Pathophysiology of Oxygen Delivery in

Respiratory Failure*
Mitchell M. Levy, MD, FCCP

Complex physiologic interactions exist between oxygenation, hemoglobin, and cardiac output
(Qt) in critically ill patients with respiratory failure. When any or all of these three critical factors
fail, clinicians are challenged to support oxygen delivery (DO2) in order to avoid tissue hypoxia,
end-organ damage, and high mortality rates. Many of the interventions performed to improve
DO2, including mechanical ventilation, blood transfusions, fluid management, and invasive
monitoring of cardiac function, are accompanied by serious risks that can exacerbate the
pathology of DO2. This article provides an overview of oxygenation, hemoglobin, and Qt in
patients with respiratory failure and highlights some of the current research that seeks safe and
effective ways to improve DO2 in these patients. (CHEST 2005; 128:547S–553S)

Key words: organ damage; oxygen delivery; oxygenation; respiratory failure; tissue hypoxia

Abbreviations: Cao2 ⫽ arterial oxygen content; CVP ⫽ central venous pressure; Do2 ⫽ oxygen delivery;
Fio2 ⫽ fraction of inspired oxygen; NO ⫽ nitric oxide; PEEP ⫽ positive end-expiratory pressure; Pio2 ⫽ inspired Po2;
Qt ⫽ cardiac output; Sao2 ⫽ arterial oxygen saturation; Svo2 ⫽ venous oxygen saturation; V̇o2 ⫽ oxygen consumption;
V̇/Q̇ ⫽ ventilation/perfusion

Learning Objectives: 1. To review important considerations for managing patients in respiratory failure, ie,
oxygenation, hemoglobin, and cardiac output. 2. To discuss several strategies for each aspect of care, with attention to
controversies and/or new trends in treatment of respiratory failure.

R espiratory failure is a condition that leads to

inadequate oxygen delivery (Do ) and requires 2
immediate clinical intervention to avoid tissue hyp-
oxia and subsequent organ damage. Global Do2 is
the product of cardiac output (Qt) and arterial
oxygen content (Cao2). Cao2 is, in turn, the product
of arterial oxygen saturation (Sao2), hemoglobin
concentration, and a constant, reflecting the hemo-
globin-oxygen binding capacity.1 A concise review of
these basic equations appears elsewhere in this
Supplement (see Huang YCT; Monitoring Oxygen
Delivery in the Critically Ill).
Despite years of study, the relationship between
Do2 and oxygen consumption (V̇o2) is still incom-
pletely understood.2,3 The theory of pathologic oxy-
gen supply dependency in critically ill patients—
*From the Division of Pulmonary and Critical Care Medicine,
Rhode Island Hospital Brown University School of Medicine,
Providence, RI.
Dr. Levy has disclosed financial relationships with a commercial
party. Grant information and company names appear as provided
by the author. Grant monies (from industry-related sources):
Involved in EPO II trial, as principal investigator at Rhode Island
This publication was supported by an educational grant from
Ortho Biotech Products, L.P.
Reproduction of this article is prohibited without written permission
from the American College of Chest Physicians (www.chestjournal.
org/misc/reprints.shtml).
Correspondence to: Mitchell M. Levy, MD, Division of Pulmo-
nary and Critical Care Medicine, Rhode Island Hospital Brown
University School of Medicine, 593 Eddy St, Main 7, Providence,
RI 02903; e-mail: mitchell_levy@brown.edu
when tissue oxygen extraction falls short of demand
over a wide range of Do2—is now believed to be an
artifact of mathematical coupling, which occurs be-
cause Do2 and V̇o2 are calculated with equations
that share common variables.3,4 The critical Do2
threshold, known as the point at which Do2 fails to
meet the metabolic demand and anaerobic metabo-
lism begins, is currently a more clinically relevant
and widely accepted concept. In humans, healthy or
otherwise, the critical Do2 value remains unknown,5
and determination of “adequate” tissue oxygenation
has long been an elusive goal for clinicians.6 Al-
though identification of oxygen utilization at the
tissue level would help clinicians decide how much
oxygen is enough, the options are limited for assess-
ing changes in global or individual organ oxygen
requirements (usually as a change in V̇o2) and tissue
perfusion.7 Of the relationships that are known,
three major, interdependent components must be
monitored and supported in the critically ill patient
with respiratory failure in order to preserve Do2:
oxygenation, hemoglobin-related parameters, and Qt.
Determinants of Oxygen Exchange
Respiration is a process that involves the exchange
of oxygen and carbon dioxide between a living
organism and its environment. Figure 1 contains a
schematic diagram of normal gas transport in a

