Decision Making in Emergency Critical Care

SECTION 2 - Hemodynamic Monitoring

2
Tissue Oxygenation and Cardiac Output

Geoffrey K. Lighthall and Catherine S. Reid

BACKGROUND

The two principle determinants of tissue perfusion are, (1) a mean arterial pressure (MAP) sufficient to maintain constant blood flow within key organs (i.e., within the autoregulatory range); and, (2) tissue oxygen delivery in excess of metabolic demand. Deliberate evaluation of these physiologic relationships can help define an individual's risk for organ dysfunction and shock, as well as establish end points of resuscitation.

The balance between oxygen utilization (VO2) and oxygen delivery (DO2) provides a conceptual framework for understanding the development of organ dysfunction and for the formation of resuscitation strategies. DO2 is the product of cardiac output (CO) and arterial oxygen content and may be determined using the calculations in Table 2.1. Under normal conditions, global VO2 is approximately 25% of the delivered quantity, demonstrated by mixed venous oxyhemoglobin saturations of 70% to 75%. Factors that unilaterally increase VO2 or decrease DO2, therefore, increase the oxygen extraction ratio (VO2/DO2) and lower the body's overall oxygen reserves. In extreme cases, when DO2 falls below a critical threshold (Fig. 2.1A), DO2 limits oxygen consumption. Below this point, oxygen consumption becomes supply-dependent, mitochondrial respiration is impaired, and lactic acidosis often manifests.13 The curve is a snapshot of a dynamic situation; infection and stress raise oxygen demand, while hemorrhage, hypovolemia, or impaired cardiac function compromises DO2.

TABLE 2.1 Determinants of Tissue Oxygenation

MAP, mean arterial pressure; CO, cardiac output; SVR, systemic vascular resistance; VO2, oxygen utilization; CaO2, arterial oxygen content; CvO2, mixed venous oxygen content; Hb, hemoglobin; SaO2, hemoglobin saturation; DO2, oxygen delivery; PaO2, partial pressure of arterial oxygen; O2ER, oxygen extraction ratio.

FIGURE 2.1 The key determinants of organ perfusion are depicted. In (A), the relationship between oxygen consumption (VO2) and delivery (DO2) is indicated. Patients usually function on the rightward side of the curve, where an excess of oxygen is supplied relative to demand. As delivery decreases relative to consumption, the patient moves left on the curve. A decrease in central venous oxygen saturation accompanies leftward movement on the curve. In severe cases where delivery is unable to meet metabolic demands, the patient slips beneath the critical DO2 threshold, where oxygen consumption is limited by delivery. Organ dysfunction and lactic acidosis are regarded as evidence of pathologic oxygen supply. In (B), the autoregulatory curve describing constancy of organ blood flow over a broad range of MAP is shown. Some patients with chronic hypertension have curves shifted to the right relative to the normotensive curve as shown with the dashed line. For both relationships shown, the flat horizontal portions indicate safe ranges, indicative of adequate organ blood flow and intact homeostatic mechanisms. Movement to the down-sloping portions on the left indicates decompensation, placing the patient at risk for organ failure. VO2, oxygen uptake/minute; CaO2, oxygen content of arterial blood [mainly hemoglobin]; CO, cardiac output; MAP, mean arterial pressure; SVR, systemic vascular resistance; DO2, oxygen delivery.

Most clinicians are accustomed to thinking of shock, organ failure, and perfusion not in terms of the VO2/DO2 relationship, but rather in terms of changes in blood pressure, or MAP. In the case of cellular function, these two physiologic parameters overlap significantly. When MAP drops below the autoregulatory threshold for a given organ (Fig. 2.1B), regional imbalances between VO2 and DO2occur, yet may escape detection. Note that deficits in DO2 can occur in the setting of an apparently normal MAP—a condition termed cryptic shock4—and that a desirable level of total-body DO2 can exist in conjunction with an inadequate MAP.

