Hemodynamic assessment of critically ill patients
using echocardiography Doppler
Philippe Vignon
Purpose of review
The evaluation of hemodynamic status in critically ill
patients is a leading recommended indication of
transesophageal echocardiography in the intensive care
unit. Advantages and diagnostic yield of transesophageal
echocardiography in this setting are particularly relevant
when considering limitations and questioned prognostic
impact of pulmonary artery catheterization.
Recent findings
Recent clinical studies have been performed to validate and
assess the value of transesophageal echocardiography in
determining cardiac output, cardiac preload dependence,
right ventricular function, and left ventricular filling pressure.
In addition, diagnostic capacity and therapeutic impact of
transesophageal echocardiography have been widely
reported in various intensive care unit settings.
Summary
Transesophageal echocardiography appears well suited for
the determination of cardiac index and to track its variations
after therapeutic interventions. Although repeated
measurements of left ventricular end-diastolic dimension allows
to accurately track preload variations, a single determination is
not reliable to predict fluid responsiveness in intensive care unit
patients. Identification of preload dependence in
hemodynamically unstable patients currently tends to rely
mainly on dynamic parameters that use cardiopulmonary
interactions under mechanical ventilation. Transesophageal
echocardiography also allows to adequately assess right
ventricular function and left ventricular filling pressure using
combined Doppler modalities. Adequate education and training
of intensivists and anesthesiologists is crucial to further develop
the use of transesophageal echocardiography in the intensive
care unit setting. Despite the absence of randomized controlled
studies documenting transesophageal echocardiography
benefits on patient outcome, present evidence and experience
strongly recommend a larger use of echocardiography Doppler
for a comprehensive functional hemodynamic assessment of
critically ill patients with circulatory failure.
Keywords
critical care, diagnostic techniques and procedures,
echocardiography, echocardiography Doppler, therapeutics
—234. ª 2005 Lippincott Williams & Wilkins.
Curr Opin Crit Care 11:227—
Medical-surgical intensive care unit, Dupuytren Teaching Hospital, Limoges, France
Correspondence to Pr Philippe Vignon, Service de Réanimation Polyvalente, CHU
Dupuytren, 2 Ave. Martin Luther King, 87042 Limoges Cedex, France
Tel: +33 5 55 05 62 40; fax: +33 5 55 05 62 44; e-mail: vignon@unilim.fr
Current Opinion in Critical Care 2005, 11:227—
—234
ª 2005 Lippincott Williams & Wilkins.
1070-5295
Introduction
The treatment of critically ill patients frequently requires
a comprehensive evaluation of the hemodynamic status.
Complex monitoring systems are widely available and of-
ten rely on invasive techniques such as the pulmonary ar-
tery catheter (PAC). Hemodynamic monitoring using PAC
has not yet been shown to improve survival [1] and has
substantial limitations [2]. Echocardiography is currently
the only method that can provide real-time imaging of
the heart at bedside. As such, it appears as an alternative
modality that is ideally suited for the functional hemody-
namic assessment of patients with circulatory failure in
the intensive care unit (ICU). Nevertheless, recent rec-
ommendations for hemodynamic support of septic patients
promulgate PAC as a first-line modality for the evaluation
of hemodynamics in this setting, echocardiography being
‘useful to assess ventricular volumes and cardiac perfor-
mance’ [3].
This review will briefly discuss the current field of use and
advantages of transesophageal echocardiography (TEE)
in ICU patients with hemodynamic compromise. It will
mainly focus on clinical studies that showed practical ap-
plications of TEE in hemodynamically unstable patients
and will purposely exclude cardiac output monitoring by
esophageal Doppler that has been recently reviewed
[4,5•].
Modalities of echocardiography Doppler
Although promising results have been recently reported
using transthoracic echocardiography (TTE) for the iden-
tification of cardiac source of shock in a general ICU
population [6•], diagnostic accuracy of TEE has long been
shown to be superior, especially in ventilated patients [7].
Similarly, at its present stage of development, recently
available hand-held echocardiography is not adequately
suited for a comprehensive hemodynamic assessment in
critically ill patients [8,9••]. Hemodynamic instability is
therefore one of the main class I indications of the Amer-
ican Society of Anesthesiologists for the use of TEE in the
ICU setting [10].
Cardiac index
Serial determinations of cardiac index are widely used
in the management of patients with circulatory failure.
Echocardiography allows the measurement of cardiac out-
put using two-dimensional imaging (measurement of ven-
tricular volumes) or Doppler-based methods.
