Evaluating hypoxemia in the critically ill

May 1, 2005
Alexander G. Duarte, MD
Alexander G. Duarte, MD

Akhil Bidani, MD, PhD
Akhil Bidani, MD, PhD

Volume 5, Issue 5

Abstract: Prompt correction of hypoxemia is a basic goal in the treatment of critically ill patients. Improvements in global oxygen delivery may be achieved by several means, such as providing an adequate fraction of inspired oxygen and using packed red blood cell transfusions for volume resuscitation. Low levels of positive end-expiratory pressure often help improve arterial oxygen tension. Measurement of mixed venous oxygen saturation (Sv?248-175?O2) can be useful in patient assessment. Sv?248-175?O2 may be decreased in patients with hypoxemia, hypovolemia, or anemia and may be elevated in patients with sepsis. Serum lactate levels may not quantitate the degree of tissue hypoxia, but serial measurements can help monitor the patient's response to therapy. For patients with septic or hypovolemic shock, early fluid resuscitation with isotonic crystalloid solution is essential. Catecholamine vasopressors can be useful when fluid administration fails to restore adequate blood pressure. (J Respir Dis. 2005;26(5):209-219)

Hypoxemia frequently occurs in critically ill patients as a result of a derangement in the efficiency of pulmonary oxygen uptake from ambient air. Alterations in pulmonary oxygen uptake are caused by ventilation-perfusion (V./Q. ) abnormalities and/or intrapulmonary shunts resulting in reduced arterial oxygen content, which can lead to tissue hypoxia. Consequently, oxygen delivery and the metabolic activity of various organs are altered.

Clinical conditions associated with hypoxemia and changes in oxygen delivery and consumption include acute respiratory distress syndrome (ARDS), sepsis, and trauma. Clinical worsening leads to progressive hypoxemia that may result in tissue hypoxia, which is implicated in the development of multiorgan failure.1 Thus, critically ill patients with sepsis or ARDS are not only in pulmonary peril but are also very likely to be at increased risk for systemic failure.

Regularly evaluating pulmonary oxygen uptake and delivery in critically ill patients provides information about oxygen transport and metabolic activity of peripheral tissue beds, and may allow early detection of tissue hypoxia.

In this article, we will review the dynamics of tissue oxygen balance. We will also describe various techniques for assessing oxygen delivery and will offer suggestions that may help you to better manage critically ill patients.


The terms "acute lung injury" and "ARDS" are used to describe hyp-oxemic respiratory failure in persons who are critically ill. The proposed mechanisms of abnormal oxygen uptake are alveolar edema and atelectasis, conditions that set the stage for shunting and V./Q. mismatching.2 In patients who have ARDS, hypoxemia is caused primarily by right-to-left intrapulmonary shunting of blood, leading to inefficient pulmonary oxygen uptake. Most patients consume a normal amount of oxygen, and tissue hypoxia rarely occurs unless their condition progresses to a terminal stage.

Other conditions that exacerbate oxygen imbalance and are often present in critically ill, mechanically ventilated patients include hypotension, fever, and anemia (see "Understanding oxygen balance"). Reduced blood flow or abnormal microvascular perfusion distribution is a prominent cause of impaired oxygen delivery and uptake. Febrile patients have increased oxygen demands that may further aggravate hypoxemia. In such an environment, the quantity of oxygen available for tissue delivery is diminished. The resulting tissue hypoxia constitutes a condition in which the cells within the tissue bed use more oxygen, which leads to anaerobic metabolism.3


Initial data using cardiac output and arterial and mixed venous blood oxygen contents suggested that whole body oxygen consumption was dependent on tissue oxygen delivery. This led to the practice of trying to achieve "supranormal" oxygen delivery with the use of vasopressors to increase cardiac output and deliver a greater amount of oxygen to peripheral tissues. This created controversy regarding the best approach to managing oxygen balance in the critically ill patient.