Resuscitation from shock states focuses on moving the patient to within the normal range on these curves (Fig. 2.1A and B). Careful inspection of the determinants underlying DO2 and MAP (Table 2.1, Fig. 2.1A) reveals the common factor of CO. As the sole factor whose improvement leads to increases in both MAP and DO2, CO optimization is typically the focus of patient examination and monitoring, fluid replacement, and other resuscitative measures. Pursuit of resuscitative targets other than CO is of less clear benefit. For example, in the wrong situation, raising MAP with alpha-adrenergic agonists may actually worsen CO and have disastrous consequences for DO2 and tissue perfusion. Similarly, raising hemoglobin levels with aggressive transfusion does not necessarily improve DO2, can produce volume overload, and can precipitate acute lung injury. For these reasons, the critical care community pays close attention to interventions that modulate CO and to monitoring systems that capture these changes. This chapter provides a review of the evolution of our understanding of these concepts; details of the evidence from specific studies of these subjects are presented in the Literature Tables of Chapters 3, 4, and 5.

HISTORICAL PERSPECTIVE

Modern hemodynamic monitoring and resuscitation have evolved significantly from decades ago, when manipulation of CO and DO2 indices to supranormal values was believed to improve patient survival. In the early 1970s, enhanced ventricular performance and increased DO2 and consumption were predicted to improve survival in trauma patients.5 Subsequent studies in surgical patients appeared to confirm the survival benefit of using a pulmonary artery catheter (PAC) to facilitate increases in CO and DO2.610 Deliberate augmentation of DO2 in medical and surgical ICU patients was the natural next step; however, therapy designed to achieve such supranormal indices repeatedly failed to improve outcomes in this patient population.1114 The difference in outcomes was believed to be due to the use of less stringent and, in some cases, clinician-generated, resuscitative end points in larger and better-controlled studies comparing pulmonary artery and CVP catheters. These studies demonstrated no advantage of pursuing supranormal indices in high-risk surgical patients,15 in patients with shock and sepsis,16 or in adult patients with acute respiratory distress syndrome.17

Critics of targeted supranormal indices noted that some patients may have reached an optimum blood pressure or DO2 at lower cardiac indices than that pursued by the experimental protocol (e.g., 3.3 vs. 4.5 L/min/m2) and suffered harm from excessive use of fluids and vasoactive agents. Additionally, these supranormal indices may have been impossible to attain in some patients with advanced age or structural heart disease. One important finding in these studies was that some patients, regardless of the treatment group, were able to raise their own CO and DO2 to very high levels and that these patients had improved survival. The collective outcomes of these studies supported the qualitative goal of optimizing CO and increasing DO2 but refuted the use of specific numerical end points in restoring organ perfusion. This approach continues to define critical care resuscitative philosophy.

OPTIMIZATION OF CARDIAC OUTPUT

CO is the product of heart rate and stroke volume. In healthy individuals, stoke volume is normally a function of preload; however, with any acute or chronic disease, stroke volume is also sensitive to ventricular performance (contractility) and afterload. Optimization of tissue perfusion requires the provider to answer four questions: (1) Is the patient fluid responsive (i.e., will a fluid challenge increase stroke volume)? (2) Is contractility adequate (i.e., does the patient need an inotrope)? (3) Does the patient need a vasopressor? and (4) Does the patient need a blood transfusion? If the patient has adequate MAP (achieved using vasopressors if needed), hemoglobin within normal range and constant across serial measurements, and a relatively static demand for oxygen (VO2), then the provider need addresses only fluid responsiveness and contractility to optimize CO. Techniques used to monitor these parameters will vary by clinical circumstance and available resources; whereas some suit initial evaluation in the ED, others with greater trending capabilities may be preferred in the ICU.

FLUID RESPONSIVENESS

Fluid responsiveness describes the ability of the heart to increase its stroke volume—and consequently CO—in response to infusion of fluids. From a patient management perspective, fluid responsiveness determines the extent to which circulatory homeostasis can be maintained with fluids alone, without the addition of inotropes or vasopressors. The decision to give a patient more fluids requires an understanding of the concepts demonstrated by the Frank-Starling curve, which describes how stroke volume responds to changes in preload (Fig. 2.2). The ascending portion of the Frank-Starling curve corresponds to the fluid-responsive phase of resuscitation, seen as a fairly linear increase in CO. Once the left ventricle reaches the plateau phase of the curve, additional fluid administration will not further improve CO and may lead to adverse consequences such as hydrostatic pulmonary edema.