227
228 Cardiopulmonary monitoring
Two-dimensional echocardiography usually underesti-
mates volumes of the left ventricle (LV) when compared
with reference imaging techniques such as contrast or ra-
dionuclide ventriculography, mostly because LV long axis
is foreshortened. This has been recently confirmed in ICU
patients after coronary artery bypass surgery [11]. Never-
theless, TEE determination of LV stroke volume and ejec-
tion fraction remained accurate (mean bias <7%) since
both the end-diastolic and end-systolic LV volumes were
underestimated.
Combining the cross-sectional area and the stroke dis-
tance traveled by the column of blood at a given anatomic
site also allows to determine cardiac output. Most clinical
trials support TEE as a reliable alternative method for
measuring cardiac output and tracking its variations after
therapeutic interventions. In the absence of valvulopathy,
Doppler measurement of cardiac output at the level of the
aortic valve or LV outflow tract appears to provide a lower
failure rate and higher accuracy when compared with other
sites such as the mitral valve or pulmonary artery [12]. For
example, Perrino et al. [13] showed that intraoperative
measurement of cardiac output at the level of the aortic
valve was reliable. Mean bias with thermodilution was
ÿ0.01 L/min with an SD of the differences of 0.56 L/min
[13]. Similarly, most validation studies using the thermo-
dilution technique as a gold standard for cardiac output
measurement reported limits of agreement with TEE that
reached ±1 L/min [12]. This fairly wide range of agree-
ment is not surprising when considering the variation in-
trinsic to cardiac output determination by both methods
and potential inaccuracy of thermodilution [2]. Balik
et al. [14] recently confirmed that the severity of tricuspid
regurgitation reduced the agreement between thermodilu-
tion and TEE for cardiac output determination. In the
presence of a moderate-to-severe tricuspid regurgitation
— an abnormality frequently observed in ventilated ICU
patients [15] — the thermodilution underestimated cardi-
ac output in 95% of the cases, with a significant bias of
ÿ1.9 ± 2.3 L/min [14]. Similarly, Zink et al. [16•] showed
that the agreement between TEE and fast response
thermistor PAC for the measurement of ejection fraction
of the right ventricle (RV) significantly decreased with
pulmonary hypertension that is usually associated with tri-
cuspid regurgitation.
In visualizing cardiac structures, TEE has also the major
advantage over PAC of providing direct appreciation of
the size and function of both ventricles. Thus, the oper-
ator can readily appreciate in real time if the stroke
volume is mostly generated by a preserved myocardial
contractility or by an enlarged, yet severely depressed,
ventricle and if the low output state is secondary to an
RV systolic dysfunction rather LV failure (Fig. 1). Al-
though TEE is not yet adapted for continuous moni-
toring of cardiac output in critically ill patients, it allows
repeated assessment of the efficacy and tolerance of ther-
apeutic interventions (e.g., fluid challenge, vasopressors)
and promising automated methods of cardiac output mea-
surement are currently emerging [17•].
Preload assessment
Preload is defined as the myocardial fiber length at end-
diastole. At the organ level, preload may theoretically be
assessed by measuring LV end-diastolic volume. In a study
using thermodilution and radionuclide ventriculography
as ‘gold standard’, TEE consistently appeared to underes-
timate LV end-diastolic volume [11]. An alternative expla-
nation would be an overestimation of cardiac output by
thermodilution in low-flow states that led to overestimate
calculated LV volumes. Indeed, the early work of Parker
et al. [18] that used the same combination of techniques
to suggest that ventricular dilatation during septic shock
could be a preload adaptation in survivors was not con-
firmed by a recent TEE study in which no LV dilatation
was found at the early phase of septic shock [19].
In animal and clinical studies assessing the ability of TEE
to detect hypovolemia induced by a graded hemorrhage,
LV end-diastolic area — used as a surrogate of end-diastolic
volume — has been shown to closely correlate with blood
loss or volume expansion [20]. In addition, variations in
ventricular end-diastolic volume and stroke volume after
a fluid infusion are closely correlated in normal volunteers,
whereas no relation exists between ventricular filling pres-
sure and stroke volume variations [21••]. Accordingly, TEE
appears to adequately track changes in LV end-diastolic di-
mension throughout therapeutic interventions.
