Home Educatione-Learning Update on the Evaluation of Intravascular Fluid Status in Critically Ill Patients
  • Specialties
  • Critical Care, Pulmonary
  • Learning Categories
  • Learning Category 2: Self-Directed Learning
  • CME Credit
  • Credit not available as of July 1, 2013

Update on the Evaluation of Intravascular Fluid Status in Critically Ill Patients

PCCSU Volume 25, Lesson 7


The American College of Chest Physicians offers this lesson as a review of a previously offered self-study program. The program provides information on pulmonary, critical care, and sleep medicine issues. CME is no longer available for the PCCSU program.


  • Update your knowledge and understanding of pulmonary medicine topics.
  • Update your knowledge and understanding of critical care medicine topics.
  • Update your knowledge and understanding of sleep medicine topics.
  • Learn clinically useful practice procedures.

CME Availability

Effective July 1, 2013, PCCSU Volume 25 is available for review purposes only.

Effective December 31, 2012, PCCSU Volume 24 is available for review purposes only.

Effective December 31, 2011, PCCU Volume 23 is available for review purposes only. CME credit for this volume is no longer being offered

Effective December 31, 2010, PCCU Volume 22 is available for review purposes only. CME credit for this volume is no longer being offered.

Accreditation Statement

The American College of Chest Physicians is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians.

CME Statement

Credit no longer available as of July 1, 2013.

Disclosure Statement

The American College of Chest Physicians (CHEST) remains strongly committed to providing the best available evidence-based clinical information to participants of this educational activity and requires an open disclosure of any potential conflict of interest identified by our faculty members. It is not the intent of CHEST to eliminate all situations of potential conflict of interest, but rather to enable those who are working with CHEST to recognize situations that may be subject to question by others. All disclosed conflicts of interest are reviewed by the educational activity course director/chair, the Education Committee, or the Conflict of Interest Review Committee to ensure that such situations are properly evaluated and, if necessary, resolved. The CHEST educational standards pertaining to conflict of interest are intended to maintain the professional autonomy of the clinical experts inherent in promoting a balanced presentation of science. Through our review process, all CHEST CME activities are ensured of independent, objective, scientifically balanced presentations of information. Disclosure of any or no relationships will be made available for all educational activities.

CME Availability

Volume 25 Through June 30, 2013
Volume 24 Through December 31, 2012
Volume 23 Through December 31, 2011
Volume 22 Through December 31, 2010

Hardware/software requirements: Web browsing device with working Web browser.

PCCSU Volume 25 Editorial Board

Steven A. Sahn, MD, FCCP

Director, Division of Pulmonary and Critical Care Medicine, Allergy, and Clinical Immunology
Medical University of South Carolina
Charleston, SC

Dr. Sahn has disclosed no significant relationships with the companies/organizations whose products or services may be discussed within Volume 25.

Deputy Editor
Richard A. Matthay, MD, FCCP

Professor of Medicine
Section of Pulmonary and Critical Care Medicine
Yale University School of Medicine
New Haven, CT

Dr. Matthay has disclosed no significant relationships with the companies/organizations whose products or services may be discussed within Volume 25.

Alejandro C. Arroliga, MD, FCCP
Professor of Medicine
Texas A&M Health Science Center
College of Medicine
Temple, TX

Dr. Arroliga has disclosed no significant relationships with the companies/organizations whose products or services may be discussed within Volume 25.

Paul D. Blanc, MD, FCCP
Professor of Medicine
University of California, San Francisco
San Francisco, CA

Dr. Blanc has disclosed significant relationships with the following companies/organizations whose products or services may be discussed within Volume 25:

National Institutes of Health, Flight Attendants Medical Research Institute – university grant monies
University of California San Francisco, US Environmental Protection Agency, California Environmental Protection Agency Air Resources Board – consultant fee
Habonim-Dror Foundation Board of Trustees – fiduciary position

Guillermo A. do Pico, MD, FCCP
Professor of Medicine
University of Wisconsin Medical School
Madison, WI

Dr. do Pico has disclosed no significant relationships with the companies/organizations whose products or services may be discussed within Volume 25.

Ware G. Kuschner, MD, FCCP
Associate Professor of Medicine
Stanford University School of Medicine
Palo Alto, CA

Dr. Kuschner has disclosed no significant relationships with the companies/organizations whose products or services may be discussed within Volume 25.

Teofilo Lee-Chiong, MD, FCCP
Associate Professor of Medicine
National Jewish Medical Center
Denver, CO

Dr. Lee-Chiong has disclosed significant relationships with the following companies/organizations whose products or services may be discussed within Volume 25:

National Institutes of Health – grant monies (from sources other than industry)
Covidien, Respironics, Inc. – grant monies (from industry-related sources)
Elsevier – consultant fee

Margaret Pisani, MD, MPH, FCCP
Assistant Professor of Medicine
Yale University School of Medicine
New Haven, CT

Dr. Pisani has disclosed no significant relationships with the companies/organizations whose products or services may be discussed within Volume 25.