www.chestjournal.org CHEST / 128 / 5 / NOVEMBER, 2005 SUPPLEMENT 547S

 Figure 1. Gas transport in a healthy 70-kg adult breathing room air (Fio2 of 0.21) at standard
barometric pressure (101 kPa).1 Values in parentheses are for the normal 70-kg adult. The oxygen
partial pressures of dry, humidified, and alveolar air are assumed to be 21.3 kPa, 20 kPa, and 14.7 kPa,
respectively. Hb ⫽ hemoglobin; Cvo2 ⫽ mixed venous oxygen content; Peco2 ⫽ mixed expired Pco2;
Peo2 ⫽ mixed inspired Po2; Pio2 ⫽ inspired Po2; P50 ⫽ Po2 at which 50% of the hemoglobin is
saturated. Used with permission from Treacher et al.1

healthy adult, demonstrating the interplay of lung
alveoli, heart, blood, capillaries, and intracellular
mitochondria in the respiratory process.1 Following
oxygen uptake in the lung via inspiration and then
passive oxygen diffusion into arterial blood, Do2 next
depends on the oxygen-carrying capacity of blood, ie,
its hemoglobin content and dissociation kinetics.
Global and regional Do2 to tissues also relies on
adequate Qt and local control of blood flow to end
organs. Oxygen diffuses from capillaries to cells,
where it is utilized at the tissue level. Simultaneous
exchange of oxygen and carbon dioxide is followed by
removal of carbon dioxide from the blood into the
alveoli, where this waste gas is ultimately exhaled,
completing the normal respiratory cycle. Respiratory
failure and inadequate Do2 can result from malfunc-
tion of any aspect of the “ventilatory apparatus.” A
decrease in convective oxygen transport and in-
creased oxygen extraction by tissues can lead to
progressive decreases in venous oxygen saturation
(Svo2), rapid arterial desaturation, and an insuffi-
cient oxygen supply to the tissues.8
Oxygenation
Under normal conditions, alveolar oxygen pres-
sure drives the diffusion of oxygen into arterial
blood, measured clinically as Pao2. The transfer of
inspired oxygen and removal of waste carbon dioxide
will be limited if there is alveolar damage or injury to
respiratory muscles, carotid bodies, or the central
respiratory center. Acute lung injury and ARDS
commonly lead to hypoxemic respiratory failure.9
Arterial hypoxemia may be initiated by alveolar
hypoventilation (respiratory depression, respiratory
muscle weakness, or obstructive airway disease) but
also can result from development of a diffusion
barrier such as occurs in pulmonary edema or from
ventilation-perfusion (V̇/Q̇) mismatch.1 This results
either from inadequate ventilation of perfused alve-
oli or reduced perfusion of well-ventilated alveoli,10
and is the most frequent contributor to clinically
important oxygen desaturation.11
Pao2 and Sao2 are the main determinants of
arterial hypoxemia; in addition, clinical assessment,
pH, lactate, Do2/V̇o2, changes in Pco2, gastric mu-
cosal pH, and Svo2 have all been used to monitor
tissue oxygen status.12 There are, however, examples
in which these measures are inadequate. Although
frequently used to monitor oxygen exchange, Pao2
may not provide sufficient information about the
adequacy of Do2. A below-normal Pao2 generally
indicates V̇/Q̇ mismatch, but a normal Pao2 does not
necessarily mean that there is V̇/Q̇ homogeneity in
the lung. The alveolar Po2/Pao2 gradient reflects the
efficiency of oxygen uptake from the alveoli to
blood11 and may be a more sensitive indicator of V̇/Q̇
abnormalities. When severe V̇/Q̇ mismatch is sus-
pected as the cause of hypoxemia, the Pao2/fraction
of inspired oxygen (Fio2) ratio is a good index of
oxygenation that is easily calculated.13 Values of Pao2
and Sao2 can also be normal in a critically ill patient
who is anemic or has low Qt, and thus these param-
eters will fail to detect the presence of tissue hypoxia.
In these situations, mixed Svo2, when very low, may
be a better indicator of tissue oxygenation than Pao2
or Sao2.6 The measurement of Svo2 requires the
presence of a pulmonary arterial catheter, which is
not without risk.14,15