FIGURE 2.2 The Frank-Starling curve describing the relationship between cardiac output (CO) and preload. Volume responsiveness describes the steep portion of the curve in which modest changes in preload result in significant increases in CO and hence MAP. In the upper region of the curve, a similar change in preload would result in a negligible change in CO.

Methods of interpreting intravascular volume range from clinical assessments (e.g., inspection of veins or a passive leg-raising test), to more invasive methods (e.g., central venous and pulmonary artery catheterization), and to newer and technically sophisticated methods (e.g., echocardiography and analysis of flow parameters). When evaluating these techniques, it is helpful to consider their ability to predict a state of fluid responsiveness versus euvolemia and how their unique characteristics may be paired with different clinical situations to yield accurate and meaningful information.

Passive Leg-Raising

The passive leg-raising test (PLR), in which the legs of a supine patient are elevated to 45 degrees, delivers a reversible endogenous fluid challenge by increasing venous return; the effect on blood pressure and heart rate is subsequently evaluated. When PLR is used in concert with an existing arterial line, changes in preload leading to increased CO and blood pressure are immediately apparent. Because fluid bolus administration is the primary alternative to a PLR, the PLR can quickly identify patients for whom fluid infusion would be of no benefit and potentially harmful. The PLR has shown good correspondence with other derived indices in predicting fluid responsiveness in patients with sepsis and pancreatitis18 and has been compared favorably with transthoracic echo19 and esophageal Doppler20in mechanically ventilated patients. PLR is a valuable technique in early patient assessment, as it requires little technical skill and does not rely on the presence of a central venous and pulmonary arterial catheter for preload assessment.

Central Venous Pressure Monitoring

Central venous pressure (CVP) is the measurement of pressure within the thorax in the superior vena cava and serves as a reasonable surrogate for right atrial pressure. Historically, CVP was widely used to estimate intravascular volume in critically ill patients,21 with the implication that CVP served as a reasonable surrogate for left ventricular preload and that some correspondence between measured values and CO existed.

The standard test for volume responsiveness was to give a fluid challenge that increases the CVP by 2 mm Hg and then determine whether it increased CO.22 A study of 83 ICU patients showed that patients with an increase in CVP of 2 mm Hg following a bolus of approximately 500 mL of isotonic crystalloids over 10 to 30 minutes had a cardiac index increase of 300 mL/min/m2. Two additional findings of the study were important: (1) only 4.5 % of the patients with a CVP more than 10 mm Hg responded to a fluid challenge; and (2) of patients who had increase in CO, 42% only had a simultaneous increase in blood pressure. The study concluded that, first, patients with a CVP of more than 10 mm Hg responded poorly to volume infusion and that 10 mm Hg likely represented euvolemia in most individuals; and second, that blood pressure increase was not a good indicator of cardiac response to a fluid challenge.23 These data, which supported the notion that CVPs in the 8 to 12 mm Hg range indicated volume repletion, were incorporated into early goal-directed therapy and subsequently into the initial versions of the Surviving Sepsis Guidelines.24,25

More recent examination of central pressures has shown CVP to be a poor indicator of intravascular volume. A healthy person may have a CVP of less than zero in an upright position (due to the influence of negative intrathoracic pressure generated during spontaneous respirations) and still have an adequate CO and be euvolemic. Conversely, CVP can be high in a patient with poor ventricular function and low CO or with good ventricular function and volume overload.26 As these common scenarios illustrate, values derived from pressure readings are most usefully considered in conjunction with a dynamic clinical response—such as blood pressure or urine output—or with another measure of CO. Meta-analyses show no difference between fluid responders and nonresponders at CVPs of a given value; a poor correlation between changes in CVP and cardiac performance following a fluid challenge; and poor correlation between blood volume and CVP.27,28 Although use of CVPs in resuscitation persists, the recommendation for their application has softened in the latest Surviving Sepsis Guidelines,29 and many believe the practice of targeted CVPs should be abandoned completely.28 Despite the inability of CVP to represent the dynamic range of fluid responsiveness, a low CVP (<5 mm Hg) in a critically ill patient is generally assumed to correlate with hypovolemia.29 Ideally, however, significant hypovolemia should be suspected clinically and acted upon empirically, without the need for invasive pressure monitoring.