Stephen I. Rennard, MD, FCCP
Professor of Medicine
University of Nebraska Medical Center
Omaha, NE

Dr. Rennard has disclosed significant relationships with the following companies/organizations whose products or services may be discussed within Volume 25:

AstraZeneca, Biomark, Centocor, Novartis – grant monies (from industry-related sources)

Almirall, Aradigm, AstraZeneca, Boehringer Ingelheim, Defined Health, Dey Pharma, Eaton Associates, GlaxoSmithKline, Medacrop, Mpex, Novartis, Nycomed, Otsuka, Pfizer, Pulmatrix, Theravance, United Biosource, Uptake Medical, VantagePoint – consultant fee/advisory committee

AstraZeneca, Network for Continuing Medical Education, Novartis, Pfizer, SOMA – speaker bureau

Ex Officio
Gary R. Epler, MD, FCCP

Clinical Associate Professor of Medicine
Harvard Medical School
Brigham & Women's Hospital
Boston, MA

Dr. Epler has disclosed no significant relationships with the companies/organizations whose products or services may be discussed within Volume 25.

Lilly Rodriguez
ACCP Staff Liaison

By Sumit Singh, MD; and Geoffrey Lighthall, MD, PhD

Dr. Singh is Fellow, Department of Anesthesia and Critical Care, and Dr. Lighthall is Associate Professor, Department of Anesthesia, Stanford University Medical Center, Stanford, California.

Drs. Singh and Lighthall have disclosed no significant relationships with the companies/organizations whose products or services may be discussed within this chapter.


  1. Understand the concept of fluid responsiveness.
  2. Describe the various methods to assess intravascular volume status.
  3. Understand that dynamic parameters derived from cardiorespiratory interactions are more reliable for intravascular volume assessment than static parameters.
  4. Understand echocardiography as a method of assessment of intravascular volume.

Key words: cardiac output; central venous pressure; echocardiography; esophageal Doppler monitoring; intravascular volume assessment; passive leg raising; pulmonary artery occlusion pressure; stroke volume variability; systolic pressure variability

Abbreviations: CI = cardiac index; CO = cardiac output; CVP = central venous pressure; dIVC = distensibility index of the inferior vena cava; dPP = difference in pulse pressure; EDM = esophageal Doppler monitoring; FTc = corrected flow time; IVC = inferior vena cava; PAC = pulmonary artery catheter; PAOP = pulmonary artery occlusion pressure; PLR = passive leg raising; SPV = systolic pressure variability; Svo2 = mixed venous oxygen saturation; SVV = stroke volume variability; TEE = transesophageal echocardiogram; TTE = transthoracic echocardiogram


The ability to judge intravascular volume is fundamental to understanding the evolution and management of critical illness. Insufficient intravascular volume can result in decreased oxygen delivery to tissues and impair oxidative metabolism, while fluid excesses can lead to fluid extravasation and potential impairment of gas exchange and bowel function.

Nearly all aspects of hemodynamic management in critically ill patients revolve around explicit knowledge or assumptions regarding intraventricular volume. Two concepts are applicable to the discussion of fluids in critical illness. Euvolemia refers to a range of ventricular volumes that allow sufficient cardiac output (CO) to meet tissue oxygen needs. With euvolemia, neither diuresis nor fluid administration is necessary. Likewise, use of vasopressors and inotropes to maintain blood pressure is generally inappropriate if a patient is hypovolemic, while the use of additional fluids is counterproductive when hypervolemic. Fluid responsiveness refers to the extent to which circulatory homeostasis can be maintained with fluids alone, vs the use of inotropes and vasopressors. Most diagnostic and therapeutic paradigms related to fluid administration center around the Frank-Starling relationship, which describes changes in stroke volume resulting from changes in cardiac preload. Fluid responsiveness implies that the patient is on the ascending portion of the Frank-Starling curve and has “recruitable” CO.1 However, fluid administration does not always improve hemodynamics. Once the left ventricle is functioning near the plateau of the Frank-Starling curve, fluid loading has little effect on CO and will likely result in pulmonary edema, as well as worsen lung compliance and gas exchange. Thus, a key value in identifying fluid-responsive patients is the ability to make prospective decisions as to who should or should not receive fluids before they are given.

In a broad sense, if euvolemia is the goal of fluid use in resuscitation, then fluid responsiveness reflects the process of maintaining euvolemia. In evaluating the various techniques applicable to analyzing fluid status, it is worth contrasting their utility in predicting fluid responsiveness vs euvolemia, as some techniques perform quite well in one category but not in another. Methods of analyzing intravascular volume range from clinical assessments such as inspection of veins and passive leg raising (PLR), to more invasive methods such as central venous and pulmonary artery catheterization, to newer technically intensive methods such as echocardiography and pressure wave analysis. This paper will review the currently available techniques for evaluation of intravascular volume and fluid responsiveness.

Central Venous Pressure

Central venous pressure (CVP) is the pressure measured in the superior vena cava (SVC). Due to its proximity and continuity with the right side of the heart, pressures measured within the SVC are a surrogate measurement of cardiac preload. The technique was originally described by Hughes and Magovern in 1959 and is a general starting point for many in monitoring intravascular volume.2 In one survey of intensivists,3 93% reported using CVP to guide fluid therapy.

CVP is measured at end expiration; at this point, the chest recoil and lung elasticity forces are balanced and no net external pressure is exerted on the heart and central vasculature. Artifacts that may skew the end-expiratory CVP include forceful expiration, increased pericardial pressure, and the use of positive end-expiratory pressure.4 The part of the CVP waveform that accurately reflects the preload is called the z point, or the leading edge of the c wave—the moment in the cardiac cycle when the tricuspid valve is open and the catheter tip is in continuity with the right ventricle (Fig 1). The same concept of measuring pressure between a and c waves at end-expiration applies to both CVP and wedged pulmonary artery catheters (PACs) and during all forms of ventilation.