548S Improving Outcomes in Respiratory Failure: Ventilation, Blood Use, and Anemia Management
 Strategies for Improving Oxygenation
Oxygen may be supplemented directly with inva-
sive or noninvasive ventilatory support. While at-
tempting to improve oxygenation by mechanical
means, the potential for oxygen toxicity should be
considered. In the critically ill patient, excess Fio2
has the potential to decrease vital capacity and
increase alveolar-capillary permeability.16 The least
amount of Fio2 to achieve effective oxygenation
should be used, typically ⬍ 0.60.17 If V̇/Q̇ mismatch
is present, there will be a positive response of Pao2
to administered oxygen, but if there is a “true
shunt”—if the amount of mixed venous blood that
completely bypasses the pulmonary capillaries ex-
ceeds 30%—increasing the Fio2 will not improve
the Pao2.1
Ventilatory management is shifting from a strategy
that prioritized oxygen exchange and Do2 optimiza-
tion to a new strategy that attempts to ensure that
the lung is protected.17 The goals of the old vs new
strategies are not mutually exclusive, since functional
lung density and oxygen exchange efficiency are
correlated.18 Mainstream lung protective approaches
to mechanical ventilation include the use of lower
tidal volumes and optimized pressure settings for
positive end-expiratory pressure (PEEP). Noninva-
sive ventilation may be appropriate in some patients.
Less conventional but potentially promising strate-
gies involve prone positioning, permissive hypercap-
nia, and high-frequency ventilation that allows lower
tidal volumes while maintaining minute ventilation.
Reduced Tidal Volumes and PEEP
Conclusive evidence that a lung protective strategy
improves mortality was provided by the ARDS Net-
work.19 In a study of 861 patients with ARDS,
ventilation with low tidal volumes (6 mL/kg) com-
bined with an airway plateau pressure of ⱕ 30 cm
H2O reduced the relative risk of mortality by 22%
compared to traditional ventilation volume (12 mL/
kg) with ⱕ 50 cm H2O. The number of ventilator-
free days and percentage of patients breathing with-
out assistance by day 28 were also significantly
higher in the group with low tidal volumes
(p ⫽ 0.007 and p ⬍ 0.001, respectively). Largely be-
cause of the ARDS Network study and another
investigation by Amato et al,20 which coupled higher
PEEP settings with low tidal volumes and achieved a
47% reduction in mortality, there is general agree-
ment that low tidal volume ventilation should be the
standard of care against which other interventions
and supportive techniques are measured.21
The primary rationale for using PEEP in patients
with ARDS is to prevent end-expiratory alveolar
collapse and the hypoxemia that results from this
www.chestjournal.org
increased shunt in patients with ARDS. This occurs
for several reasons, including a qualitative and quan-
titative surfactant defect as well as the increased
weight on dependent lung regions due to increased
lung density.18 There is also evidence that repetitive
recruitment and derecruitment during tidal ventila-
tion may result in shear stress injury, which may
contribute to ventilator-associated lung injury. The
exact method for determining “optimal PEEP” in
ARDS remains uncertain.
Other Lung Protective Strategies for Improving
Oxygenation
Inhaled nitric oxide (NO) has the potential to
improve V̇/Q̇ matching by dilating the pulmonary
vasculature and has been suggested as a means to
augment arterial and tissue oxygenation in patients
receiving mechanical ventilation, with an added ben-
efit of reduced inflammatory mediators and platelet
aggregation.9 Although improved oxygenation with
NO has been observed, outcomes in large multi-
center studies have not improved,22 except in a small
subgroup of patients who inhaled low doses (5 ppm)
of NO.23
Prone positioning in patients with ARDS has been
reported to improve oxygenation in 70 to 80% of
patients.24 The mechanisms involved relate to in-
creased V̇/Q̇ matching due to increased lung volume,
redistribution of perfusion, and dorsal lung recruit-
ment. The positive effects of prone positioning on
oxygenation diminish after approximately 1 week of
mechanical ventilation and are incompletely under-
stood.24 In one study25 of patients with pulmonary
aspiration, oxygenation was significantly improved in
the patients who were subjected to prone position-
ing. The authors24 concluded that early prone posi-
tioning favorably altered V̇/Q̇ relationships, aided
drainage of secretions, opened up alveoli, and pre-
vented progression of pneumonitis. In a large, pro-
spective, randomized, controlled study of prone po-