Pulmonary Artery Catheterization

In the setting of cardiac and lung pathology, aberrations in right heart compliance and pulmonary vascular resistance can drastically alter the relationship between CVP and left atrial pressure. In critically ill patients with such underlying pathology, pulmonary artery catheterization remains a reasonable option for measuring both right and left heart and pulmonary artery pressures. As with the CVP, however, pulmonary artery occlusion pressure (PAOP, or “wedge” pressure) measurements are dependent on myocardial compliance. Multiple studies of ICU patients with acute illness have shown PAOP to correlate poorly or inconsistently with left ventricular end-diastolic volume (LVEDV).3033 Studies in mechanically ventilated patients receiving positive-end expiratory pressure (PEEP) show that PEEP drastically alters the relationship between PAOP and recruitable stoke volume, and surprisingly, that a right heart parameter (right ventricular end-diastolic volume; RVEDV) correlated more reliably with changes in the cardiac index.34,35

As noted, use of the PAC was historically justified by the assumption that it was desirable to target improvements in physiologic parameters to supranormal end points in critically ill patients. Use of PACs for this purpose has fallen over the last 10 years because of the reasons cited, because of relative success with CVP-based methods for resuscitation in septic shock,36 and because of unique complications of the PAC, including ventricular arrhythmias, right bundle-branch block, thromboembolism, pulmonary artery rupture,3739 and frequent misinterpretation of PAC-generated data.40,41 Despite these issues, debate continues as to whether the PAC can assist in fluid and hemodynamic management in patients with severe cardiac or pulmonary pathology. A number of randomized trials have assessed protocols for fluid and inotrope management, both with and without the PA catheter in patients undergoing major surgery,15 in patients with congestive heart failure,42 shock and ARDS,16,39 and in general ICU populations,4345 and have demonstrated no hospital or mortality benefit ascribable to the device. Future research is needed to demonstrate whether the PAC retains a useful niche in the management of the very few critically ill patients (e.g., those with severe pulmonary hypertension) in whom maintenance of CO requires careful manipulation of pulmonary artery pressures.

CARDIAC CONTRACTILITY

Central Venous Oxygen Saturation and Lactate Clearance

While a central venous catheter is of limited value in assessing fluid responsiveness and of no use in direct measurement of CO, analysis of CVP-derived central venous oxygen saturation (ScvO2) can provide insight into the adequacy of cardiac contractility and CO. ScvO2 provides a dynamic measure of the VO2/DO2 relationship; ScvO2 values of >70% are consistent with an adequate CO and perfusion status. When VO2 is constant among several serial Fick Equation measurements (i.e., all obtained with the patient at a similar level of activity and with a similar body temperature), increases and decreases in ScvO2 indicate horizontal movement along the VO2/DO2 curve (Fig. 2.1A) and thus indicate a change in CO or hemoglobin, or both. If the hemoglobin is also constant during serial measurements of ScvO2, then changes in the latter value (again, indicating horizontal movement along the curve) indicate changes in CO. In the absence of significant bradycardia, CO problems are typically due to inadequate stroke volume, either from low preload or from problems with contractility. Continuing with the condition in which VO2 and hemoglobin are constant along several measurements, if a patient is on the upper reaches of the Starling curve, and therefore is euvolemic, then abnormalities in ScvO2 are indicative of inadequate contractility and suggest the need for inotropic support. The ScvO2 target of 70% is an integral part of the Surviving Sepsis campaign's resuscitation bundle for severe sepsis,36 and careful inspection of the algorithm reveals the logic detailed above, in which ScvO2 is used to judge adequacy of contractility. Intermittent blood samplings via CVP, dialysis, or peripherally inserted central catheters lines, or the use of catheters with oximetric sensors are equally valid means of analyzing ScvO2.