Figure 1. A typical central venous pressure (CVP) waveform (lower panel) with accompanying ECG waveform (above). The a, c, and v waves are shown, along with the z point, indicating the appropriate time in the cardiac cycle for CVP measurement. CVP measurements should be obtained at end-expiration.

Two concepts related to CVP monitoring are important for its most meaningful use. First, single-point estimates of CVP are of limited clinical value. Multiple values and trends are useful but need to be interpreted in conjunction with some reference to CO such as ventricular volume, blood pressure trends, or central oxyhemoglobin saturation. Second, it is important to develop some idea regarding what CVP values represent euvolemic and hypervolemic states. A failing ventricle will display a flattened Starling curve, with a plateau portion reached for lower preload values. Fluid loading beyond the plateau portion of the Starling curve will do little to increase stroke volume.5,6

The resuscitation scheme advanced by the Society of Critical Care Medicine’s Surviving Sepsis Campaign recommends a CVP of 8 to 12 mm Hg, and 12 to 15 mm Hg in patients supported by mechanical ventilation.5 This recommendation is based, in part, on experimental work in critically ill patients demonstrating, on the whole, that fluid loading in excess of a 10 mm Hg CVP failed to create meaningful increases in CO.5 While use of these CVP values and other parameters as part of a care bundle has certainly improved survival, we believe it is erroneous to interpret such findings in isolation. Some patients will be fully resuscitated at CVPs of 6 to 7 mm Hg, while others may be hypovolemic at 10 mm Hg. Additional caution should be exercised when using values from femoral catheterization. Femoral central line catheterization has been shown to be of good correlation in patients who are supine and spontaneously breathing; however, the readings are around 0.5 mm Hg lower than those from the SVC but rarely more than 3 mm Hg different. Femoral/inferior vena caval (IVC) measurements are not likely to be reliable in patients with high intraabdominal pressures, high airway pressures, or high positive end-expiratory pressure.7

In critically ill patients, CVP may not accurately reflect left ventricular preload.6 Limitations include the differences between right and left volume compliance, and therefore, Starling curves, and interference from abdominal pressure changes. Likewise, catecholamine fluxes due to changes in wakefulness, stress, or therapeutic infusion can raise tone in venous capacitance vessels. Such fluxes will temporarily redistribute some blood centrally to produce central pressures that may overestimate circulating blood volume.

An additional advantage of central vein catheterization is the ability to measure central venous oxyhemoglobin saturation (Scvo2). Changes in saturation reflect changes in systemic oxygen extraction from hemoglobin; the value will decrease with increased uptake by metabolic demand (Vo2), decreased CO, or loss of hemoglobin. An Scvo2 of 70% is consistent with normal function of the cardiovascular system. At stable levels of temperature and activity (Vo2), and hemoglobin, changes in (Scvo2) can be attributed to changes in CO. Scvo2 monitoring is an integral component of early goal-directed therapy for severe sepsis and septic shock.8 Overall, while there are come caveats to central pressure monitoring, per se, the ability to associate pressure readings and trends with Scvo2 values provides valuable information regarding euvolemia and adequacy of tissue perfusion.

Pulmonary Artery Catheters and Pulmonary Artery Occlusion (Wedge) Pressures

More than 40 years ago, Harold Swan and William Ganz introduced a balloon-tipped catheter that was directed by flow to a proximal branch of the pulmonary artery.9 PACs have the theoretical advantage of bypassing the right side of the heart and establishing continuity with the left ventricle, a more appropriate measurement of systemic preload. The “wedge” or pulmonary artery occlusion pressure (PAOP) describes the pressure reading obtained when the balloon is inflated to isolate pulmonary artery pressures from downstream pressures. These catheters also allow for direct measurement of pulmonary artery pressures, mixed venous oxyhemoglobin saturation (Svo2), and CO. The first iterations of the catheter used thermodilution to measure CO; however, newer options include thermistors that permit continuous measurements of CO and Svo2. Software packages from most monitoring systems allow automated calculation of derived indices such as stroke volumes, right and left ventricular ejection stroke work indices, and systemic vascular resistance.

In principal, the ability to correlate functional values such as CO and Svo2 with ventricular filling pressures should allow for more precise characterization of ventricular function and differentiation between states that would be best treated with either fluids, inotropes, or vasopressors. In clinical practice, there is no comparative study demonstrating superiority of the PAC over other techniques.10 The limitations with CVP monitoring also exist with PAC, namely, that the correlation of end-diastolic pressure and volume is curvilinear and undergoes right or left shift with sepsis, shock, myocardial ischemia, positive pressure ventilation, and vasoactive drugs.11 Second, the assumption catheter to ventricle continuity is valid only when the catheter is in West zone 3 of the lung (ie, where pulmonary venous pressure is greater than airway pressure). Third, the compliance of the left ventricle may be reduced in conditions such as myocardial ischemia, infiltrative diseases, and left ventricle hypertrophy, leading to overestimations of preload. Finally mitral valve stenosis and regurgitation can produce difficulty in interpreting wedge pressures.12