Supranormal delivery describes a strategy in which global oxygen delivery is increased to obtain oxygen delivery, oxygen consumption, and oxygen extraction values that can sustain life. It has been test- ed in clinical trials, with dubious results.

Early studies showed that infusing prostacyclin would increase oxygen delivery without increasing oxygen consumption, improving survival in some critically ill patients.4 This response to prostacyclin was believed to identify patients who had occult tissue hypoxia. The results of this investigation and other nonrandomized studies promulgated the use of vasopressors and inotropes as a means of achieving supranormal oxygen delivery in critically ill patients.5-7

Various investigators noted difficulties in achieving supranormal oxygen delivery.5-7 One problem is the failure to attain target values despite administration of increasing doses of catecholamines. In older or severely ill patients, achieving the target value is difficult. In addition, administration of large doses of dobutamine and norepinephrine to achieve a high cardiac output may lead to tachyarrhythmias, may worsen maldistribution of blood flow, and/or may produce myocardial ischemia.

More recent studies using independent measurements of oxygen consumption and delivery have not revealed a "pathologic" oxygen-supply dependence.8-11 Instead, there is a marked variability in basal whole body oxygen consumption related to the metabolic state of the patient.8-11

Several randomized controlled trials examined the survival benefit of increasing the patient's oxygen delivery to a target cardiac index of more than 4.5 mL/min/m2 and/or mixed venous oxygen saturation (Sv?248-175?O2) of 70% or more. Treatment of patients according to this strategy showed no benefit; to the contrary, mortality was increased.12-16 These studies also showed that using inotropes and vasopressors to increase oxygen to supranormal levels is difficult.14,15

Currently, the use of supranormal oxygen delivery in critically ill patients with existing hypoxia remains largely unjustified by clinical evidence.16 Improvements in global oxygen delivery may be achieved by several means. The first goal is to provide an adequate fraction of inspired oxygen to maintain oxygen saturation at more than 92% or PaO2 at more than 60 mm Hg.

In addition, oxygen-carrying capacity can be improved through the use of packed red blood cell transfusions for volume resuscitation. While the optimal hemoglobin concentration remains unknown, a target value of 10 to 12 g/dL would appear to be appropriate. However, data indicate that lower hemoglobin concentrations (7 to 10 g/dL) may be adequate and that routine transfusions of packed red blood cells to achieve a preselected value are not warranted.17

Alkalemia should be avoided because it impairs dissociation to the tissue. Simultaneously, mechanical ventilatory assistance should be provided using a tidal volume of 6 to 8 mL/kg of ideal body weight, and an attempt should be made to increase end-expiratory lung volume using positive end-expiratory pressure (PEEP) (5 cm H2O). Use of PEEP to increase mean alveolar pressure often succeeds in maintaining an adequate lung volume, thereby improving PaO2.

Low levels of PEEP reduce alveolar atelectasis in dependent portions of the lung and increase PaO2. However, at higher levels of PEEP, cardiac output may decline, leading to decreased oxygen delivery. Thus, optimal PEEP has been defined as the level that optimizes global oxygen delivery.18

PEEP-induced decline in cardiac output may be managed with volume infusion or packed red blood cell transfusion in anemic patients. Other measures involve decreasing oxygen consumption by giving antipyretics for fever and rigors. Minimizing agitation with the use of sedatives may also allow a patient's respiratory pattern to synchronize with the ventilator and avoid excessive muscle activity, thereby diminishing the work and oxygen cost of breathing.


Monitoring the patient for tissue hypoxia may allow early identification and implementation of specific therapies to reverse the process. Tissue hypoxia is assessed on a clinical basis, using biochemical indices, physiologic markers, and specific tissue probes.

Clinical assessment

Assessment of tissue hypoxia begins with a bedside evaluation. Although the physical findings of hypoxia are nonspecific, the presence of hypotension, tachycardia, tachypnea, mental obtundation, oliguria, cyanosis, pallor, and cool extremities provides strong supportive evidence and suggests organ dysfunction. In patients with sepsis, organ dysfunction may occur even in the absence of systemic hypoperfusion. In this case, systemic blood pressure measurements may not reflect specific organ perfusion and oxygen delivery.