sitioning conducted by Gattinoni and colleagues,26
survival was unaffected despite improved oxygen-
ation. However, a survival benefit was suggested in
the subset of patients with the lowest Pao2/Fio2
ratio.
Largely investigated in pediatrics, permissive hy-
percapnia combined with high ventilation rates is
another potentially beneficial lung protective strat-
egy in patients receiving ventilation.27 The oxygen-
hemoglobin dissociation curve shifts to the right
when Paco2 is elevated and favors peripheral tissue
oxygen unloading.28 Hypocapnic alkalosis alters the
balance between global Do2 and V̇o2, decreasing
supply and increasing demand for oxygen.29 In con-
trast, at the cellular level, lower Paco2 increases the
CHEST / 128 / 5 / NOVEMBER, 2005 SUPPLEMENT 549S
metabolic requirement for oxygen through increases
in cell excitation or contraction, and contributes to
cell death, the pathogenesis of acute lung injury,
systemic inflammation, and the risk of permanent
brain damage.28,29 Data reporting a lower incidence
of barotrauma during mechanical ventilation for
near-fatal asthma30 and the ability of permissive
hypercapnia to ameliorate stretch-induced lung in-
jury in neonates31 have contributed to a growing
acceptance of permissive hypercapnia in the ventila-
tion strategy of adult patients with ARDS.32
Hemoglobin
The second component of Do2 apt to malfunction
in the patient with respiratory failure is the oxygen-
carrying capacity of blood. The concentration of
circulating hemoglobin is the primary determinant of
Cao2;33 therefore, all things being equal, an ade-
quate hemoglobin concentration should delay the
onset of anaerobic metabolism. For numerous rea-
sons, including phlebotomy, hemorrhage secondary
to trauma and/or surgery, inflammation, nutritional
deficiencies, and decreased erythropoietin produc-
tion,34 –36 anemia is prevalent in critically ill patients.
During the course of an ICU stay, hemoglobin may
fall from a baseline average of 11 g/dL to 8 to 10
g/dL.37
Optimal hemoglobin levels in critically ill patients
and how they should be achieved has been a topic of
interest and debate since the landmark work of
Freudenberger and Carson38 and Hebert et al.39
One factor that complicates prediction of “optimal
hemoglobin” is the dissociation profile of oxygen
from hemoglobin, which is likely to be altered and/or
changing throughout the course of critical illness.
The kinetics of oxygen binding to hemoglobin are
affected by temperature, pH, and inorganic phos-
phates in RBCs (eg, 2,3-diphosphoglycerate).6 Aci-
dosis and increased temperature decrease the hemo-
globin affinity for oxygen and tend to improve tissue
oxygenation; however, the effects of hypophos-
phatemia on 2,3-diphosphoglycerate are opposing
and tend to shift the hemoglobin dissociation curve
back to the left.6
Strategies for Improving Hemoglobin
Parameters
Theoretically, administration of RBC transfusions
should supply hemoglobin and improve Do2 in
critically ill patients, although studies40 – 43 have
failed to show improvement in tissue oxygenation or
outcomes following RBC transfusions. Raising he-
550S Improving Outcomes in Respiratory Failure: Ventilation, Blood Use, and Anemia Management
moglobin concentration is a key factor for adequate
tissue oxygenation, but the bulk of evidence seems to
indicate that the Do2, V̇o2, or survival benefits of
using RBC transfusions as a vehicle to deliver hemo-
globin do not outweigh their risks. Perhaps the most
frequently cited evidence questioning the effective-
ness of RBC transfusions was provided by Hebert et
al,39 who demonstrated, in a well-designed, random-
ized clinical trial, that a liberal transfusion policy
conferred no survival benefit over a restrictive trans-
fusion policy, with the exception of a subset of
patients with advanced cardiac disease. Spahn44 has
recently acknowledged that there is no “magic num-
ber of hemoglobin” at which transfusions should be
administered. The College of American Pathologists
recommends RBC transfusions based on an oxygen
extraction rate ⬎ 50%, Pvo2 ⬍ 25 mm Hg (⬍ 3.3
kPa), and a reduction in V̇o2 to ⬍ 50% of baseline.45
An important consideration with regard to the
oxygen-delivering capability of RBC transfusions is
that allogeneic blood typically administered to ICU
patients has been stored an average of 21 days.37
Older stored blood lacks 2,3-diphosphoglycerate,
which increases the hemoglobin-oxygen binding af-
finity, as noted above, and impairs its ability to
unload oxygen.6 The RBC membranes of older blood
may also be less deformable and lead to microthrom-