Compared with the dynamic parameters of preload estimation discussed below, CVP and PAOP (called static parameters) are not as accurate in predicting fluid responsiveness. They cannot differentiate fluid responsive from nonresponsive with baseline values, and a strong correlation has not been defined between cardiac filling pressures before volume expansion and the hemodynamic response to volume expansion.13 A normal or high PAOP does not necessarily mean that the heart is maximally filled, and intermediate readings do not indicate the adequacy of cardiac filling.14 PAC data are notorious for misinterpretation by intensivists, leading many to question the safety of its use.12

Indices Derived From Cardiorespiratory Interactions

CO and blood pressure interact with the respiratory system in a dynamic manner according to the relationships indicated in Table 1. With positive pressure ventilation, blood in the pulmonary circulation is pushed toward the left ventricle, causing a rise in preload, CO, and blood pressure during early inspiration. Later, the decrease in right ventricle preload caused by positive intrathoracic pressure causes a drop in left ventricle preload and systemic blood pressure.15

Table 1General Cardiorespiratory Interactions Used to Predict Volume Responsiveness

Mode of Inspiration RV Preload RV Afterload LV Preload LV Afterload

RV = right ventricle; LV = left ventricle.

It has been long known that pulsus paradoxus, or an inspiratory fall in systolic blood pressure by more than 10 mm Hg, is seen in critically ill patients with hypovolemia. With hypovolemia, the myocardium is at the steep portion of the Frank-Starling curve, so any minor variation in the preload with inspiration or expiration can cause appreciable changes in CO and blood pressure. Indices of intravascular fluid and preload assessment derived from positive pressure ventilator-induced arterial blood pressure changes include systolic pressure variability (SPV), the respiratory systolic variation test, stroke volume variability (SVV), and respiratory changes in arterial pulse pressure.

Systolic Pressure Variability
SPV is defined as the range of systolic blood pressure measured during a single positive pressure breath. Using the pressure at end-expiration to define baseline systolic pressure, the SPV has two components: delta up and delta down, corresponding to systolic pressure waves read at peak amplitude of early inspiration (the upward component) and at end-inspiration (downward). The total amplitude variation, or the sum of delta up and delta down, is thus the SPV. Figure 2 indicates how these components are identified and measured.


Figure 2. A, Esophageal Doppler tracings from descending aorta blood flow. Flow time and peak velocity are indicated by dashed and vertical arrows, respectively. Parameters derived from the latter measurements include corrected flow-time (FTc), acceleration, and stroke distance. FTc is a derived parameter analogous to the corrected QT interval. The area under the curve is equivalent to stroke distance, the length traveled by an erythrocyte during a single cardiac cycle. B, The vertical arrow in this panel also demonstrates the stroke distance. Assuming that flow is via a cylindrical path, the stroke volume (SV) is the product of aortic cross-sectional area and the stroke distance. SV is indicated as the white volume within the cylinder. Aortic blood flow (L/min) is calculated from the product of heart rate and calculated SV, and is not exactly equal to true cardiac output, as approximately 10% of total cardiac output is diverted to subclavian and cerebral arteries upstream to the area of measurement. AUC = area under the curve.

SPV can be helpful in diagnosing and managing hypovolemia as well as revealing the presence of heart failure.16,17,19 No clear threshold values have been determined for SPV to indicate hypovolemia, although many use a value of 12 mm Hg to indicate hypovolemia. Rooke and colleagues18 showed that absence of hypovolemia is confirmed by SPV value <5 mm Hg and delta-down component below 2 mm Hg. The presence of SPV in patients believed to be euvolemic may lead to the consideration of myocardial dysfunction.16,17 The afterload reduction from positive pressure variation leads to decreased wall stress and more efficient ejection as reflected by an increase in the delta-up component.16

Additional modes of analyzing respiratory-induced changes in arterial blood pressure have been developed and validated; three will be briefly noted. The respiration systolic variation test uses the slope of the line relating systolic blood pressures to airway pressures during pressure control ventilation at inspiratory pressures of 5, 10, 15, and 20 cm H2O.20 In one trial, the calculated downslope was able to reflect the degree of hemorrhage and resuscitation after volume expansion.21 Respiratory changes in arterial pulse pressure (dPP) is the difference between pulse pressures of greatest (PPmax) and least magnitude (PPmin) over several respiratory cycles, divided by the mean of the two values and expressed as a percentage. In a study of patients with sepsis, a dPP of 13% was able to distinguish between responders and nonresponders to a intravascular vole expansion with reasonable accuracy.22 SVV is an automatically derived index based on real time CO and SV measurements from a commercially available device. One study found that SVV and SPV were more sensitive indicators of fluid responsiveness (decreased variation after a fluid challenge) than CVP.23 Devices that incorporate SVV and CO measurements are likely to be more valuable than single-modality devices and may prove to be safer than pulmonary artery catheterization. With the exception of SVV, all of the indices above can be derived from a manual printout of an arterial line trace. It is important to reconfigure and rescale the waveforms from a chart recorder in order to produce these measurements (Fig 2).

Echocardiography has historically been used to estimate cardiac function and guide therapy in outpatients. Despite its early use in the late 1970s, bedside ultrasound has seen rapid growth in the last 5 years in emergency room and ICU settings. In the ICU, where the causes of circulatory failure often overlap, the ability to rapidly evaluate contractility, intraventricular volume, and structural abnormalities has made its use indispensable. Both transthoracic echocardiogram (TTE) and transesophageal echocardiogram (TEE) are applicable to this discussion.