The chief limits of bedside clinical assessment are the nonspecific and unquantifiable nature of these findings, as well as interobserver variability. Nonetheless, serial bedside examinations provide information that can be correlated with specific physiologic and biochemical measurements. One report highlighted the finding that the detection of cool extremities can identify patients with hypoperfusion as noted by elevated lactate levels and a relatively low cardiac output.19

Use of a pulmonary artery catheter allows bedside measurement of hemodynamics and Sv?248-175?O2. Measurements of central venous pressure, pulmonary artery pressure, pulmonary capillary wedge pressure, and cardiac output are routinely obtained. The pulmonary artery catheter is useful in determining intravascular volume status and guiding therapy for shock.

While pulmonary artery catheters are widely used in ICUs, there is concern regarding the lack of improvement in patient outcomes.20,21 An observational study involving more than 5700 critically ill medical and surgical patients raised issues concerning patient safety.22 Various reports have indicated that potential problems with the pulmonary artery catheter may be related to patient selection and data interpretation.23-25 Moreover, a prospective randomized trial reported no benefit to therapy directed by the pulmonary artery catheter compared with standard care of elderly, high-risk surgical patients.26

Mixed venous oxygenation

Sv?248-175?O2 can be used to assess the balance between systemic oxygen delivery and consumption. Continuous Sv?248-175?O2 monitoring was developed to provide the physician with immediate measurements that could be used as a basis for timely interventions.27

Intermittent blood sampling may be performed every 4 to 6 hours or at the discretion of the bedside clinician. Sv?248-175?O2 can be measured with the use of a specially equipped catheter (within a pulmonary artery catheter) that contains fiberoptic bundles that transmit and receive light to give a reading of venous blood oxygen content. This approach is widely available.

Blood can also be sampled from the distal port of the pulmonary artery catheter to ensure adequate mixing from the superior and inferior vena cava and, therefore, more accurate Sv?248-175?O2 measurements. Normally, Sv?248-175?O2 values obtained from the superior vena cava are lower than those obtained from the inferior vena cava; the reverse tends to occur in patients with shock.28,29

Sv?248-175?O2 is influenced by cardiac output, hemoglobin, oxygen consumption, and arterial oxygen saturation (SaO2). In hemodynamically stable patients, Sv?248-175?O2 ranges from 65% to 77%, with 75% generally considered to be normal.30 Keep in mind that a normal Sv?248-175?O2 value might not reflect inadequate oxygen delivery to specific organs.

Measurement of Sv?248-175?O2 using a pulmonary artery catheter provides a global assessment of Sv?248-175?O2 from different organs. However, an individual organ, such as the GI tract, may have a low Sv?248-175?O2 because of regional hypoperfusion or locally elevated oxygen consumption, while another vascular bed, such as muscle, may demonstrate an elevated Sv?248-175?O2, and the average of these values may provide a near-normal central Sv?248-175?O2.

Above-normal Sv?248-175?O2 values indicate conditions--sepsis, cirrhosis, and peripheral arteriovenous fistula, for example--in which oxygen delivery is increased relative to consumption. When measuring Sv?248-175?O2, keep in mind that problems with blood sampling, inaccurate calibration, or migration of a continuous oximetric catheter may lead to falsely elevated values.

Sv?248-175?O2 values of less than 65% reflect conditions, such as hypovolemia, cardiogenic shock, hypoxemia, and anemia, in which oxygen delivery is decreased relative to consumption. Diminished Sv?248-175?O2 values may also be observed under clinical circumstances, such as fever, shivering, seizures, and hyperthyroidism, in which oxygen demand exceeds oxygen delivery.