In contrast, no threshold value of LV end-diastolic area can
be used to predict fluid responsiveness in most ICU
patients [22–25•]. First, for a given transmural filling pres-
sure, end-diastolic dimension depends on LV compliance
that may greatly vary among critically ill patients. Second,
relative LV compliance may abruptly change in ICU
patients (e.g., acute coronary disease or silent myocardial
ischemia, acute RV overload, increased pericardial pres-
sure). Third, cardiac diseases are frequently associated
with structural abnormalities that invalidate the use of
LV end-diastolic dimensions as a reliable index of preload
(e.g., dilated cardiomyopathy or reduced venticular cavity
associated with LV hypertrophy). Finally, cardiovascular
effects of general anesthesia may result in smaller LV
end-diastolic areas than expected in the general popu-
lation, with a quite wide range of normal values [26].
Accordingly, low LV end-diastolic dimension does not
imply necessarily preload dependence, and even less in-
dication for fluid therapy. Dynamic indices using cardio-
pulmonary interactions in ventilated patients have there-
fore been proposed as alternative markers of preload
dependence.
 Hemodynamic assessment using echocardiography Doppler Vignon 229

Figure 1. Potential advantages of echocardiography Doppler over pulmonary artery catheterization for the determination
of cardiac index

(a) In contrast to invasive hemodynamic monitoring, transesophageal echocardiography has the unparalleled advantage of assessing both the pump
function and cavity size of the left ventricle (black boxes). Accordingly, it determines which of these two parameters predominantly generates left
ventricular stroke volume. Adapted from Braunwald E. Regulation of the circulation. N Engl J Med 1974; 200:1124. (b) Illustrative example of discrepant
results provided by the pulmonary artery catheter and transesophageal echocardiography in a patient placed under ventilator for secondary pulmonary
edema associated with severe systemic hypoxemia. Under continuous infusion of dobutamine, two-dimensional echocardiography revealed the presence
of an underlying dilated cardiomyopathy with depressed pump function (left) while Doppler measurement confirmed a markedly decreased left ventricular
stroke volume (right). In this patient, cardiac index was overestimated by the thermodilution technique (see text for details).

Prediction of preload dependence
Pulse pressure variation under volume-controlled ventila-
tion has been successfully used to predict the response to
a fluid challenge in septic patients [27]. Since the pulse
pressure variation concept is mainly based on respiratory
change in LV stroke volume, Feissel et al. [24] hypothe-
sized that variation of aortic blood flow velocity measured
by Doppler could predict preload responsiveness in septic
ventilated patients. Variation of aortic Doppler velocities
was calculated as the ratio of the difference between
maximal (inspiratory) and minimal (expiratory) velocities
to the mean of these two velocities and responders were
defined as patients increasing their cardiac index of at
least 15% after a fluid challenge, as previously proposed
for pulse pressure variation [27]. Not surprisingly, the
authors [24] showed that a respiratory variation of aortic
blood flow velocity >12% when measured using TEE
pulsed Doppler at the level of the LV outflow tract accu-
rately identified responders with a positive predicting
value of 91%, while a threshold value of 13% was proposed
for pulse pressure variation [27]. As importantly, respira-
tory variation in aortic Doppler velocities #12% had
a 100% negative predictive value, thus avoiding inefficient
and potentially deleterious volume expansion. In addition,
the magnitude of respiratory variation of aortic Doppler
velocities was closely related to the increase of cardiac in-
dex induced by fluid therapy [24].
Since cardiac output must equal the volume of blood en-
tering the heart, the cyclic effect of tidal ventilation on
venous return appears also useful to detect preload depen-
dence [28]. In 66 ventilated septic patients, Vieillard-
Baron et al. [29••] showed that a collapsibility index
(maximal diameter on expiration ÿ minimal diameter on
inspiration/maximal diameter on expiration expressed as
a percentage) of the superior vena cava >36% allowed
to accurately distinguish responders from non responders
to a fluid challenge with a sensitivity and specificity of
230 Cardiopulmonary monitoring
90% and 87%, respectively. Diameter variation of the supe-
rior vena cava had a bimodal distribution, most responders
exhibiting a collapsibility index >50% while all but two
non responders had a collapsibility index <30% [29••].
Similarly, respiratory variation of the inferior vena cava di-
ameter has been successfully used to detect preload de-
pendence in septic patients examined with TTE
[30•,31•].
All limitations of pulse pressure variation as an indicator
of fluid responsiveness [32] also apply when using ‘dy-
namic’ echocardiographic markers of preload dependence
[28]. In addition, the proposed threshold values have
been determined retrospectively in septic patients, but
not yet tested prospectively in larger cohorts of patients
and in other conditions than sepsis. Importantly, preload
dependence does not necessarily imply need of volume
expansion, the indication of fluid therapy in a critically
ill patient remaining based on the clinical evaluation [28].