Transthoracic Echocardiography: The noninvasiveness, ease of use, and improved image quality produced by portable ultrasound machines are making TTE an increasingly popular tool for fluid and functional evaluation of the heart. Basic instruction in TTE is obtainable at most critical care workshops in the United States. Preload assessments generally involve direct measurement of vena cava diameter (Fig 3) and its response to respiration, and inspection of ventricular filling and end systolic volumes. In patients who are spontaneously breathing, Kircher and colleagues24 showed that more than 50% decrease in the IVC diameter with inspiration (caval respiratory index) was related to right atrium pressure <10 mm Hg (mean ± standard deviation, 6±5). A recent study25 of emergency department patients found this correlation to be useful in the initial assessment of patients.


Figure 3. Transthoracic echocardiogram with the subcostal view of the inferior vena cava passing though the liver and draining into the right atrium.

With positive pressure ventilation, variation in IVC diameter can represent fluid responsiveness. In patients supported by mechanical ventilation, greater IVC variations with respiration [DeltaD (IVC)] are noted in patients who respond to fluid boluses with an increase in CO. A 12% DeltaD IVC cutoff value allowed the identification of responders with positive and negative predictive values of 93% and 92%, respectively.26 Barbier and colleagues27 studied IVC variations in response to a fluid challenge in 23 patients with sepsis supported by mechanical ventilation, using the subxiphoid view of the heart. Three of the 23 patients were excluded because of poor visualization, a result of obesity in one patient and abdominal surgery in two patients. Measurements of the IVC were performed before and again immediately after a 7 mL/kg volume expansion using gelatin. The distensibility index of the IVC (dIVC), which reflects the increase in its diameter on inspiration, was calculated and compared with cardiac index (CI), CVP, blood pressure, and heart rate.27 Ten patients were responders and another 10 were nonresponders, depending on whether the CI increased by 15% or not. Only dIVC was found to reliably correlate with the CI in responders and nonresponders; a dIVC >18% predicted fluid responsiveness with both sensitivity and specificity of 90%. Interestingly, CVP failed to predict responsiveness to volume expansion, and when CVP did change, it correlated poorly with CI (r=0.17, P=.45) and dIVC.27 Limitations to the use of TTE and its derived indices are difficulty in IVC visualization in obese patients and in some surgical patients, anatomic variation, and artifacts from abdominal pressure variations. In spite of these limitations, it is our experience that visualizing the IVC is possible in approximately 80% of patients, and it has been an increasingly helpful and popular method for rapid assessment of intraventricular volume in the ICU. Here are additional video images demonstrating an echocardiographic-based fluid assessment.

Transesophageal Echocardiography: Since its introduction in the early 1980s, TEE has rapidly established itself as a diagnostic and monitoring tool in the operating room and ICU settings. In critical care, it is recommended that TEE be used when diagnostic information expected to alter management cannot be obtained by TTE or other modalities in a timely fashion.28 In practical terms, the availability of equipment and personnel capable of performing TEE might make this a technique to resort to earlier. TEE is especially practical and much easier in patients who are intubated; however, gastric and feeding tubes may need to be removed.

In the same fashion that IVC variation is used from TTE data, SVC is easily visualized by TEE and offers the same information. In a study of 22 patients supported by mechanical ventilation with acute lung injury related to septic shock, a marked decrease in right ventricle stroke volume and SVC collapsibility was noted during inspiration in patients who were hypovolemic. Blood volume expansion corrected the collapse, significantly decreased right ventricle stroke volume variations, and increased CI.29 Furthermore, in 66 patients with sepsis and acute lung injury supported by mechanical ventilation, the same group demonstrated that a collapsibility index above 36% predicted a >11% increase in CI following volume expansion (specificity of 100% and sensitivity of 90%).30

For both TTE and TEE, sonography allows direct visualization and assessment of left and right ventricular function and thus helps guide the decision between fluid challenge and use of vasopressors or/and inotropes. Visual estimates of euvolemia can be made with reasonable confidence, while numerical measures of fluid responsiveness depend upon computationally intensive serial measurements of two- and three-dimensional images. For example, large left ventricular volumes with minimal change between systolic and diastolic dimensions generally indicate a patient who will not increase CO with additional fluids. Patients with near obliteration of the ventricular cavity at end systole are generally fluid responsive. Fluid challenges in the latter cases, followed by inspection of ventricular filling by echocardiogram and central venous pressures can be used in concert to establish pressures that reflect inadequate and euvolemic ventricular filling. Additionally, patients with empty left ventricles and full or enlarged right ventricles should entertain thoughts of high pulmonary artery pressures either from vascular disease or from thromboembolism.