Examination of venous oxygen levels from different organs demonstrates a wide variation in local oxygen delivery and consumption. Thus, mixed Sv?248-175?O2 levels reflect a collection of end-capillary tissue oxygen levels from organs with different degrees of oxygen extraction, and may not detect hypoxia occurring within a specific organ. Without clinical correlation, a change in Sv?248-175?O2 suggests an imbalance in global oxygen delivery and consumption but does not necessarily identify the presence or cause of tissue hypoxia.

Concerns regarding Sv?248-175?O2 monitoring in patients with sepsis have been raised. Sudden changes in Sv?248-175?O2 that are unrelated to identifiable therapeutic interventions or physiologic changes have been reported. These changes have been attributed to fluctuations in tissue oxygen demand without a concomitant change in oxygen delivery.31

Furthermore, the accuracy of the measurement obtained using a continuous Sv?248-175?O2 oximetric catheter is affected by blood flow, hemoglobin concentration, blood coagulability, catheter position in the vessel lumen, and the plasma refractive index.32 Thus, formation of a thrombus at the catheter tip or malposition of the catheter within the vessel lumen may result in false changes in Sv?248-175?O2 values.

Serum lactate

Blood lactate levels reflect anaerobic metabolism and have been studied as a biochemical marker of tissue hypoxia.33 Most lactate originates from tissue beds that have a high rate of glycolysis: GI tract, muscle, brain, and red blood cells. Lactate is eliminated primarily through the liver, although the kidneys and myocardium also play a role. Total lactate production is 15 to 20 mEq/kg/d, and normal blood levels are 1 mEq/L (normal range, 0.7 to 1.3 mEq/L). A lactate level of more than 2 mEq/L suggests inadequate tissue perfusion.

Experimental and clinical studies of patients with sepsis have not convincingly demonstrated that elevated lactate levels result from tissue hypoxia alone.34-36 However, hyperlactemia can result from the effect of endotoxin on pyruvate dehydrogenase.Lactate production and elimination in critically ill patients is complex and probably represents altered cellular metabolism resulting from hypoxia and unrelated biochemical abnormalities.

Elevated blood lactate levels have been associated with poor outcome in critically ill patients, and changes in blood lactate levels appear to be a useful prognostic variable.37,38 Several reports have noted improved survival in critically ill patients who demonstrate a decline in elevated lactate levels.38-41 Although blood lactate levels may not quantitate the degree of tissue hypoxia, serial measurements can be useful for following the patient's response to therapy.


Regional venous oxygen levels depend on tissue perfusion and organ oxygen consumption. Regional assessment of oxygen delivery can be used to measure local tissue oxygen delivery. Ideally, assessment of oxygen delivery to a particular organ could be used to determine tissue hypoxia and to monitor the effects of therapy. Diminished tissue oxygenation to the splanchnic region, for example, may contribute to multiple organ dysfunction.42,43 Based on this evidence, noninvasive measures of regional perfusion have been proposed and developed for clinical practice.

The technique of gastric tonometry measures gastric mucosal pH. A gas-permeable balloon is placed onto the stomach mucosa. After allowing enough time for the PCO2 in the gut lumen to equilibrate with the balloon contents, a sample is removed and the PCO2 is analyzed. Subsequent measurement of intramucosal pH (pHi) permits evaluation of gastric carbon dioxide levels. Measurements of pHi are based on observations that the mucosal layer of the stomach is vulnerable to alterations in blood flow, decreased hemoglobin concentration, and increased splanchnic oxygen consumption and, therefore, may provide an early means of detecting tissue hypoxia.44

Gastric tonometry may be a prognostic indicator of survival in patients who are at risk for tissue hypoxia.45,46 However, the modality is expensive, operator-dependent, and cumbersome to use. Further clinical investigations and modifications of the current technique are required before it becomes more useful in daily patient care.