Identification of right ventricular dysfunction
Early diagnosis of RV dysfunction is crucial for adequate
treatment of ventilated ICU patients. First, RV failure
may result in increased pulse pressure variation due to
the cyclic rise in outflow impedance during the insuffla-
tion of the tidal volume [29••]. In this case, pulse pressure
variation does not reflect a marked inspiratory drop in ve-
nous return due to hypovolemia, but rather an RV systolic
dysfunction exacerbated by positive pressure ventilation.
TEE usually confirms RV dysfunction associated with in-
creased venous pressure, as reflected by reduced or abol-
ished respiratory variation of vena cava diameter. This
scenario qualifies the patient as a non-responder to vol-
ume expansion since there is no LV fluid responsiveness
without RV preload dependence. Second, echocardiogra-
phy allows immediate diagnosis of acute cor pulmonale
in hypotensive patients with suddenly afterloaded RV
secondary to massive pulmonary embolism or acute respi-
ratory distress syndrome (ARDS) [33]. Diagnostic algo-
rithm is dramatically simplified and echocardiography
provides insights in RV functional impairment that may
guide patient treatment. Third, in patients with ARDS re-
quiring aggressive respiratory support, TEE may reveal
deleterious effects of positive-pressure ventilation on as-
sociated RV failure and may potentially help in guiding
ventilatory strategy [34]. Fourth, RV dysfunction has been
commonly reported in various settings, such as sepsis
[29••] or ischemic cardiac disease [35], and may be asso-
ciated with a poor outcome [35]. Finally, echocardiography
allows a comprehensive functional assessment of both RV
systolic and diastolic function [36], which is frequently
impaired in ventilated ICU patients.
Left ventricular filling pressure
Since diastolic LV pressure-volume relation is not linear,
there is no more reason to use Doppler assessment of
LV filling pressure than measurement of pulmonary artery
occlusion pressure (PAOP) as an index of preload. Accord-
ingly, the main indication for evaluating LV filling pressure
with Doppler is to differentiate secondary pulmonary
edema from ARDS.
Both mitral and pulmonary venous Doppler profiles have
been successfully used to assess PAOP in ICU patients
[37]. Nevertheless, these conventional Doppler indices
can be altered by numerous additional parameters, such
as LV diastolic properties that are better characterized
by tissue or color Doppler indices. Accordingly, animal
and clinical studies have recently shown that the com-
bined use of tissue or color Doppler and conventional
Doppler indices allowed predicting PAOP non invasively.
In 60 ventilated patients monitored with a PAC for post-
operative shock and acute lung injury, Bouhemad et al.
[38] showed that an E/Ea ratio $7 (E: maximal mitral
Doppler velocity at early diastole; Ea: maximal tissue
Doppler velocity of the lateral mitral annulus at early di-
astole) predicted a PAOP $13 mm Hg with a sensitivity of
86% and a specificity of 92%. This ratio appeared less ac-
curate for predicting PAOP #8 mm Hg. The ratio E/Vp
(Vp: early diastolic propagation velocity measured by color
M-mode) was a less accurate predictor of LV filling pres-
sure in this study, with fairly large limits of agreement of
ÿ6 to +6 mm Hg [38]. Similarly, Combes et al. [39•]
reported in 23 ventilated ICU patients that an E/Ea ra-
tio >7.5 estimated a PAOP $15 mm Hg, with a sensitivity
and a specificity of 86% and 81%, respectively. Limits of
agreement were also fairly large (±5 mm Hg). Importantly,
several studies suggest that Doppler evaluation of LV fill-
ing pressure is more accurate in the presence of LV systolic
dysfunction [38,40••]. The concept of combining new
Doppler indices — considered to be preload independent
— with conventional pulsed Doppler parameters to pre-
dict more accurately LV filling pressure has been recently
challenged. Indeed, animal and clinical studies have shown
that both Ea velocity and Vp were in fact significantly al-
tered by a marked and rapid preload reduction, such as ul-
trafiltration [41,42•,43••]. Finally, no clinical study has yet
been focused on TEE prediction of a PAOP #18 mm Hg,
to obtain the hemodynamic criterion of ARDS without
PAC [44].