Esophageal Doppler Monitoring
Estimates of stroke volume and CO can now be made via small flexible probes of the same approximate size and tolerability as a nasogastric tube.31 According to the Doppler principle, blood flow velocity can be estimated by analysis of ultrasound waves reflected from flowing blood cells. Figure 4 shows the flow-time tracing generated by esophageal Doppler. The flow time represents the time needed for the left ventricle to eject the stroke volume, and it is calculated from the start of waveform upstroke to its return to baseline; because it is also dependent on the cycle time or heart rate, a derived parameter—analogous to that applied to the QT interval on electrocardiograms (QTc)—called the “corrected flow-time” (FTc) is used. Clinical preload assessments are made according to the presumption that the larger the left ventricle end-diastolic volume or preload, the longer the flow time. Indeed, FTc was found to be decreased with decreases in preload created by hemorrhage and vasodilators.32 Similar to systolic pressure analysis, use of cardiorespiratory interactions to predict volume responsiveness are equally applicable to Doppler flow measurements. Monnet and colleagues33 studied 38 patients supported by mechanical ventilation and found that fluid responsiveness can be predicted by the respiratory variation in aortic blood flow and/or the flow time (FTc). A respiratory variation in aortic flow before volume expansion of at least 18% predicted fluid responsiveness with a sensitivity of 90% and a specificity of 94%. In hemodynamically unstable patients with spontaneous respiration, heart-lung interaction indices such as stroke volume and arterial pulse pressure variations are less reliable in predicting volume responsiveness.


Figure 4. Arterial line tracings showing systolic pressure variation with respiration. The pressure at end-expiration defines baseline systolic pressure. SPV has two components: delta up and delta down, corresponding to systolic pressure waves reading at peak amplitude of early inspiration (the upward component or delta up) and at end-inspiration (downward component or delta down). The total amplitude variation or the sum of delta up and delta down is, thus, the SPV, shown with the arrow on the right of the diagram. SPV = systolic pressure variability.

Additional information regarding the circulatory system can be obtained with esophageal Doppler. The peak velocity and its upslope (acceleration) can be used to evaluate both contractility and left ventricular sensitivity to afterload. Preload and afterload can affect both acceleration and flow-time, so all measurements should be interpreted in the context of a patient’s underlying physiology and the interventions being performed.

Esophageal Doppler monitoring (EDM) has been used in a variety of clinical settings to measures CO with acceptable accuracy and precision. The area under the flow-time curve is the “stroke distance,” which, combined with direct or indexed measurements of aortic cross-sectional area, allows measurement of stroke volume. Stroke volume can therefore be derived by multiplying flow velocity and ejection time with the cross-sectional area of the aorta. Unlike other modalities of echocardiogram, the low profile and stable placement of the esophageal probe allows for serial measurements and trending over time, and as various interventions are carried out to optimize circulatory flow.

In surgical patients, esophageal Doppler has been used to guide fluid resuscitation and more rapidly achieve clinical milestones such as bowel motility, feeding, and hospital discharge following large operations.34,35 Studies in intensive care patients have mainly compared EDM with other modalities, such as PAC, and have been less focused upon discrete outcomes. Overall, the ICU experience demonstrates EDM to be a viable alternative to traditional measurements when brief probe insertion is feasible.36

Limitations to both esophageal Doppler and TEE include the potential for esophageal damage, especially in patients with existing disease. Esophagectomy, head/neck surgery, and other recent surgery should lead a clinician to other alternatives. The presence of severe lung disease, high intrathoracic pressures, severe aortic valve disease, thoracic flow abnormalities, and intraaortic balloon pump can interfere with the accuracy of EDM.31

Passive Leg Raising
PLR is an easily performed bedside test. The concept behind the test is an “auto fluid challenge” that evaluates the blood pressure response to an increase in venous return. A simple way to perform PLR involves transferring a semirecumbent patient positioned at 45° into the PLR position by using the pivotal motion of the bed.37 PLR is a quick, stand-alone test, but it can also be used to validate impressions gained by any and all of the techniques mentioned previously. The correspondence between PLR and other derived indices to predict fluid responsiveness has been specifically evaluated using TTE in patients supported by mechanical ventilation with spontaneous respiration,38 esophageal Doppler in patients supported by a ventilator,39 and in patients with sepsis and pancreatitis.40 PLR is best accomplished with cooperative or adequately sedated patients; wakefulness and excitation precipitated by the maneuver creates all sorts of interference that effectively invalidates any information obtained.

Putting It All Together

From the discussion above, it can be understood that there is no single technique of volume monitoring that is ideal for both assessments of euvolemia and volume responsiveness. Critical care clinicians rarely depend upon a single monitor to do it all. Rather, most patient resuscitations pass through phases where one technique may serve the specific needs of that situation better than others, as seen in the following discussion of two patients.

Case 1
For example, consider a patient with mild congestive heart failure and diverticulosis who is experiencing hypotension a few days following a colectomy. One liter of fluid is infused over 15 min in an attempt to improve his blood pressure; during this time, an arterial line is placed. With the fluid, the patient’s blood pressure has responded some, but is still low. Raising the patient’s legs increases the mean arterial pressure from 60 to 73 mm Hg, so an additional 3 L is infused over the next 1.5 h. A CVP monitor is placed to follow pressure changes with fluid and to evaluate the adequacy of tissue oxygen delivery by vena cava oxygen saturation. After further consideration of the situation, there is now a strong suspicion that the patient has developed sepsis from a urinary catheter. The patient’s blood pressure and central venous saturation are acceptable, with a CVP of 13 mm Hg and without the need for an inotrope or vasopressor. A TEE confirmed that a CVP of 12 to 13 mm Hg correlates with an adequate ventricular volume, and mild global impairment of left ventricular function was also found. Throughout the night, the patient’s CVP and blood pressure dropped in parallel; fluid administration was guided by the earlier echocardiographic demonstration of euvolemia at a CVP of 12 to 13 mm Hg. If the blood pressure drop was not accompanied by a change in CVP, most clinicians would worry about worsening ventricular function and repeat a caval oxygen saturation, and if it was grossly abnormal, repeat the echocardiogram.