boses that could exacerbate tissue hypoxia.46 Despite
these clinical and physiologic observations that ought
to discourage the liberal transfusion of RBCs, a
recent study of critically ill patients by Corwin et al37
(of whom 32% had a primary diagnosis of respiratory
failure) showed that transfusions were still very
commonly administered.
Transfusion practices were recently evaluated in a
retrospective study47 of patients receiving mechani-
cal ventilation from the study by Corwin et al.37 In
that study,37 it was observed that patients receiving
mechanical ventilation tended to receive transfusions
more often than patients not receiving mechanical
ventilation (49% vs 33%, p ⬍ 0.0001) [mean ⫾ SD],
and patients receiving mechanical ventilation had
significantly higher pretransfusion hemoglobin levels
(8.7 ⫾ 1.7 g/dL) than patients not receiving mechan-
ical ventilation (8.2 ⫾ 1.7 g/dL, p ⬍ 0.0001). These
data suggest an a priori assumption on the part of
intensivists that maintaining hemoglobin at a higher
level in patients receiving mechanical ventilation is
beneficial. Among these patients, however, transfu-
sions were not associated with better outcomes.
A few reports48 –50 have supported the notion that
successful attempts to wean from mechanical venti-
lation are associated with higher hemoglobin levels.
However, a post hoc analysis of the mechanical
ventilation subset from the population investigated
by Hebert et al51 revealed no difference between
liberal and restrictive RBC transfusion groups with
respect to the duration of mechanical ventilation or
number of ventilator-free days. Given the present
evidence, transfusing patients receiving mechanical
ventilation with RBCs aggressively and at a higher
hemoglobin threshold than in patients not receiving
mechanical ventilation does not appear to be justi-
fied.
Administration of epoetin alfa (40,000 U/wk) sig-
nificantly increased hemoglobin levels and reduced
transfusion requirements in a randomized study of
1,302 critically ill patients, although no significant
effects on ventilator outcomes, hospital or ICU
length of stay, or 28-day mortality were detected.52
Further studies are necessary to investigate whether
raising hemoglobin with epoetin alfa can improve
Do2 and outcomes in the specific population of
patients receiving mechanical ventilation.
Qt
Beyond optimizing oxygenation and hemoglobin,
achieving adequate Do2 in the patient with respira-
tory failure also requires that Qt be maintained. For
the heart to function properly, a constant replenish-
ment of oxygen is needed, since the coronary circu-
lation has limited oxygen reserve and extracts 60 to
75% of what is delivered.38 Cardiac dysfunction in
the ICU population is one of the primary predictors
of mortality,53 and in studies37,54 of the critically ill,
approximately 40% have significant cardiac disease
as a comorbidity. Cardiac dysfunction may result
from underlying organic heart disease; insufficient
Do2 to the coronary circulation, which can be pre-
cipitated/exacerbated by anemia, subendocardial
ischemia from left ventricular hypertrophy, compro-
mised myocardial contractility from the effects of
inflammatory cytokines, inappropriate intravascular
fluid status, or a combination of factors.38,55
recently published results from a large, randomized
study in which mortality and other outcomes were
similar in patients administered 4% albumin or 0.9%
saline solution, both overall and in the subgroup of
patients with acute respiratory failure.
Ventricular end-diastolic pressure, measured indi-
rectly as pulmonary artery occlusion pressure with a
pulmonary artery catheter, can provide information
about progressive increases in Qt resulting from fluid
challenges. However, the benefit of the pulmonary
artery catheter is somewhat controversial. Because
permeability of the pulmonary capillary endothelium
may be increased due to inflammation, the pressure
corresponding to edema formation may be more
accurately determined by bedside pulmonary capil-
lary pressure measurements than by pulmonary ar-
tery occlusion pressure.56 Measurement of arterial
pulse pressure or arterial pulse contour analysis has
also been suggested as a means to predict which
patients will respond favorably to a fluid challenge by
increasing Qt.59
The inotropic agents dopamine and dobutamine
(primarily ␤1-agonists) are frequently employed in
patients with respiratory failure when the response
to fluid challenge is inadequate.60 Rivers et al61
examined the effects of early goal-directed therapy
in patients with severe sepsis and shock, combining
sequential protocol-driven (Fig 2) administration of a