Careful clinical evaluation requires attention to changes in such physical findings as blood pressure, adequate urinary output, normal mentation, absence of cyanosis, presence of warm skin with good capillary refill, and adequate SaO2. Patients with fever should be managed with antipyretics (acetaminophen or ibuprofen) to diminish oxygen consumption.47 Excessive cooling, whether by a cooling blanket or fan, may cause the patient to shiver. This is counterproductive because it increases musculoskeletal activity and systemic oxygen consumption.

If the patient is agitated, check for evidence of barotrauma, myocardial ischemia, hypoxemia, ventilator-patient asynchrony, pain, or delirium; correct these problems as appropriate. Sedation of the delirious patient is useful because it decreases energy expenditure and excessive muscular activity.

Treatment of hypotension with early fluid resuscitation is crucial for patients with hypovolemia or septic shock. The promptness and adequacy of fluid resuscitation is more important than the selection of the fluid.

For resuscitation of hypovole-mic shock following a hemorrhage, initial volume replacement with isotonic crystalloid solution (0.9% saline or Ringer lactate) is appropriate; a packed red blood cell transfusion should be given as soon as typed blood is available. Ringer lactate solution should not be given in the setting of renal failure because of the risk of inducing hyperkalemia. Hypovolemic shock related to uncontrolled diabetes mellitus, diarrhea, or acute adrenal insufficiency should be treated with isotonic crystalloid solution.

In patients with septic shock, early fluid resuscitation with isotonic crystalloid solution is appropriate and should be followed by transfusion of blood products for correction of anemia or significant coagulopathy. The use of crystalloid or colloid solutions (dextran, albumin, hydroxyethyl starch) is controversial. A meta-analysis suggested increased mortality after colloid resuscitation, but the validity of this conclusion and the underlying mechanisms are unclear.48

A trial of early fluid resuscitation in a group of critically ill patients was shown to improve survival.49 As part of fluid resuscitation, packed red blood cell transfusions should be considered in critically ill patients with a hemoglobin concentration below 7 g/dL.

In critically ill patients who do not have acute blood loss, maintenance of the hemoglobin concentration between 7 and 9 g/dL does not appear to adversely affect outcome.17 However, patients with acute myocardial infarctions or unstable angina may require transfusion at hemoglobin values of less than 10 g/dL.50

Use of catecholamine vasopressors may be useful when fluid administration fails to restore adequate blood pressure.51 Although these agents may reduce blood flow to certain organs, their overall effect is aimed at restoring flow to hypoperfused organs. However, attempts to augment oxygen delivery or flow with the use of inotropes and/or vasopressors are discouraged, because this approach appears to increase mortality.



1. Bernard GR, Artigas A, Brigham KL, et al. The American-European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination.

Am J Respir Crit Care Med.

1994;149(3 pt 1):818-824.
2. Duarte AG, Bidani A. Monitoring patients with ARDS: pulmonary oxygen uptake.

J Crit Illness.

3. Third European Consensus Conference in Intensive Care Medicine. Tissue hypoxia: how to detect, how to correct, how to prevent. Societe de Reanimation de Langue Francaise, American Thoracic Society, European Society of Intensive Care Medicine.

Am J Respir Crit Care Med.

1996;154: 1573-1578.
4. Bihari D, Smithies M, Gimson A, Tinker J. The effects of vasodilation with prostacyclin on oxygen delivery and uptake in critically ill patients.

N Engl J Med.

5. Vincent JL, Roman A, De Backer D, Kahn RJ. Oxygen uptake/supply dependency. Effects of short-term dobutamine infusion.

Am Rev Respir Dis.

6. Shoemaker WC, Appel PL, Kram HB. Role of oxygen debt in the development of organ failure, sepsis, and death in high-risk surgical patients.


7. Astiz ME, Rackow EC, Falk JL, et al. Oxygen delivery and consumption in patients with hyperdynamic septic shock.

Crit Care Med.

1987;15: 26-28.
8. Ronco JJ, Fenwick JC, Tweeddale MG, et al. Identification of critical oxygen delivery for anaerobic metabolism in critically ill septic and nonseptic humans.