Therapeutic impact
Overall therapeutic impact of TEE is substantial in all
clinical studies involving ICU populations since it may
reach 60% of the cases [45••], especially in hemodynam-
ically unstable patients [46]. Although changes in medical
therapy is the most frequent, cardiovascular surgery may
be indicated without further work-up based on TEE find-
ings [7,45••,47–49]. Increasing use of TEE in an institu-
tion does not appear to reduce its yield in ICU patients
[50]. The rate of discrepancies between TEE and PAC
is usually high, mainly due to limitations of invasive
 Hemodynamic assessment using echocardiography Doppler Vignon 231

Figure 2. Examples of hemodynamic disturbances frequently overlooked by pulmonary artery catheterization but accurately
identified when using echocardiography Doppler in hemodynamically unstable patients

(a) Excessive use of inotropic drugs in this hypotensive patient resulted in a functional left ventricular outflow tract obstruction (maximal pressure
gradient of ;90 mm Hg; arrows) due to the combination of non corrected hypovolemia and underlying septal hypertrophy. (b) Acute cor pulmonale
observed in a ventilated patient with unstable hemodynamics and marked respiratory variation of arterial pressure. In this case, the insufflation of tidal
volume revealed a severe right ventricular systolic dysfunction rather than preload dependence of the heart (see text for details). (c) In this hypertensive
patient who was placed under ventilator for secondary pulmonary edema, transesophageal echocardiography depicted preserved left ventricular
systolic function together with elevated filling pressures. Pure diastolic dysfunction was diagnosed based on the presence of a marked concentric left
ventricular hypertrophy together with an abnormal tissue Doppler diastolic profile and decreased propagation velocity measured by color M-mode.
Please note the reduced left ventricular end-systolic volume (20 mL). (d) In this septic patient presenting with a pulmonary edema and elevated cardiac
output, transesophageal echocardiography revealed a massive aortic regurgitation (arrows) secondary to an acute valvular endocarditis that required
prompt surgical repair. (e) In this hypotensive patient ventilated for a penetrating trauma, transesophageal echocardiography depicted a defect of the
interventricular septal wall that was responsible for a substantial left-to-right shunt (arrow). (f) In this patient who sustained severe hypotension with
elevated central venous pressure shortly after cardiac surgery, transesophageal echocardiography immediately identified a compression of right
cardiac cavities by a large hematoma (double-headed arrow). Please note that right ventricular cavity became virtual (arrow) due to extrinsic
compression that resulted in tamponade. Hemodynamics rapidly improved after immediate surgery. RA, right atrium; LA, left atrium; RV, right ventricle;
LV, left ventricle.

monitoring [50–53]. Importantly, TEE frequently depicts
abnormalities that are consistently overlooked by PAC,
such as LV outflow tract obstruction, ventricular interde-
pendency, LV diastolic dysfunction, acute valvulopathy,
acquired anatomic shunt, or localized cardiac compression
(Fig. 2). Accordingly, TEE frequently allows modifying in-
adequate PAC-guided therapy [50–53].
Prognostic impact
It is currently unclear as to whether acute management
guided by repeated assessment of hemodynamic status
improves survival in critically ill adults. Few studies have
evaluated the potential impact of TEE on the outcome of
ICU patients examined for unstable hemodynamics.
Schmidlin et al. [47] reported that patients evaluated with
TEE after cardiac surgery had a significantly worse out-
come than those who were not. However, in the absence
of randomization, hypotensive patients assessed with ul-
trasound were at higher risk of poor outcome, the use
of TEE only reflecting the presence of postoperative com-
plications as opposed to the unmatched control group.
Heidenreich et al. [54] reported that TEE could help pre-
dict the mortality in critically ill patients according to the
identified mechanism of hypotension. Regrettably, the
impact of TEE on patients’ outcome could not be evalu-
ated in the absence of randomization. Of note, TEE-
induced morbidity remains fairly low since complication
rates in the ICU setting ranges between 1 and 5% [45••],
and is even lower in ventilated patients [7].
Advantages, cost effectiveness, and impact on outcome of
TEE when used as a first-line modality for the assessment
232 Cardiopulmonary monitoring
of hemodynamically unstable ICU patients have not yet
been documented in prospective, randomized, controlled
trials. However, ultrasound has become a standard of care
in many institutions and depriving hypotensive patients of
a TEE examination could be potentially harmful and un-
ethical, when considering its potential benefit and thera-
peutic impact in the ICU settings.
Educational issues
Echocardiography is an operator-dependent technique that
requires immediate and around-the-clock availability of
experienced examiners to obtain and interpret adequate
images. Insufficient training may lead to harmful assump-
tions, diagnoses, and therapy, especially in the setting of
unstable ICU patients. The influence of the learning curve
on TEE diagnostic accuracy has been previously demon-
strated [55]. Currently, the access of intensivists and anes-
thesiologists to a formal education and practical training
is in progress in many countries [56–58]. Several studies
have showed that adequately trained intensivists may
safely perform TEE examinations that frequently alter pa-
tient treatment, provided that supervision by an experi-
enced operator is available [7,46–48].