Central vein oximetry is based on the assumption that the cellular machinery responsible for oxygen uptake and utilization functions normally and that changes in measured values reflect oxygen supply and demand. This is not, however, always the case. In sepsis, mitochondrial function can be impaired as a result of depletion of high-energy substrates related to the inflammatory burst; in this case, oxygen consumption is disrupted in the setting of high demand—a state termed cytopathic hypoxia.46,47 Under these conditions, central venous saturation will be deceptively normal because tissues cannot fully utilize the oxygen delivered. Organ function and adequacy of blood flow should therefore be simultaneously assessed. To this end, lactate clearance has recently been studied as an assessment of cellular function during resuscitation. In two recent multicenter trials, patient inability to normalize lactate with resuscitation was found to be an independent predictor of mortality.48,49 In a recent landmark trial, a lactate clearance of 10% was found to be equivalent to ScvO2 as both the resuscitative end point and predictor of mortality.50 Importantly, resuscitation to a ScvO2 saturation of >70% can still be associated with lactate nonclearance for the reasons stated above; in these instances, clinical assessment of organ function and lactate clearance should be used to guide ongoing resuscitation.49

Echocardiography

Transthoracic echocardiography (TEE), with its increased portability and affordability, has found a place in all acute care settings, and the modern intensivist and emergency physician are expected to be skilled in its use. Where the PAC uses pressure measurements to make volume determinations, echocardiography relies on direct visualization of the cardiac anatomy and flow dynamics. In patients with overlapping causes of circulatory failure, echocardiography can evaluate structural abnormalities, contractility, and intravascular volume in a single efficient exam.

In the last decade, the improved image quality of portable echocardiography machines has made TTE a popular tool for intravascular fluid assessment. Right heart preload can be reliably obtained by direct measurement of variation in the diameter of the inferior vena cava (IVC) with respiration and by measurement of right and left ventricular end-diastolic volumes. In one study, a 50% decrease in IVC diameter (caval index), seen in the subcostal views with spontaneous breathing, correlated with an RA pressure of <10 mm Hg (mean SD 6±5) as measured by CVP.51 Recent studies in ED settings found caval index measurement to be a useful noninvasive tool for initial estimation of CVP and, more importantly, fluid responsiveness.52 In mechanically ventilated patients, an IVC variation of 12% with respiration (delta IVC) differentiated fluid responders from nonresponders.53 In another study of mechanically ventilated septic patients, the CVP and the IVC diameter increase on inspiration (distensibility index [dIVC]) was measured before and after a gelatin fluid challenge of 7 mL/kg. Response was measured as an increase in cardiac index (CI) of 15% or more. A dIVC of >18% predicted fluid responsiveness with a sensitivity and specificity of 90%. Changes in CVP, however, correlated poorly with changes in CI or dIVC.54 Although it can be challenging to visualize the IVC in patients who are obese or post–abdominal surgery, TTE is usually able to provide a quick, noninvasive, and reliable method of assessing intravascular volume status and fluid responsiveness in most patients.

TTE is also a promising tool for noninvasive measurements of global cardiac contractility and left ventricular function. A goal-oriented exam can provide a rapid assessment of the adequacy of contractility and aid in resuscitation decision making.55,56 While this technique (discussed in detail in Chapter 6) requires an initial investment in training of medical staff, it has a subsequent high success rate and requires little time to perform. Studies have demonstrated that after initial training—including basic echocardiography, review of images, and demonstration of image acquisition and interpretation techniques—intensivists were able to successfully perform and interpret (84% correct) a limited TTE in a mean time of 11 minutes.56 Goal-directed TTE can also aid in the diagnosis of specific pathologies contributing to a patient's hemodynamic instability. TTE is safe and noninvasive, which make it an ideal tool for repeated assessment of hemodynamic variables that change as a result of interventions or the disease process itself.

