ70% has a signifi-
oxygen delivery falls, tissue O2 ER increases and ScvO2 cant positive impact on outcome.68 Both of those studies
falls. When evaluated in human trauma patients, 39% evaluated the effects of goal-directed therapy during the
of patients had saturation values less than 65% despite first hours after diagnosis, a situation potentially much
normalized and stable vital signs including heart rate different from development of sepsis as a complication
and pulse pressure.63 These patients, with saturation in hospitalized patients. Furthermore, more recent at-
values less than 65%, had more serious injuries, larger tempts to replicate the findings of the Rivers study have
estimated blood losses and required more blood trans- failed to demonstrate benefit. The PROCESS trial eval-
fusions than those with saturation values greater than uated protocolized goal-directed therapy for manage-
65%. This clinical finding in people has been at least par- ment of septic shock and found no difference in all-cause
tially explained with findings in animal models of shock. mortality compared with usual resuscitation in patients
In an anesthetized pig model of hemorrhagic shock with septic shock, but did demonstrate higher costs of
ScvO2 correlated well with changes in stroke volume and care associated with the technique.69 Other recent trials
the venous-arterial pCO2 gap.64 Whereas the pCO2 gap demonstrated that EGDT drove significant differences in
normalized with restoration of stroke volume by crys- treatment with vasopressors, inotropes, and blood trans-
talloid resuscitation, the observed reduction of ScvO2 fusions but did not reduce mortality.70,71 Although these
persisted, prompting the investigators to conclude that studies call into question the practice and the use of
ScvO2 may not be strictly a function of hemodynamic EGDT as a treatment strategy, low ScvO2 remains as-
performance. When viewed as a function of oxygen sup- sociated with higher mortality in patients with septic
ply and demand, a persistently low ScvO2 can alert shock and therefore should continue to be evaluated as
the clinician to look for other causes such as excessive a potential prognostic tool.71


C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749 7
R.A.L. Walton & B.D. Hansen

Cardiogenic shock reduction in ScvO2 during CPR has prognostic signifi-

Cardiogenic shock is characterized by inadequate DO2 cance; in one study people whose ScvO2 failed to reach
resulting in tissue hypoxia and organ failure, and is as- 30% never achieved return of spontaneous circulation
sociated with mortality rates in humans ranging from 50 whereas all patients with a ScvO2 value greater than 72%
to 80%.72 It is a consequence of circulatory failure due to did.79
cardiomyopathies, ventricular outflow obstruction (aor-
tic stenosis), valvular heart disease, and/or arrhythmias.
ScvO2 in Veterinary Medicine
As cardiac pump failure reduces DO2 , tissues compen-
sate by increasing the O2 ER, resulting in a proportional There is a growing body of evidence that physical signs
reduction in venous SO2 . Central venous hemoglobin and blood pressure may not predict recovery of opti-
saturation with oxygen was found to be an independent mal oxygenation following initial resuscitation in some
predictor of treatment failure in a small study of people people and experimental models. It seems plausible that
with acute decompensated heart failure who required in- veterinary patients are subject to the same phenomenon.
otropic support, in addition to standard care, including A normal ScvO2 range has never been established in
diuretic therapy, where an ROC-derived cutoff of <60% veterinary patients; however, the cutoff of 70% has been
predicted treatment failure in 81% in patients receiving suggested based on the experimental work and experi-
dobutamine.73 Measurement of both blood lactate con- ence in human medicine. Using that cutoff in one study,
centration and continuous ScvO2 have also been used to 11/30 critically ill dogs were suspected to have occult
stratify and treat patients with acutely decompensated shock despite normalization of heart rate, blood pres-
end-stage chronic heart failure. This combination was sure, mentation, and perfusion parameters, suggesting
superior to routine monitoring in detecting cardiogenic persistent severe illness despite clinical findings suggest-
shock in a subset of patients whose vital signs suggested ing otherwise.80 Dogs that had a ScvO2 >70% were also
they were relatively well compensated. The mean ScvO2 noted to have improved normalization of lactate, sug-
in patients with elevated blood lactate concentration was gesting better reversal of shock in those patients.