Strategies for Improving Qt

During resuscitation maneuvers in shock patients,
fluids, vasopressors, vasodilators, and inotropes are
administered, often in combination, to augment Qt
and overall Do2.56 Patients in respiratory failure
supported with mechanical ventilation are frequently
hemodynamically unstable, with low BP, tachycar-
dia, and decreased Qt and urinary output.57 Clinical
decisions must take into account the benefits vs risks
of the type (colloid vs crystalloid) and volume of fluid
chosen to increase circulating blood volume and
Figure 2. Protocol for early goal-directed therapy for patients in
venous return. With regard to fluid choice, the Saline shock.61 MAP ⫽ mean arterial pressure; Scvo2 ⫽ central venous
vs Albumin Fluid Evaluation study investigators58 oxygen saturation; Used with permission from Rivers et al.61

www.chestjournal.org CHEST / 128 / 5 / NOVEMBER, 2005 SUPPLEMENT 551S

crystalloid fluid bolus, vasopressors, and dobutamine
with continuous monitoring of central venous pres-
sure (CVP) and central Svo2 to achieve hemody-
namic goals, which included a CVP of 8 to 12 mm
Hg, mean arterial pressure of 65 to 90 mm Hg, and
Svo2 of 70%. Patients randomly assigned to receive
early goal-directed therapy for the first 6 h had
significantly better outcomes (in-hospital mortality,
organ failure, physiologic indicators of Do2) than
those who received standard care during this time.61
Although this study61 was conducted in septic shock
patients and not specifically in patients with respira-
tory failure, the marked reduction in mortality
achieved with this aggressive approach is noteworthy
and may also be relevant in the setting of respiratory
failure.
Critical Do2 Thresholds and Do2
Requirements
In critically ill patients with and without sepsis in
whom independent measurements of Do2 and V̇o2
were made, Ronco and colleagues62 reported that
the critical Do2 was approximately 4 mL of oxygen
per kilogram per minute, noting that this was much
lower than any values previously derived using non-
independent variables for anesthetized or critically ill
patients. Although initially encouraging, prospective
studies63 that have evaluated increasing Do2 have
failed to consistently demonstrate benefit. In one
study by Gattinoni and colleagues,63 achievement of
supranormal hemodynamic values and patterns of
increased oxygen transport in medical patients with
ARDS produced no difference in mortality when
compared to patients in whom normal Do2 was
achieved. Several meta-analyses have confirmed this
disappointing finding. The important findings of
Rivers et al61 in septic patients suggest that the
failure of these early trials may have been due to the
timing of therapy. Most of the early studies did not
increase Do2 until ⬎ 24 h after hospital admission.
Conclusions
Further study is needed to determine the optimal
oxygenation requirements in respiratory failure, as
well as to improve our understanding of the pathol-
ogies underlying oxygen deficits. More efficient and
safer ways to deliver oxygen to tissues through
manipulations of Qt and hemoglobin parameters are
also needed. Although a source of hemoglobin, RBC
transfusions may not be effective in improving oxy-
genation or outcomes in critically ill patients with
respiratory failure. Further studies are needed to
552S Improving Outcomes in Respiratory Failure: Ventilation, Blood Use, and Anemia Management
evaluate the appropriate levels of Do2 during respi-
ratory failure and the best methods for restoring
adequate tissue perfusion.
References
1 Treacher DF, Leach RM. Oxygen transport: 1. Basic princi-
ples. BMJ 1998; 317:1302–1306
2 Dorinsky PM, Costello JL, Gadek JE. Relationships of oxygen
uptake and oxygen delivery in respiratory failure not due to
the adult respiratory distress syndrome. Chest 1988; 93:1013–
1019
3 Raghuram J, Eng P, Ong YY. The oxygen delivery debate: a
review. Ann Acad Med Singapore 1998; 27:404 – 408
4 Hameed SM, Aird WC, Cohn SM. Oxygen delivery. Crit Care
Med 2003; 31:S658 –S667
5 Lieberman JA, Weiskopf RB, Kelley SD, et al. Critical oxygen
delivery in conscious humans is less than 7.3 ml O2 x kg(-1)
x min(-1). Anesthesiology 2000; 92:407– 413
6 Leach RM, Treacher DF. Oxygen transport: 2. Tissue hyp-
oxia. BMJ 1998; 317:1370 –1373
7 Boyd O. Optimisation of oxygenation and tissue perfusion in
surgical patients. Intensive Crit Care Nurs 2003; 19:171–181