As noted above, fluid responsiveness and contractility are the key factors influencing CO; an emergency physician experienced in TTE can evaluate both without the need for CVP monitoring. For the many patients who do receive central catheter placement—to facilitate safe administration of vasopressors or serial assessment of ScvO2—the volumetric views provided by echocardiography may help contextualize waveform or catheter-based values.

INFLUENCE OF CARDIOPULMONARY INTERACTIONS ON CARDIAC OUTPUT

Cardiac output and MAP interact with the respiratory system in a predictable manner. With positive pressure ventilation, venous return to the left ventricle is initially augmented, causing a rise in cardiac output and MAP during early inspiration. Following this, a decrease in RV preload caused by positive intrathoracic pressure will manifest as a drop in LV preload. The varying effects of positive pressure ventilation on preload and cardiac output are influenced by the patient's intravascular volume status. For example, with hypovolemia, the myocardium is on the steep portion of the Frank-Starling curve, such that minor variations in left ventricular preload with inspiration or expiration can cause appreciable changes in CO and MAP.

In mechanically ventilated patients, changes in arterial pressure waveforms and Doppler analysis of aortic blood flow during the respiratory cycle can be used to evaluate euvolemia and fluid responsiveness. Where CVP and other pressure-based measures of fluid responsiveness are suspect in these patients, flow-based measurements using Doppler achieve their highest accuracy. Similarly proven are several indices of fluid responsiveness derived from the interaction between positive pressure ventilation and arterial blood pressure waveforms (e.g., systolic and pulse pressure variation). Measurement of these indices is described in Table 2.2 and discussed fully in Chapter 3. Reliable parameters for predicting fluid responsiveness have not been developed for patients breathing spontaneously.

TABLE 2.2 Hemodynamic Monitoring Devices

Common means of circulatory monitoring are listed and compared for their ability to estimate intravascular volume, contractility, and cardiac output.

CO, cardiac output; MV, mechanical ventilation; PPV, pulse pressure variation; SPV, systolic pressure variation; dPAWP, change in (delta) pulmonary artery wedge pressure; TDCO, thermodilution cardiac output.

CONCLUSION

Optimization of CO is central to the maintenance of circulatory homeostasis. The emergency and critical care communities have moved away from resuscitation based on targeted values of CO and toward resuscitation based on adequacy of CO. Evaluating adequacy of CO requires assessment of fluid responsiveness and contractility. A number of invasive and noninvasive tools can be used in a flexible manner to provide rapid answers to these fundamental questions.

REFERENCES

1.Astiz ME, et al. Oxygen delivery and consumption in patients with hyperdynamic septic shock. Crit Care Med. 1987;15(1):26–28.

2.Astiz ME, et al. Relationship of oxygen delivery and mixed venous oxygenation to lactic acidosis in patients with sepsis and acute myocardial infarction. Crit Care Med. 1988;16(7):655–658.

3.Shibutani K, et al. Critical level of oxygen delivery in anesthetized man. Crit Care Med. 1983;11(8):640–643.

4.Puskarich MA, et al. Outcomes of patients undergoing early sepsis resuscitation for cryptic shock compared with overt shock. Resuscitation. 2011;82(10):1289–1293.

5.Shoemaker WC, Montgomery ES, Kaplan E, et al. Physiologic patterns in surviving and nonsurviving shock patients. Use of sequential cardiorespiratory variables in defining criteria for therapeutic goals and early warning of death. Arch Surg. 1973(106):630–636.

6.Rady MY, Edwards JD, Nightingale P. Early cardiorespiratory findings after severe blunt thoracic trauma and their relation to outcome. Br J Surg. 1992;79(1):65–68.

7.Tuchschmidt J, et al. Early hemodynamic correlates of survival in patients with septic shock. Crit Care Med. 1989;17(8):719–723.

8.Shoemaker WC, et al. Prospective trial of supranormal values of survivors as therapeutic goals in high-risk surgical patients. Chest. 1988;94(6):1176–1186.