9. Ronco JJ, Fenwick JC, Wiggs BR, et al. Oxygen consumption is independent of increases in oxygen delivery by dobutamine in septic patients who have normal or increased plasma lactate.

Am Rev Respir Dis.

10. Manthous CA, Schumacker PT, Pohlman A, et al. Absence of supply dependence of oxygen consumption in patients with septic shock.

J Crit Care.

11. Mira JP, Fabre JE, Baigorri F, et al. Lack of oxygen supply dependency in patients with severe sepsis. A study of oxygen delivery increased by military antishock trouser and dobutamine.


. 1994;106:1524-1531.
12. Yu M, Levy MM, Smith P, et al. Effect of maximizing oxygen delivery on morbidity and mortality rates in critically ill patients: a prospective, randomized, controlled study.

Crit Care Med.

1993; 21:830-838.
13. Bone RC, Slotman G, Maunder R, et al. Randomized double-blind, multicenter study of prostaglandin E1 in patients with adult respiratory distress syndrome. Prostaglandin E1 Study Group.


. 1989;96:114-119.
14. Hayes MA, Timmins AC, Yau EH, et al. Elevation of systemic oxygen delivery in the treatment of critically ill patients.

N Engl J Med

. 1994;330: 1717-1722.
15. Gattinoni L, Brazzi L, Pelosi P, et al. A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO


Collaborative Group.

N Engl J Med.

16. Russell JA, Phang PT. The oxygen delivery/consumption controversy. Approaches to management of the critically ill.

Am J Respir Crit Care Med.

17. Hebert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group [published correction appears in

N Engl J Med.


N Engl J Med.

18. Suter PM, Fairley B, Isenberg MD. Optimum end-expiratory airway pressure in patients with acute pulmonary failure.

N Engl J Med

. 1975;292: 284-289.
19. Kaplan LJ, McPartland K, Santora TA, Trooskin SZ. Start with a subjective assessment of skin temperature to identify hypoperfusion in intensive care unit patients.

J Trauma.

20. Zion MM, Balkin J, Rosenmann D, et al. Use of pulmonary artery catheters in patients with acute myocardial infarction. Analysis of experience in 5841 patients in SPRINT registry. SPRINT Study Group.


21. Gore JM, Goldberg RJ, Spodick DH, et al. A community-wide assessment of the use of pulmonary artery catheters in patients with acute myocardial infarction.


22. Connors AF Jr, Speroff T, Dawson NV, et al. The effectiveness of right heart catheterization in the initial care of critically ill patients. SUPPORT Investigators.


23. Bernard GR, Sopko G, Cerra F, et al. Pulmonary artery catheterization and clinical outcomes: National Heart, Lung, and Blood Institute and Food and Drug Administration Workshop Report.


. 2000;283:2568-2572.
24. Gnaegi A, Feihl F, Perret C. Intensive care physicians' insufficient knowledge of right-heart catheterization at the bedside: time to act?

Crit Care Med.

25. Hoyt JD, Leatherman JW. Interpretation of the pulmonary artery occlusion pressure in mechanically ventilated patients with large respiratory excursions in intrathoracic pressure.

Intensive Care Med.

26. 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.

27. Fahey PJ, Harris K, Vanderwarf C. Clinical experience with continuous monitoring of mixed venous saturation in respiratory failure.


. 1984; 86:748-752.
28. Lee J, Wright F, Barber R, Stanley L. Central venous oxygen saturation in shock: a study in man.


29. Edwards JD, Mayall RM. Importance of the sampling site for measurement of mixed venous oxygen saturation in shock.

Crit Care Med.

1998; 26:1356-1360.
30. Fahey PJ. Clinical and physiologic rationale for continuous measurement of mixed venous oxygen saturation. In: Tobin MJ, ed.

Contemporary Management in Critical Care.