32% despite vital signs that were similar to a control More recently, some attempts have been made to di-
group of clinically stable heart failure patients presented rect resuscitation with global indicators of perfusion in-
for outpatient procedures with a mean ScvO2 of 60%.74 cluding lactate, base excess, and ScvO2 in addition to as-
These findings suggest a potential role for ScvO2 moni- sessment of physical perfusion parameters. In dogs with
toring of patients with myocardial dysfunction to detect sepsis and septic shock secondary to pyometra, ScvO2
occult shock. By allowing for earlier intervention and and base deficit were the best predictors of survival.81
therapy targeted at improving DO2 and ScvO2 , investi- In that study, there were no significant changes in ScvO2
gators have speculated that using ScvO2 to guide acute over time, despite therapeutic interventions, however
care of patients with cardiogenic shock will lead to im- values were significantly different between groups with
proved outcomes. survivors demonstrating higher values than nonsur-
vivors. One possible explanation for failure of treatment
to increase ScvO2 is the limited repertoire of adjunctive
Cardiac arrest monitoring (eg, no invasive blood pressure monitoring
Cardiac arrest rapidly produces a global oxygen debt was performed) and circulatory treatments (consisting
associated with increased plasma lactate concentration of saline, colloid, and dopamine titrated to maintain
and development of a metabolic acidosis.75 Factors as- blood pressure). A second recent study demonstrated
sociated with improved outcome from resuscitation in- that low ScvO2 values appeared to be associated with an
clude early recognition of patients at risk, and earlier increased probability of death in critically ill dogs.82 A
and technically optimized resuscitation efforts.76 These total of 126 dogs were enrolled into this observational
factors reduce the severity and duration of cellular hy- study after admission to the intensive care unit follow-
poxia, therefore a primary goal of cardiopulmonary re- ing initial evaluation (and in some cases, resuscitation)
suscitation (CPR) is rapid restoration of adequate sys- in the emergency department or routine appointment;
temic oxygen delivery. Because it may be implemented many of these dogs were referred after initial treatment
quickly and offers nearly real-time assessment, contin- at primary care facilities. Central venous hemoglobin sat-
uous ScvO2 monitoring may be used to assess effec- uration with oxygen values ranged from 59.3 to 84.7%
tiveness of CPR and therapy during the critical hours in survivors and 44.7 to 82.3% in nonsurvivors, and a
following initial resuscitation. During CPR, cardiac out- ROC-derived cutoff of 68% predicted increased risk of
put is often less than 30% of normal and very low poor outcome. The results of both of these studies sug-
ScvO2 values reflect significantly increased oxygen ex- gest that dogs with ScvO2 values that are low after initial
traction by peripheral tissues.77,78 The severity of that resuscitation and remain low despite ongoing care have

8 
C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749
 Venous SO2 monitoring

a higher mortality risk than dogs with ScvO2 values that 10. Akmal AH, Hasan M, Mariam A. The incidence of complications of
central venous catheters at an intensive care unit. Ann Thorac Med
normalize. 2007; 2(2):61–63.
These recent veterinary study findings suggest the 11. The SVO2 collaborative group. The SvO2 study: general design and
utility of ScvO2 as a prognostic indicator and perhaps results of the feasibility phase of a multicenter, randomized trial
of three different hemodynamic approaches and two monitoring
also as a therapeutic target in canine shock patients. techniques in the treatment of critically ill patients. The SvO2 Col-
Hemorrhagic shock, cardiopulmonary arrest, and sep- laborative Group. Control Clin Trials 1995; 16(1):74–87.
sis are common presentations in emergency and critical 12. Rivers EP, Ander DS, Powell D. Central venous oxygen saturation
monitoring in the critically ill patient. Curr Opin Crit Care 2001;
care medicine and improved monitoring through utiliza- 7(3):204–211.
tion of ScvO2 and more focused therapeutic targets may 13. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in
improve treatment and outcome for significant numbers the treatment of severe sepsis and septic shock. N Engl J Med 2001;
345(19):1368–1377.
of animals. Whether such monitoring offers a substan- 14. Nelson DP, King CE, Dodd SL, et al. Systemic and intestinal lim-
tial improvement in outcome compared to more con- its of O2 extraction in the dog. J Appl Physiol (1985) 1987; 63(1):
ventional methods applied by a competent critical care 387–394.