9.Boyd O, Grounds RM, Bennett ED. A randomized clinical trial of the effect of deliberate perioperative increase of oxygen delivery on mortality in high-risk surgical patients. JAMA. 1993;270(22):2699–2707.

10.Bishop MH, et al. Prospective, randomized trial of survivor values of cardiac index, oxygen delivery, and oxygen consumption as resuscitation endpoints in severe trauma. J Trauma. 1995;38(5):780–787.

11.Gattinoni L, et al. A trial of goal-oriented hemodynamic therapy in critically ill patients. SvO2 Collaborative Group. N Engl J Med. 1995;333(16):1025–1032.

12.Tuchschmidt J, et al. Elevation of cardiac output and oxygen delivery improves outcome in septic shock. Chest. 1992;102(1):216–220.

13.Yu M, 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(6):830–838.

14.Hayes MA, et al. Elevation of systemic oxygen delivery in the treatment of critically ill patients. N Engl J Med. 1994;330(24):1717–1722.

15.Sandham JD, et al. A randomized, controlled trial of the use of pulmonary-artery catheters in high-risk surgical patients. N Engl J Med. 2003;348(1):5–14.

16.Richard C, et al. Early use of the pulmonary artery catheter and outcomes in patients with shock and acute respiratory distress syndrome: a randomized controlled trial. JAMA. 2003;290(20):2713–2720.

17.Wheeler AP, et al. Pulmonary-artery versus central venous catheter to guide treatment of acute lung injury. N Engl J Med. 2006;354(21):2213–2224.

18.Preau S, et al. Passive leg raising is predictive of fluid responsiveness in spontaneously breathing patients with severe sepsis or acute pancreatitis. Crit Care Med. 2010;38(3):819–825.

19.Lamia B, et al. Echocardiographic prediction of volume responsiveness in critically ill patients with spontaneously breathing activity. Intensive Care Med. 2007;33(7):1125–1132.

20.Monnet X, et al. Passive leg raising predicts fluid responsiveness in the critically ill. Crit Care Med. 2006;34(5):1402–1407.

21.Boldt J, et al. Volume replacement strategies on intensive care units: results from a postal survey. Intensive Care Med. 1998;24(2):147–151.

22.Magder SA, Georgiadis GS, Cheong T. Respiratory variations in right atrial pressure predict response to fluid challenge. J Crit Care. 1992;7:76–85.

23.Magder S, Bafaqeeh F. The clinical role of central venous pressure measurements. J Intensive Care Med. 2007;22(1):44–51.

24.Dellinger RP, et al. Surviving sepsis campaign guidelines for management of severe sepsis and septic shock. Crit Care Med. 2004;32(3):858–873.

25.Dellinger RP, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med. 2008;36(1):296–327.

26.Magder S. How to use central venous pressure measurements. Curr Opin Crit Care. 2005;11(3):264–270.

27.Marik PE, Baram M, Vahid B. Does central venous pressure predict fluid responsiveness? A systematic review of the literature and the tale of seven mares. Chest. 2008;134(1):172–178.

28.Marik PE, Cavallazzi R. Does the central venous pressure predict fluid responsiveness? An updated meta-analysis and a plea for some common sense. Crit Care Med. 2013;41(7):1774–1781.

29.Dellinger RP, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med. 2013;41(2):580–637.

30.Fontes ML, et al. Assessment of ventricular function in critically ill patients: limitations of pulmonary artery catheterization. Institutions of the McSPI Research Group. J Cardiothorac Vasc Anesth. 1999;13(5):521–527.

31.Spinelli L, et al. Losartan treatment and left ventricular filling during volume loading in patients with dilated cardiomyopathy. Am Heart J. 2002;143(3):433–440.

32.Tousignant CP, Walsh F, Mazer CD. The use of transesophageal echocardiography for preload assessment in critically ill patients. Anesth Analg. 2000;90(2):351–355.

33.Hansen RM, et al. Poor correlation between pulmonary arterial wedge pressure and left ventricular end-diastolic volume after coronary artery bypass graft surgery. Anesthesiology. 1986;64(6):764–770.