New York: Churchill Livingstone; 1995:101-118.
31. Krafft P, Steltzer H, Hiesmayr M, et al. Mixed venous oxygen saturation in critically ill septic shock patients. The role of defined events.


. 1993;103:900-906.
32. Cariou A, Monchi M, Dhainaut JF. Continuous cardiac output and mixed venous oxygen saturation monitoring.

J Crit Care.

33. Vincent JL. The available clinical tools--oxygen derived variables, lactate, and pHi. In: Sibbald WJ, Messmer K, Fink MP, eds.

Tissue Oxygenation in Acute Medicine.

New York: Springer-Verlag; 1998:193-203.
34. Gutierrez G, Clark C, Brown SD, et al. Effect of dobutamine on oxygen consumption and gastric mucosal pH in septic patients.

Am J Respir Crit Care Med.

35. Hotchkiss RS, Rust RS, Dence CS, et al. Evaluation of the role of cellular hypoxia in sepsis by the hypoxic marker [18F]fluoromisonidazole.

Am J Physiol.

36. Gore DC, Jahoor F, Hibbert JM, DeMaria EJ. Lactic acidosis during sepsis is related to increased pyruvate production, not deficits in tissue oxygen availability.

Ann Surg.

37. Bakker J, Gris P, Coffernils M, et al. Serial blood lactate levels can predict the development of multiple organ failure following septic shock.

Am J Surg.

38. Roumen RM, Redl H, Schlag G, et al. Scoring systems and blood lactate concentrations in relation to the development of adult respiratory distress syndrome and multiple organ failure in severely traumatized patients.

J Trauma.

1993; 35:349-355.
39. Bakker J, Coffernils M, Leon M, et al. Blood lactate levels are superior to oxygen-derived variables in predicting outcome in human septic shock.


. 1991;99:956-962.
40. Friedman G, Berlot G, Kahn RJ, Vincent JL. Combined measurements of blood lactate concentrations and gastric intramucosal pH in patients with severe sepsis.

Crit Care Med.

1995; 23:1184-1193.
41. Bernardin G, Pradier C, Tiger F, et al. Blood pressure and arterial lactate level are early indicators of short-term survival in human septic shock.

Intensive Care Med.

42. Jakob SM, Takala J. ARDS. Monitoring tissue perfusion.

Crit Care Clin.

43. Brown SD, Gutierrez G. Gut mucosal pH monitoring. In: Tobin MJ, ed.

Principles and Practice of Intensive Care Monitoring.

New York: McGraw Hill Co; 1998:351-368.
44. Gutierrez G, Palizas F, Doglio G, et al. Gastric intramucosal pH as a therapeutic index of tissue oxygenation in critically ill patients.


. 1992; 339:195-199.
45. Carrico CJ, Meakins JL, Marshall JC, et al. Multiple-organ-failure syndrome.

Arch Surg.

46. Dantzker DR. The gastrointestinal tract. The canary of the body?


47. Manthous CA, Hall JB, Olson D, et al. Effect of cooling on oxygen consumption in febrile critically ill patients.

Am J Respir Crit Care Med.

48. Choi PT, Yip G, Quinonez LG, Cook DJ. Crystalloids vs colloids in fluid resuscitation: a systematic review.

Crit Care Med.

49. 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.
50. Hebert PC, Yetisir E, Martin C, et al. Is a low transfusion threshold safe in critically ill patients with cardiovascular diseases?

Crit Care Med.

51. Practice parameters for hemodynamic support in adult patients with sepsis. Task Force of the American College of Critical Care Medicine, Society of Critical Care Medicine.

Crit Care Med.

52. Nunn JF. Oxygen. In: Nunn JF, ed.

Nunn's Applied Respiratory Physiology.

Oxford, UK: Butterworth-Heinemann; 1993:247-305.
53. Phang P, Cunningham K, Ronco J, et al. Mathematical coupling explains dependence of oxygen consumption on oxygen delivery in ARDS.

Am J Respir Crit Care Med.

1994;150: 318-323.