15. Shimizu S, Enoki Y, Kohzuki H, et al. Determination of Hufner’s fac-
specialist remains to be determined with larger studies. tor and inactive hemoglobins in human, canine, and murine blood.
Jpn J Physiol 1986; 36(5):1047–1051.
16. Curtis SE, Cain SM. Regional and systemic oxygen delivery/uptake
relations and lactate flux in hyperdynamic, endotoxin-treated dogs.
Conclusions Am Rev Respir Dis 1992; 145(2 Pt 1):348–354.
17. Marx G, Reinhart K. Venous oximetry. Curr Opin Crit Care 2006;
Venous oxygen saturation can provide valuable infor- 12(3):263–268.
mation regarding oxygen delivery and utilization in a 18. Gutierrez JA, Theodorou AA. Oxygen delivery and oxygen con-
variety of shock states and critical illness. Venous oxime- sumption in pediatric critical care. In: Lucking SE, Maffei FA, Tam-
burro RF, Thomas NJ. eds. Pediatric Critical Care Study Guide.
try has been utilized in several settings of human critical London: Springer-Verlag; 2012, pp. 19–38.
illness, and while its utility for monitoring people with 19. Haskins S, Pascoe PJ, Ilkiw JE, et al. Reference cardiopulmonary
these disorders is unlikely to completely translate into values in normal dogs. Comp Med 2005; 55(2):156–161.
20. Grieb P, Forster RE, Strome D, et al. O2 exchange between blood
use in veterinary medicine, it likely does have value as and brain tissues studied with 18O2 indicator-dilution technique. J
both a monitoring tool and therapeutic target for crit- Appl Physiol (1985) 1985; 58(6):1929–1941.
ically ill animals. Future veterinary studies of venous 21. von RW, Holtz J, Bassenge E. Exercise induced augmentation of
myocardial oxygen extraction in spite of normal coronary dilatory
oxygenation should include evaluation of the accuracy of capacity in dogs. Pflugers Arch 1977; 372(2):181–185.
continuous in vivo monitoring, the sensitivity and speci- 22. Mills PC. A model to investigate hepatic extraction of oxygen during
ficity of abnormal ScvO2 to identify clinically important anaesthesia in the dog. Res Vet Sci 2003; 75(3):179–183.
23. Schlichtig R, Kramer DJ, Pinsky MR. Flow redistribution during
problems compared with less invasive monitoring, and progressive hemorrhage is a determinant of critical O2 delivery. J
its suitability as a treatment endpoint for resuscitation. Appl Physiol 1991; 70(1):169–178.
24. Fujita Y. Effects of PEEP on splanchnic hemodynamics
and blood volume. Acta Anaesthesiol Scand 1993; 37(4):427–
431.
References 25. Samsel RW, Schumacker PT. Systemic hemorrhage augments local
O2 extraction in canine intestine. J Appl Physiol 1994; 77(5):2291–
1. Wacker DA, Winters ME. Shock. Emerg Med Clin North Am 2014; 2298.
32(4):747–758. 26. Schlichtig R, Kramer DJ, Boston JR, et al. Renal O2 consumption
2. Laszlo G. Respiratory measurements of cardiac output: from elegant during progressive hemorrhage. J Appl Physiol 1991; 70(5):1957–
idea to useful test. J Appl Physiol (1985) 2004; 96(2):428–437. 1962.
3. Swan HJ, Ganz W, Forrester J, et al. Catheterization of the heart in 27. Ilkiw JE, Rose RJ, Martin IC. A comparison of simultaneously col-
man with use of a flow-directed balloon-tipped catheter. N Engl J lected arterial, mixed venous, jugular venous and cephalic venous
Med 1970; 283(9):447–451. blood samples in the assessment of blood-gas and acid-base status
4. Bizouarn P, Blanloeil Y, Pinaud M. Comparison between oxygen in the dog. J Vet Intern Med 1991; 5(5):294–298.
consumption calculated by Fick’s principle using a continuous ther- 28. Lee J, Wright F, Barber R, et al. Central venous oxygen saturation in
modilution technique and measured by indirect calorimetry. Br J shock: a study in man. Anesthesiology 1972; 36(5):472–478.