34.Cheatham ML, et al. Right ventricular end-diastolic volume index as a predictor of preload status in patients on positive end-expiratory pressure. Crit Care Med. 1998;26(11):1801–1806.

35.Diebel L, et al. End-diastolic volume versus pulmonary artery wedge pressure in evaluating cardiac preload in trauma patients. J Trauma. 1994;37(6):950–955.

36.Rivers E, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345(19):1368–1377.

37.Coulter TD, Wiedemann HP. Complications of hemodynamic monitoring. Clin Chest Med. 1999;20(2):249–267. vii.

38.Hadian M, Pinsky MR. Evidence-based review of the use of the pulmonary artery catheter: impact data and complications. Crit Care. 2006;10(suppl 3):S8.

39.Wheeler AP, Bernard GR, Thompson BT, et al. Pulmonary-artery versus central venous catheter to guide treatment of acute lung injury. N Engl J Med. 2006;354(21):2213–2224.

40.Jacka MJ, et al. Pulmonary artery occlusion pressure estimation: how confident are anesthesiologists? Crit Care Med. 2002;30(6):1197–1203.

41.Summerhill EM, Baram M. Principles of pulmonary artery catheterization in the critically ill. Lung. 2005;183(3):209–219.

42.Binanay C, et al. Evaluation study of congestive heart failure and pulmonary artery catheterization effectiveness: the ESCAPE trial. JAMA. 2005;294(13):1625–1633.

43.Guyatt G. A randomized control trial of right-heart catheterization in critically ill patients. Ontario Intensive Care Study Group. J Intensive Care Med. 1991;6(2):91–95.

44.Harvey S, et al. Assessment of the clinical effectiveness of pulmonary artery catheters in management of patients in intensive care (PAC-Man): a randomised controlled trial. Lancet. 2005;366(9484):472–477.

45.Rhodes A, et al. A randomised, controlled trial of the pulmonary artery catheter in critically ill patients. Intensive Care Med. 2002;28(3):256–264.

46.Fink MP. Bench-to-bedside review: cytopathic hypoxia. Crit Care. 2002;6(6):491–499.

47.Fink M. Cytopathic hypoxia in sepsis. Acta Anaesthesiol Scand Suppl. 1997;110:87–95.

48.Puskarich MA, et al. Whole blood lactate kinetics in patients undergoing quantitative resuscitation for severe sepsis and septic shock. Chest. 2013;143(6):1548–1553.

49.Arnold RC, et al. Multicenter study of early lactate clearance as a determinant of survival in patients with presumed sepsis. Shock. 2009;32(1):35–39.

50.Jones AE, et al. Lactate clearance vs central venous oxygen saturation as goals of early sepsis therapy: a randomized clinical trial. JAMA. 2010;303(8):739–746.

51.Kircher BJ, Himelman RB, Schiller NB. Noninvasive estimation of right atrial pressure from the inspiratory collapse of the inferior vena cava. Am J Cardiol. 1990;66(4):493–496.

52.Nagdev AD, et al. Emergency department bedside ultrasonographic measurement of the caval index for noninvasive determination of low central venous pressure. Ann Emerg Med. 2010;55(3):290–295.

53.Feissel M, et al. The respiratory variation in inferior vena cava diameter as a guide to fluid therapy. Intensive Care Med. 2004;30(9):1834–1837.

54.Barbier C, Loubières Y, Schmit C. Respiratory changes in inferior vena cava diameter are helpful in predicting fluid responsiveness in ventilated septic patients. Intensive Care Med. 2004;30:1740–1746.

55.Jensen MB, et al. Transthoracic echocardiography for cardiopulmonary monitoring in intensive care. Eur J Anaesthesiol. 2004;21(9):700–707.

56.Manasia AR, et al. Feasibility and potential clinical utility of goal-directed transthoracic echocardiography performed by noncardiologist intensivists using a small hand-carried device (SonoHeart) in critically ill patients. J Cardiothorac Vasc Anesth. 2005;19(2):155–159.