In this example, the CVP line provided a general target for fluid resuscitation that was crosschecked with 2D echocardiogram; this approach prevented misassumptions regarding ventricular filling in a patient with likely altered ventricular compliance. The echocardiogram also provided some reassurance that ventricular function was adequate and that an inotrope was not needed. As the patient was not intubated, the use of TEE would have been difficult and time consuming in this case. Spontaneous ventilation made volume response estimates from systolic pressure variation less reliable, but the arterial line provided instant confirmation of volume responsiveness upon raising the legs. Esophageal Doppler would have required frequent repositioning and focusing of the probe in this postoperative patient, but it would have been quite useful to estimate fluid needs and prevent overinfusion of fluids during his operation.

Case 2
For another example, consider a recently admitted cardiac surgery patient in the ICU who is found to be hypotensive. Examination of data trends shows a decline in MAP over the previous 30 min from the mid 80s to 58. This time frame also demonstrated concomitant declines in CVPs (from 14 mm Hg to 7 mm Hg), pulmonary artery diastolic pressure (from 18 mm Hg to 9 mm Hg), urine output, and CI (1.7 from 2.3). All signs point to a decline in CO from hypovolemia with subsequent decrease in blood pressure. Likely sources are interstitial fluid shifts, bleeding, and vasodilation and redistribution. PLR is likely to confirm the diagnosis, but regardless of result, fluid should be given to this patient. Normalization of blood pressure is very likely to accompany restoration of pulmonary artery diastolic and CVPs to euvolemic values. If, in a similar patient, the same declines in blood pressure, CO, and urine output were accompanied by increases in central venous and pulmonary artery pressures, thoughts would focus on diagnoses of cardiac tamponade, myocardial ischemia, or a cardiopulmonary interaction, such as chest wall rigidity, dynamic hyperinflation, or pneumothorax. A rapid clinical assessment with possible chest radiography or TEE would provide the most rapid means of sorting out these diagnoses and establishing definitive therapy.


Changes in solid organ function in critical illness are common, but this is especially true for the cardiovascular system, where drastic changes in cardiac filling and vascular tone can occur in minutes. Thus, one needs to maintain flexibility in the perception of illness and cardiovascular function and use multiple data sources to confirm and revise such perceptions. While the perfect monitor for fluid status awaits development, careful use of current monitoring systems—especially in combination—is sufficient to answer the key questions regarding fluid status. As with most decisions related to diagnosis and therapy, the quality of the thoughts and questions related to fluid administration are of much greater value than the actual tools used to answer the questions.