Anaesth 1995; 75(6):719–723. 29. Barratt-Boyes BG, Wood E. The oxygen saturation of blood in the ve-
5. Shoemaker WC, Appel PL, Kram HB, et al. Prospective trial of supra- nae cavae, right-heart chambers, and pulmonary vessels of healthy
normal values of survivors as therapeutic goals in high-risk surgical subjects. J Lab Clin Med 1957; 50(1):93–106.
patients. Chest 1988; 94:1176–1186. 30. Bassett DR, Jr., Howley ET. Limiting factors for maximum oxygen
6. Molnar KL. SvO2 monitoring: limitations and considerations. Can uptake and determinants of endurance performance. Med Sci Sports
Crit Care Nurs J 1991; 8(1):4–7. Exerc 2000; 32(1):70–84.
7. Harvey S, Harrison DA, Singer M, et al. Assessment of the clinical 31. Blomstrand E, Radegran G, Saltin B. Maximum rate of oxygen
effectiveness of pulmonary artery catheters in management of pa- uptake by human skeletal muscle in relation to maximal activ-
tients in intensive care (PAC-Man): a randomised controlled trial. ities of enzymes in the Krebs cycle. J Physiol 1997; 501 (Pt 2):
Lancet 2005; 366(9484):472–477. 455–460.
8. Harvey SE, Welch CA, Harrison DA, et al. Post hoc insights from 32. Tuman KJ. Tissue oxygen delivery: the physiology of anemia. Anes
PAC-Man—the U.K. pulmonary artery catheter trial. Crit Care Med Clin N Amer 1990;8:451–469.
2008; 36(6):1714–1721. 33. Messmer K. Blood rheology factors and capillary blood flow.
9. Kelso LA. Complications associated with pulmonary artery In: Gutierrez G, Vincent JL. eds. Update in Intensive Care and
catheterization. New Horiz 1997; 5(3):259–263.


C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749 9
R.A.L. Walton & B.D. Hansen

Emergency Medicine 12: Tissue Oxygen Utilization. 12 ed. Berlin: 56. De BD, Donadello K, Sakr Y, et al. Microcirculatory alterations in
Springer-Verlag; 1991, pp. 103–113. patients with severe sepsis: impact of time of assessment and rela-
34. van BP, Wietasch G, Scheeren T, et al. Clinical review: use of venous tionship with outcome. Crit Care Med 2013; 41(3):791–799.
oxygen saturations as a goal—a yet unfinished puzzle. Crit Care 57. Ellis CG, Bateman RM, Sharpe MD, et al. Effect of a maldistribution
2011; 15(5):232. of microvascular blood flow on capillary O(2) extraction in sepsis.
35. Granger HJ, Nyhof RA. Dynamics of intestinal oxygenation: inter- Am J Physiol Heart Circ Physiol 2002; 282(1):H156–H164.
actions between oxygen supply and uptake. Am J Physiol 1982; 58. Vallet B. Vascular reactivity and tissue oxygenation. Intensive Care
243(2):G91–G96. Med 1998; 24(1):3–11.
36. Schlichtig R, Kramer DJ, Pinsky MR. Flow redistribution during 59. Arnold RC, Shapiro NI, Jones AE, et al. Multicenter study of early
progressive hemorrhage is a determinant of critical O2 delivery. J lactate clearance as a determinant of survival in patients with pre-
Appl Physiol 1991; 70:169–178. sumed sepsis. Shock 2009;32(1):35–39.
37. Cain SM, Curtis SE. Experimental models of pathologic oxygen sup- 60. Fink M. Cytopathic hypoxia in sepsis. Acta Anaesthesiol Scand
ply dependency. Crit Care Med 1991; 19(5):603–612. Suppl 1997; 110:87–95.