Conclusion
Ease of use, versatility, and instantaneous diagnostic capa-
bility of echocardiography allow accurate and expeditious
assessment of the hemodynamic status in critically ill
patients. Although TEE is not currently adapted for a con-
tinuous monitoring, it appears ideally suited for a compre-
hensive functional hemodynamic evaluation that can be
repeated after therapeutic interventions in ICU patients.
When performed by adequately trained intensivists, TEE
has a relevant therapeutic impact in a large proportion of
hemodynamically unstable patients. One crucially lacks
large scale randomized controlled studies to document
the superiority of TEE over other methods on patient
treatment and outcome. Nevertheless, present evidence
and experience strongly recommend a larger availability
of equipment and expertise in ICUs for further use of
echocardiography in critically ill patients with hemody-
namic compromise.
References and recommended reading
Papers of particular interest, published within the annual period of review, have
been highlighted as:
• of special interest
•• of outstanding interest
1 Richard C, Warszawski J, Anguel N, et al. Early use of the PAC and outcomes
in patients with shock and acute respiratory distress syndrome. A randomized
controlled trial. JAMA 2003; 290:2713— —2720.
2 Jardin F, Bourdarias JP. Right heart catheterization at bedside: a critical view.
Intensive Care Med 1995; 21:291— —295.
3 Hollenberg SM, Ahrens TS, Annane D, et al. Practice parameters for hemo-
dynamic support of sepsis in adult patients: 2004 update. Crit Care Med
2004; 32:1928——1948.
4 Cholley BP, Singer M. Esophageal Doppler: noninvasive cardiac output mon-
itor. Echocardiography 2003; 20:763—
—769.
5 Dark PM, Singer M. The validity of transesophageal Doppler ultrasonography
• as a measure of cardiac output in critically ill adults. Intensive Care Med
2004; 30:2060— —2066.
An exhaustive review of validation studies evaluating the accuracy of esophageal
Doppler to monitor changes in cardiac output in the critically ill.
6 Joseph MX, Disney PJS, Da Costa R, et al. Transthoracic echocardiography
• to identify or exclude cardiac cause of shock. Chest 2004; 126:1592— —1597.
This clinical study performed in ICU patients evaluated the diagnostic ability of
transthoracic echocardiography to identify the cardiac source of shock when us-
ing a full-feature platform. A high rate of adequate imaging quality was obtained,
resulting in a remarkable diagnostic accuracy of TTE.
7 Vignon P, Mentec H, Terré S, et al. Diagnostic accuracy and therapeutic im-
pact of transthoracic and transesophageal echocardiography in mechanically
ventilated patients in the ICU. Chest 1994; 106:1829— —1834.
8 Vignon P, Chastagner C, Francxois B, et al. Diagnostic ability of hand-held echo-
cardiography in ventilated critically-ill patients. Crit Care 2003; 7:R84——R91.
9 Vignon P, Frank MBJ, Lesage J, et al. Hand-held echocardiography with
•• Doppler capability for the assessment of critically-ill patients: is it reliable?
Intensive Care Med 2004; 30:718— —723.
First randomized descriptive study evaluating the diagnostic capacity of a portable
ultrasound device with spectral Doppler capability in ICU patients. Experienced
operators were randomized to conduct the ultrasound study with either the por-
table device or an upper-end platform, which was used as reference imaging mo-
dality. Despite spectral Doppler capability, hand-carried echocardiography missed
relevant valvulopathies and yielded lower diagnostic accuracy compared with con-
ventional TTE. Technological limitations of portable devices are discussed.
10 American Society of Anesthesiologists and the Society of Cardiovascular
Anesthesiologists Task Force on Transesophageal Echocardiography. Prac-
tice guidelines for perioperative transesophageal echocardiography. Anes-
thesiology 1996; 84:986— —1006.
11 Ryan T, Burwash I, Lu J, et al. The agreement between ventricular volumes
and ejection fraction by transesophageal echocardiography or a combined
radionuclear and thermodilution technique in patients after coronary artery
surgery. J Cardiothorac Vasc Anesth 1996; 10:323— —328.
12 Poelaert J, Schmidt C, Van Aken H, et al. A comparison of transesophageal
echocardiographic Doppler across the aortic valve and the thermodilution
technique for estimating cardiac output. Anaesthesia 1999; 54:128——136.
13 Perrino AC, Harris SN, Luther MA. Intraoperative determination of cardiac
output using multiplane transesophageal echocardiography. A comparison
to thermodilution. Anesthesiology 1998; 89:350——357.