8 Jubran A, Mathru M, Dries D, et al. Continuous recordings of
mixed venous oxygen saturation during weaning from me-
chanical ventilation and the ramifications thereof. Am J
Respir Crit Care Med 1998; 158:1763–1769
9 Zimmerman JL. Respiratory failure. Blood Purif 2002; 20:
235–238
10 Appel PL, Shoemaker WC. Relationship of oxygen consump-
tion and oxygen delivery in surgical patients with ARDS.
Chest 1992; 102:906 –911
11 Greene KE, Peters JI. Pathophysiology of acute respiratory
failure. Clin Chest Med 1994; 15:1–12
12 American Thoracic Society. Tissue hypoxia: how to detect,
how to correct, how to prevent? J Crit Care 1997; 12:39 – 47
13 Huang Y-CT. Arterial blood gases. In: Hess DR, MacIntyre
NR, Mishoe SC, et al, eds. Respiratory care: principles and
practice. Philadelphia, PA: W.B. Saunders Company, 2002;
362–396
14 Valentine RJ, Duke ML, Inman MH, et al. Effectiveness of
pulmonary artery catheters in aortic surgery: a randomized
trial. J Vasc Surg 1998; 27:203–211
15 Sandham JD, Hull RD, Brant RF, et al. A randomized,
controlled trial of the use of pulmonary-artery catheters in
high-risk surgical patients. N Engl J Med 2003; 348:5–14
16 Capellier G, Maupoil V, Boussat S, et al. Oxygen toxicity and
tolerance. Minerva Anestesiol 1999; 65:388 –392
17 Marini JJ. Evolving concepts in the ventilatory management
of acute respiratory distress syndrome. Clin Chest Med 1996;
17:555–575
18 Gattinoni L, Pesenti A, Bombino M, et al. Relationships
between lung computed tomographic density, gas exchange,
and PEEP in acute respiratory failure. Anesthesiology 1988;
69:824 – 832
19 Acute Respiratory Distress Syndrome Network. Ventilation
with lower tidal volumes as compared with traditional tidal
volumes for acute lung injury and the acute respiratory
distress syndrome. N Engl J Med 2000; 342:1301–1308
20 Amato MB, Barbas CS, Medeiros DM, et al. Effect of a
protective-ventilation strategy on mortality in the acute respi-
ratory distress syndrome. N Engl J Med 1998; 338:347–354
21 Pinhu L, Whitehead T, Evans T, et al. Ventilator-associated
lung injury. Lancet 2003; 361:332–340
22 Lundin S, Mang H, Smithies M, et al. Inhalation of nitric
oxide in acute lung injury: results of a European multicentre
 study. The European Study Group of Inhaled Nitric Oxide.
Intensive Care Med 1999; 25:911–919
23 Dellinger RP, Zimmerman JL, Taylor RW, et al. Effects of
inhaled nitric oxide in patients with acute respiratory distress
syndrome: results of a randomized phase II trial. Inhaled
Nitric Oxide in ARDS Study Group. Crit Care Med 1998;
26:15–23
24 Pelosi P, Brazzi L, Gattinoni L. Prone position in acute
respiratory distress syndrome. Eur Respir J 2002; 20:1017–
1028
25 Easby J, Abraham BK, Bonner SM, et al. Prone ventilation
following witnessed pulmonary aspiration: the effect on oxy-
genation. Intensive Care Med 2003; 29:2303–2306
26 Gattinoni L, Tognoni G, Pesenti A, et al. Effect of prone
positioning on the survival of patients with acute respiratory
failure. N Engl J Med 2001; 345:568 –573
27 Laffey JG, O’Croinin D, McLoughlin P, et al. Permissive
hypercapnia: role in protective lung ventilatory strategies.
Intensive Care Med 2004; 30:347–356
28 Thome UH, Carlo WA. Permissive hypercapnia. Semin Neo-
natol 2002; 7:409 – 419
29 Laffey JG, Kavanagh BP. Hypocapnia. N Engl J Med 2002;
347:43–53
30 Dhuper S, Maggiore D, Chung V, et al. Profile of near-fatal
asthma in an inner-city hospital. Chest 2003; 124:1880 –1884
31 Varughese M, Patole S, Shama A, et al. Permissive hypercap-
nia in neonates: the case of the good, the bad, and the ugly.
Pediatr Pulmonol 2002; 33:56 – 64
32 Kavanagh B. Normocapnia vs hypercapnia. Minerva Anesthe-
siol 2002; 68:346 –350
33 Prittie JE. Triggers for use, optimal dosing, and problems
associated with red cell transfusions. Vet Clin North Am
Small Anim Pract 2003; 33:1261–1275
34 Rodriguez RM, Corwin HL, Gettinger A, et al. Nutritional
deficiencies and blunted erythropoietin response as causes of
the anemia of critical illness. J Crit Care 2001; 16:36 – 41
35 Drews RE. Critical issues in hematology: anemia, thrombo-
cytopenia, coagulopathy, and blood product transfusions in
critically ill patients. Clin Chest Med 2003; 24:607– 622
36 Vincent JL, Baron J-F, Reinhart K, et al. Anemia and blood
transfusion in critically ill patients. JAMA 2002; 288:1499 –
1507
37 Corwin HL, Gettinger A, Pearl RG, et al. The CRIT study:
anemia and blood transfusion in the critically ill; current