38. Bruegger D, Kemming GI, Jacob M, et al. Causes of metabolic acido- 61. Pope JV, Jones AE, Gaieski DF, et al. Multicenter study of central
sis in canine hemorrhagic shock: role of unmeasured ions. Crit Care venous oxygen saturation (ScvO(2)) as a predictor of mortality in
2007; 11(6):R130. patients with sepsis. Ann Emerg Med 2010; 55(1):40–46.
39. Shoemaker WC, Appel PL, Kram HB. Role of oxygen debt in the 62. Kauvar DS, Lefering R, Wade CE. Impact of hemorrhage on trauma
development of organ failure sepsis, and death in high-risk surgical outcome: an overview of epidemiology, clinical presentations, and
patients. Chest 1992; 102(1):208–215. therapeutic considerations. J Trauma 2006; 60(6 Suppl):S3–11.
40. Haymond S. Oxygen saturation. Clinical Laboratory News 2006:10– 63. Scalea TM, Hartnett RW, Duncan AO, et al. Central venous oxygen
12. saturation: a useful clinical tool in trauma patients. J Trauma 1990;
41. Haymond S, Cariappa R, Eby CS, et al. Laboratory assessment of 30(12):1539–1543.
oxygenation in methemoglobinemia. Clin Chem 2005; 51(2):434– 64. Nemeth M, Tanczos K, Demeter G, et al. Central venous oxygen
444. saturation and carbon dioxide gap as resuscitation targets in a hem-
42. Iodice FG, Ricci Z, Haiberger R, et al. Fiberoptic monitoring of cen- orrhagic shock. Acta Anaesthesiol Scand 2014; 58(5):611–619.
tral venous oxygen saturation (PediaSat) in small children undergo- 65. Mayr FB, Yende S, Angus DC. Epidemiology of severe sepsis. Viru-
ing cardiac surgery: continuous is not continuous. F1000Res 2014; lence 2014; 5(1):4–11.
3:23. 66. Krafft P, Steltzer H, Hiesmayr M, et al. Mixed venous oxygen satu-
43. Breuer HW, Groeben H, Breuer J, et al. Oxygen saturation calcu- ration in critically ill septic shock patients: the role of defined events.
lation procedures: a critical analysis of six equations for the de- Chest 1993; 103:900–906.
termination of oxygen saturation. Intensive Care Med 1989; 15(6): 67. Varpula M, Tallgren M, Saukkonen K, et al. Hemodynamic vari-
385–389. ables related to outcome in septic shock. Intensive Care Med 2005;
44. Shepherd SJ, Pearse RM. Role of central and mixed venous oxygen 31(8):1066–1071.
saturation measurement in perioperative care. Anesthesiology 2009; 68. de Oliveira CF, de Oliveira DS, Gottschald AF, et al. ACCM/PALS
111(3):649–656. haemodynamic support guidelines for paediatric septic shock: an
45. Weber H, Grimm T, Albert J. [The oxygen saturation of blood in outcomes comparison with and without monitoring central venous
the venae cavae, right-heart chambers, and pulmonary artery, com- oxygen saturation. Intensive Care Med 2008; 34(6):1065–1075.
parison of formulae to estimate mixed venous blood in healthy 69. Yealy DM, Kellum JA, Huang DT, et al. A randomized trial of
infants and children (author’s transl)]. Z Kardiol 1980; 69(7): protocol-based care for early septic shock. N Engl J Med 2014;
504–507. 370(18):1683–1693.
46. Dahn MS, Lange MP, Jacobs LA. Central mixed and splanchnic 70. Peake SL, Delaney A, Bailey M, et al. Goal-directed resuscitation for
venous oxygen saturation monitoring. Intensive Care Med 1988; patients with early septic shock. N Engl J Med 2014; 371(16):1496–
14(4):373–378. 1506.
47. Reinhart K, Rudolph T, Bredle DL, et al. Comparison of central- 71. Shin TG, Jo IJ, Hwang SY, et al. Comprehensive interpretation of
venous to mixed venous oxygen saturation during changes in oxy- central venous oxygen saturation and blood lactate levels during
gen supply/demand. Chest 1989; 95:1216–1221. resuscitation of patients with severe sepsis and septic shock in the
48. Varpula M, Karlsson S, Ruokonen E, et al. Mixed venous oxy- emergency department. Shock 2016; 45(1):4–9.
gen saturation cannot be estimated by central venous oxy- 72. Topalian S, Ginsberg F, Parrillo JE. Cardiogenic shock. Crit Care
gen saturation in septic shock. Intensive Care Med 2006; 32(9): Med 2008; 36(1 Suppl):S66–S74.