14 Balik M, Pachl J, Hendl J. Effects of the degree of tricuspid regurgitation on
cardiac output measurements by thermodilution. Intensive Care Med 2002;
28:1117— —1121.
15 Jullien T, Valtier B, Hongnat JM, et al. Incidence of tricuspid regurgitation and
vena cava backward flow in mechanically ventilated patients. A color Doppler
and contrast echocardiographic study. Chest 1995; 107:488— —493.
16 Zink W, Nöll J, Rauch H, et al. Continuous assessment of right ventricular
• ejection fraction: new pulmonary artery catheter versus transesophageal
echocardiography. Anaesthesia 2004; 59:1126— —1132.
Serial intraoperative measurements of right ventricular fractional area change ob-
tained by TEE were compared with right ventricular ejection fraction continuously
determined by fast response thermistor pulmonary artery catheter. Authors
showed that mean bias between both methods significantly increased with elevated
pulmonary artery pressure and discussed potential source of errors of the modified
pulmonary artery catheter, including the role of tricuspid regurgitation.
17 Akamatsu S, Oda A, Terazawa E, et al. Automated cardiac output measure-
• ment by transesophageal color Doppler echocardiography. Anesth Analg
2004; 98:1232— —1238.
This study validated intraoperatively a new color Doppler technique for automated
cardiac output measurement using temporal and spatial integration of color Doppler
velocity profiles over time within a region of interest, when using the thermodilution
technique as a reference method. Authors discussed the accuracy of this new meth-
od according to the anatomical site of measurement and its advantages and limita-
tions compared to conventional echocardiographic determination of cardiac output.
18 Parker M, Shelhamer J, Bacharach S, et al. Profound but reversible myocar-
dial depression in patients with septic shock. Ann Intern Med 1984; 100:
483— —490.
19 Vieillard-Baron A, Schmitt JM, Beauchet A, et al. Early preload adaptation in
septic shock? A transesophageal echocardiography study. Anesthesiology
2001; 94:400— —406.
20 Dalibon N, Schlumberger S, Saada M, et al. Haemodynamic assessment of
hypovolemia under general anesthesia in pigs submitted to graded haemor-
rhage and retransfusion. Br J Anaesth 1999; 82:97—
—103.
 Hemodynamic assessment using echocardiography Doppler Vignon 233
21 Kumar A, Anel R, Bunnell E, et al. Pulmonary artery occlusion pressure and
•• central venous pressure fail to predict ventricular filling volume, cardiac per-
formance, or the response to volume infusion in normal subjects. Crit Care
Med 2004; 32:691— —699.
This study performed in normal volunteers demonstrates that ventricular filling
pressures fail to predict ventricular end-diastolic volumes and the variation in
stroke volume induced by a standardized volume expansion. In contrast, variations
in ventricular end-diastolic volume and stroke volume induced by the volume
expansion were closely correlated when measured using echocardiography
Doppler. Physiological mechanisms potentially responsible for these results are
discussed, as well as clinical implications.
22 Tavernier B, Makhotine O, Lebuffe G, et al. Systolic pressure variation as
a guide to fluid therapy in patients with sepsis-induced hypotension. Anesthe-
siology 1998; 89:1313— —1321.
23 Tousignant CP, Walsh F, Mazer CD. The use of transesophageal echocardi-
ography for preload assessment in critically ill patients. Anesth Analg 2000;
90:351——355.
24 Feissel M, Michard F, Mangin I, et al. Respiratory changes in aortic blood ve-
locity as an indicator of fluid responsiveness in ventilated patients with septic
shock. Chest 2001; 119:867— —873.
25 Rex S, Brose S, Metzelder S, et al. Prediction of fluid responsiveness in pa-
tients during cardiac surgery. Br J Anaesth 2004; 93:782— —788.
In 14 ventilated patients studied immediately after cardiac artery bypass surgery
using both TEE and the transpulmonary thermodilution technique, the authors
showed that stroke volume variation was the only parameter predictive of changes
in stroke volume index induced by tilting the patients from 30
head up to 30
head
down, in contrast to static parameters (invasive measurement of filling pressure
and left ventricular end-diastolic area index).
26 Skarvan K, Lambert A, Filipovic M, et al. Reference values for left ventricular
function in subjects under general anaesthesia and controlled-ventilation
assessed by two-dimensional transoesophageal echocardiography. Eur
J Anaesthesiol 2001; 18:713— —722.
27 Michard F, Boussat S, Chemla D, et al. Relation between respiratory
changes 0in arterial pulse pressure and fluid responsiveness in septic pa-
tients with acute circulatory failure. Am J Respir Crit Care Med 2000;
162:134— —138.