  1. Michard F, Teboul JL. Predicting fluid responsiveness in ICU patients: a critical analysis of the evidence. Chest. 2002;121(6):2000-2008.
  2. Hughes RE, Magovern GJ. The relationship between right atrial pressure and blood volume. AMA Arch Surg. 1959;79(2):238-243.
  3. Boldt J, Lenz M, Kumle B, Papsdorf M. Volume replacement strategies on intensive care units: results from a postal survey. Intensive Care Med. 1998;24(2):147-151.
  4. Magder S. How to use central venous pressure measurements. Curr Opin Crit Care. 2005;11(3):264-270.
  5. Magder S, Bafaqeeh F. The clinical role of central venous pressure measurements. J Intensive Care Med. 2007;22(1):44-51.
  6. 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.
  7. Desmond J, Megahed M. Is the central venous pressure reading equally reliable if the central line is inserted via the femoral vein. Emerg Med J. 2003;20(5):467-469.
  8. Dellinger RP, Carlet JM, Masur H, et al. Surviving Sepsis Campaign guidelines for management of severe sepsis and septic shock. Crit Care Med. 2004;32(6):858-873.
  9. Swan HJ, Ganz W, Forrester J, Marcus H, Diamond G, Chonett D. Catheterization of the heart in man with use of a flow-directed balloon-tipped catheter. N Engl J Med. 1970;283(9):447-451.
  10. Shah MR, Hasselbad V, Stevenson LW, et al. Impact of the pulmonary artery catheter in critically ill patients: meta-analysis of randomized clinical trials. JAMA. 2005;294(13):1664-1670.
  11. Johnston WE. Complications from the use of pulmonary artery catheters. In: Lumb PD, Bryan-Brown CW, eds. Complications in Critical Care Medicine. Chicago, IL: Year Book Medical Publishers; 1988:45-46.
  12. Jain M, Canham M, Upadhyay D, Corbridge T. Variability in interventions with pulmonary artery catheter data. Intensive Care Med. 2003;29(11):2059-2062.
  13. Michard F, Teboul JL. Predicting fluid responsiveness in ICU patients: a critical analysis of the evidence. Chest. 2002;121(6):2000-2008.
  14. Palazzo M. Circulating volume and clinical assessment of the circulation. Br J Anaesth. 2001;86(6):743-746.
  15. Coyle JP, Teplick RS, Long MC, Davidson JK. Respiratory variations in systemic arterial pressure as an indicator of volume status [Abstract]. Anesthesiology. 1983;59:A53.
  16. Pizov R, Ya'ari Y, Perel A. The arterial pressure waveform during acute ventricular failure and synchronized external chest compression. Anesth Analg. 1989;68(2):150-156.
  17. Szold A, Pizov R, Segal E, Perel A. The effect of tidal volume and intravascular volume state on systolic pressure variation in ventilated dogs. Intensive Care Med. 1989;15(6):368-371.
  18. Rooke GA, Schwid HA, Shapira Y. The effect of graded hemorrhage and intravascular volume replacement on systolic pressure variation in humans during mechanical and spontaneous ventilation. Anesth Analg.1995;80(5):925-932.
  19. Tavernier B, Makhotine O, Lebuffe G, Dupont J, Scherpereel P. Systolic pressure variation as a guide to fluid therapy in patients with sepsis-induced hypotension. Anesthesiology.1998;89(6):1313-1321.
  20. Perel A, Minkovich L, Priesman S, Abiad M, Segal E, Coriat P. Assessing fluid-responsiveness by a standardized ventilatory maneuver: the respiratory systolic variation test. Anesth Analg. 2005;100(4):942-945.
  21. Perel A, Minkovich L, Abiad M, Coriat P, Viars P. Respiratory systolic variation test: a new method for assessing preload [abstract]. Br J Anaesth.1995;74:A137.
  22. Michard F, Boussat S, Chemla D, et al. Relation between respiratory changes in arterial pulse pressure and fluid responsiveness in septic patients with acute circulatory failure. Am J Respir Crit Care Med. 2000;162(1):134-138.
  23. Reuter DA, Felbinger TW, Kilger E, Schmidt C, Lamm P, Goetz AE. Optimizing fluid therapy in mechanically ventilated patients after cardiac surgery by on-line monitoring of left ventricular stroke volume variations: comparison with aortic systolic pressure variations. Br J Anaesth. 2002;88(1):124-126.
  24. 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.
  25. Nagdev AD, Merchant RC, Tirado-Gonzalez A, Sisson CA, Murphy MC. 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.
  26. Feissel M, Michard F, Faller JP, Teboul JL. The respiratory variation in inferior vena cava diameter as a guide to fluid therapy. Intensive Care Med. 2004;30(9):1834-1837.
  27. Barbier C, Loubieres Y, Schmit C, et al. Respiratory changes in inferior vena cava diameter are helpful in predicting fluid responsiveness in ventilated septic patients. Intensive Care Med. 2004;30(9):1740-1746.
  28. Thys DM, Brooker RF, Cahalan MK, et al. Practice Guidelines for Perioperative Transesophageal Echocardiography: an updated report by the American Society of Anesthesiologists and the Society of Cardiovascular Anesthesiologists Task Force on Transesophageal Echocardiography. Anesthesiology. 2010;112(5):1084-1096.
  29. Vieillard-Baron A, Augarde R, Prin S, Page B, Beauchet A, Jardin F. Influence of superior vena caval zone condition on cyclic changes in right ventricular outflow during respiratory support. Anesthesiology. 2001;95(5):1083-1088.
  30. Vieillard-Baron A, Chergin K, Rabiller A, et al. Superior vena caval collapsibility as a gauge of volume status in ventilated septic patients. Intensive Care Med. 2004;30(9):1734-1739.
  31. Singer M. Esophageal Doppler monitoring of aortic blood flow: beat-by-beat cardiac output monitoring. Int Anesthesiol Clin. 1993;31(3):99-125.
  32. Singer M, Allen MJ, Webb AR, Bennett ED. Effects of alterations in left ventricular filling, contractility, and systemic vascular resistance on the ascending aortic blood velocity waveform of normal subjects. Crit Care Med. 1991;19(9):1138-1145.
  33. Monnet X, Rienzo M, Osman D, et al. Esophageal Doppler monitoring predicts fluid responsiveness in critically ill ventilated patients. Intensive Care Med. 2005;31(9):1195-1201.
  34. Gan TJ, Soppitt A, Mariot M, et al. Goal-directed intraoperative fluid administration reduces length of hospital stay after major surgery. Anesthesiology. 2002;57(9):845-849.
  35. Conway DH, Mayall R, Abdul-Latif MS, Gilligan S, Tackaberry C. Randomised controlled trial investigating the influence of intravenous fluid titration using oesophageal Doppler monitoring during bowel surgery. Anaesthesia. 2002;97(4):820-826.
  36. Schober P, Loer SA, Schwarte LA. Perioperative hemodynamic monitoring with transesophageal Doppler technology. Anesth Analg. 2009;109(2):340-353.
  37. Jabot J, Teboul JL, Richard C, Monnet X. Passive leg raising for predicting fluid responsiveness: importance of the postural change. Intensive Care Med. 2009;35(1):85-90.
  38. Lamia B, Ochagavia A, Monnet X, Chemla D, Richard C, Teboul JL. Echocardiographic prediction of volume responsiveness in critically ill patients with spontaneously breathing activity. Intensive Care Med. 2007;33(7):1125-1132.
  39. Monnet X, Rienzo M, Osman D, et al. Passive leg raising predicts fluid responsiveness in the critically ill. Crit Care Med. 2006;34(5):1402-1407.
  40. Preau S, Saulnier F, Dewavrin F, Durocher A, Chagnon JL. 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.