1336–1343. 73. Gallet R, Lellouche N, Mitchell-Heggs L, et al. Prognosis value of
49. Turnaoglu S, Tugrul M, Camci E, et al. Clinical applicability of the central venous oxygen saturation in acute decompensated heart
substitution of mixed venous oxygen saturation with central ve- failure. Arch Cardiovasc Dis 2012; 105(1):5–12.
nous oxygen saturation. J Cardiothorac Vasc Anesth 2001; 15(5): 74. Ander DS, Jaggi M, Rivers E, et al. Undetected cardiogenic shock in
574–579. patients with congestive heart failure presenting to the emergency
50. Cavaliere F, Zamparelli R, Martinelli L, et al. Blood from the right department. Am J Cardiol 1998; 82(7):888–891.
atrium may provide closer estimates of mixed venous saturation 75. Rivers EP, Rady MY, Martin GB, et al. Venous hyperoxia after cardiac
than blood from the superior vena cava. A pilot study. Minerva arrest. Characterization of a defect in systemic oxygen utilization.
Anestesiol 2014; 80(1):11–18. Chest 1992; 102(6):1787–1793.
51. Kopterides P, Bonovas S, Mavrou I, et al. Venous oxygen saturation 76. Sandroni C, Ferro G, Santangelo S, et al. In-hospital cardiac arrest:
and lactate gradient from superior vena cava to pulmonary artery survival depends mainly on the effectiveness of the emergency re-
in patients with septic shock. Shock 2009; 31(6):561–567. sponse. Resuscitation 2004; 62(3):291–297.
52. Walley KR. Use of central venous oxygen saturation to guide ther- 77. Jackson RE, Freeman SB. Hemodynamics of cardiac massage. Emerg
apy. Am J Respir Crit Care Med 2011; 184(5):514–520. Med Clin North Am 1983; 1(3):501–513.
53. Reinhart K, Kuhn HJ, Hartog C, et al. Continuous central venous 78. Nakazawa K, Hikawa Y, Saitoh Y, et al. Usefulness of central ve-
and pulmonary artery oxygen saturation monitoring in the critically nous oxygen saturation monitoring during cardiopulmonary resus-
ill. Intensive Care Med 2004; 30(8):1572–1578. citation. A comparative case study with end-tidal carbon dioxide
54. Joosten A, Alexander B, Cannesson M. Defining goals of resus- monitoring. Intensive Care Med 1994; 20(6):450–451.
citation in the critically ill patient. Crit Care Clin 2015; 31(1): 79. Rivers EP, Martin GB, Smithline H, et al. The clinical implications of
113–132. continuous central venous oxygen saturation during human CPR.
55. Singer M. The role of mitochondrial dysfunction in sepsis-induced Ann Emerg Med 1992; 21:1094–1101.
multi-organ failure. Virulence 2014; 5(1):66–72.

10 
C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749
 Venous SO2 monitoring

80. Young BC, Prittie JE, Fox P, et al. Decreased central venous oxygen shock in response to goal-directed hemodynamic optimization at
saturation despite normalization of heart rate and blood pressure admission to ICU and the relation to outcome. J Vet Emerg Crit Care
post shock resuscitation in sick dogs. J Vet Emerg Crit Care 2014; 2012; 22(4):409–418.
24(2):154–161. 82. Hayes GM, Mathews K, Boston S, et al. Low central venous oxygen
81. Conti-Patara A, de Araujo CJ, de Mattos-Junior E, et al. Changes saturation is associated with increased mortality in critically ill dogs.
in tissue perfusion parameters in dogs with severe sepsis/septic J Small Anim Pract 2011; 52(8):433–440.


C Veterinary Emergency and Critical Care Society 2018, doi: 10.1111/vec.12749 11