clinical practice in the United States. Crit Care Med 2004;
32:39 –52
38 Freudenberger RS, Carson JL. Is there an optimal hemoglo-
bin value in the cardiac intensive care unit? Curr Opin Crit
Care 2003; 9:356 –361
39 Hebert PC, Wells G, Blajchman MA, et al. A multicenter,
randomized, controlled clinical trial of transfusion require-
ments in critical care. Transfusion Requirements in Critical
Care Investigators, Canadian Critical Care Trials Group.
N Engl J Med 1999; 340:409 – 417
40 Kahn RC, Zaroulis C, Goetz W, et al. Hemodynamic oxygen
transport and 2,3-diphosphoglycerate changes after transfu-
sion of patients in acute respiratory failure. Intensive Care
Med 1986; 12:22–25
41 Shah DM, Gottlieb ME, Rahm RL, et al. Failure of red blood
cell transfusion to increase oxygen transport or mixed venous
Po2 in injured patients. J Trauma 1982; 22:741–746
42 Casutt M, Seifert B, Pasch T, et al. Factors influencing the
individual effects of blood transfusions on oxygen delivery and
oxygen consumption. Crit Care Med 1999; 27:2194 –2200
43 Dietrich KA, Conrad SA, Hebert CA, et al. Cardiovascular
www.chestjournal.org
and metabolic response to red blood cell transfusion in
critically ill volume-resuscitated nonsurgical patients. Crit
Care Med 1990; 18:940 –944
44 Spahn DR. Strategies for transfusion therapy. Best Pract Res
Clin Anaesthesiol 2004; 18:661– 673
45 Simon TL, Alverson DC, AuBuchon J, et al. Practice param-
eter for the use of red blood cell transfusions: developed by
the red blood cell administration practice guideline develop-
ment task force of the College of American Pathologists. Arch
Pathol Lab Med 1998; 122:130 –138
46 Chin-Yee I, Arya N, d’Almeida MS. The red cell storage
lesion and its implication for transfusion. Transfus Sci 1997;
18:447– 458
47 Levy MM, Abraham E, Zilberberg M, et al. A descriptive
evaluation of transfusion practices in mechanically ventilated
patients. Chest 2004 (in press)
48 Khamiees M, Raju P, DeGirolamo A, et al. Predictors of
extubation outcome in patients who have successfully com-
pleted a spontaneous breathing trial. Chest 2001; 120:1262–
1270
49 Rady MY, Ryan T. Perioperative predictors of extubation
failure and the effect on clinical outcome after cardiac
surgery. Crit Care Med 1999; 27:340 –347
50 Schonhofer B, Bohrer H, Kohler D. Blood transfusion facil-
itating difficult weaning from the ventilator. Anaesthesia
1998; 53:169 –191
51 Hebert PC, Blajchman MA, Cook DJ, et al. Do blood
transfusions improve outcomes related to mechanical venti-
lation? Chest 2001; 119:1850 –1857
52 Corwin HL, Gettinger A, Pearl RG, et al. Efficacy of
recombinant human erythropoietin in critically ill patients: a
randomized controlled trial. JAMA 2002; 288:2827–2835
53 Kollef MH, Sherman G. Acquired organ system derange-
ments and hospital mortality: are all organ systems created
equally? Am J Crit Care 1999; 8:180 –188
54 Hebert PC, Yetisir E, Martin C, et al. Is a low transfusion
threshold safe in critically ill patients with cardiovascular
diseases? Crit Care Med 2001; 29:227–234
55 Murray MJ. Sepsis: clinical dilemmas. Yale J Biol Med 1998;
71:485– 491
56 Levy MM. Pulmonary capillary pressure and tissue perfusion:
clinical implications during resuscitation from shock. New
Horiz 1996; 4:504 –518
57 Qvist J, Brynjolf I, Munck O. The effects of dopamine and
noradrenaline on cardiovascular function in patients with
acute respiratory failure. Eur J Anaesthesiol 1994; 11:107–110
58 Finfer S, Bellomo R, Boyce N, et al. A comparison of albumin
and saline for fluid resuscitation in the intensive care unit.
N Engl J Med 2004; 350:2247–2256
59 Pinsky MR. Rationale for cardiovascular monitoring. Curr
Opin Crit Care 2003; 9:222–224
60 Molloy DW, Ducas J, Dobson K, et al. Hemodynamic
management in clinical acute hypoxemic respiratory failure:
dopamine vs dobutamine. Chest 1986; 89:636 – 640
61 Rivers E, Nguyen B, Havstad S, et al. Early goal-directed
therapy in the treatment of severe sepsis and septic shock.
N Engl J Med 2001; 345:1368 –1377
62 Ronco JJ, Fenwick JC, Tweeddale MG, et al. Identification of
the critical oxygen delivery for anaerobic metabolism in
critically ill septic and nonseptic humans. JAMA 1993; 270:
1724 –1730
63 Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-oriented
hemodynamic therapy in critically ill patients. N Engl J Med
1995; 333:1025–1032
CHEST / 128 / 5 / NOVEMBER, 2005 SUPPLEMENT 553S