28 Vignon P. Evaluation of fluid responsiveness in ventilated septic patients:
back to venous return [editorial]. Intensive Care Med 2004; 30:1699— —
1701.
29 Vieillard-Baron A, Chergui K, Rabiller A, et al. Superior vena cava collapsibility
•• as a gauge of volume status in ventilated septic patients. Intensive Care Med
2004; 30:1734— —1739.
An important clinical study determining a simple TEE index to indirectly assess
intrathoracic volume status in septic patients under mechanical ventilation. A col-
lapsible superior vena cava during the insufflation of tidal volume was consistent
with a preload dependence of the heart. False positive results of pulse pressure
variation were observed in patients with acute cor pulmonale.
30 Barbier C, Loubières Y, Schmit C, et al. Respiratory changes in inferior vena
•
cava diameter are helpful in predicting fluid responsiveness in ventilated sep-
tic patients. Intensive Care Med 2004; 30:1740— —1746.
Using variation of inferior vena cava diameter under mechanical ventilation as
a marker of venous return, the authors proposed a distensibility index to accurately
predict fluid responsiveness in septic patients. The physiological background is
discussed.
31 Feissel M, Michard F, Faller JP, et al. The respiratory variation in inferior vena
•
cava diameter as a guide to fluid therapy. Intensive Care Med 2004;
30:1834——1837.
Using a similar approach, this study also showed the value of the variation in
inferior vena cava diameter quantified using M-mode transthoracic echo-
cardiography to accurately predict fluid responsiveness in septic ventilated pa-
tients.
32 Pinsky MR. Use of ventilation-induced aortic pressure and flow variation to
diagnose preload responsiveness [editorial]. Intensive Care Med 2004; 30:
1008——1010.
33 Vieillard-Baron A, Prin S, Chergui K, et al. Echo-Doppler demonstration of
acute cor pulmonale at the bedside in the medical intensive care unit. Am
J Respir Crit Care Med 2002; 166:1310— —1319.
34 Vieillard-Baron A, Jardin F. Why protect the right ventricle in patients with
acute respiratory distress syndrome? Curr Opin Crit Care 2003; 9:15— —21.
35 Maslow AD, Regan MM, Panzica P, et al. Precardiopulmonary bypass right
ventricular function is associated with poor outcome after coronary artery by-
pass grafting in patients with severe left ventricular systolic dysfunction.
Anesth Analg 2002; 95:1507— —1518.
36 Vignon P, Weinert L, Mor-Avi V, et al. Quantitative assessment of regional
right ventricular function with Color Kinesis. Am J Respir Crit Care Med
1999; 159:1949— —1959.
37 Boussuges A, Blanc P, Molenat F, et al. Evaluation of left ventricular filling
pressure by transthoracic Doppler echocardiography in the intensive care
unit. Crit Care Med 2002; 30:362——367.
38 Bouhemad B, Nicolas-Robin A, Benois A, et al. Echocardiographic Doppler
assessment of pulmonary capillary wedge pressure in surgical patients with
postoperative circulatory shock and acute lung injury. Anesthesiology 2003;
98:1091——1100.
39 Combes A, Arnoult F, Trouillet JL. Tissue Doppler imaging estimation of pul-
• monary artery occlusion pressure in ICU patients. Intensive Care Med 2004;
30:75——81.
Similar study design in a smaller and more heterogeneous population evaluating
noninvasive estimation of pulmonary artery occlusion pressure with tissue Doppler
imaging of the mitral annulus, using both transthoracic and transesophageal echo-
cardiography.
40 Dokainish H, Zoghbi WA, Lakkis NM, et al. Optimal noninvasive assessment
•• of left ventricular filling pressure. A comparison of tissue Doppler echocardi-
ography and B-type natriuretic peptide in patients with pulmonary artery cath-
eters. Circulation 2004; 109:2432— —2439.
First clinical study to compare the respective diagnostic value of transthoracic tis-
sue Doppler imaging and brain natriuretic peptide level to estimate pulmonary ar-
tery occlusion pressure in spontaneously breathing patients hospitalized in a
coronary care unit. The impact of left ventricular systolic function on the accuracy
of the Doppler estimate of filling pressures is specifically addressed.
41 Ie EHY, Vletter WB, Ten Cate FJ, et al. Preload dependence of new Doppler
techniques limits their utility for left ventricular diastolic function assessment
in hemodialysis patients. J Am Soc Nephrol 2003; 14:1858— —1862.
42 Seo Y, Ishimitsu T, Ishizu T, et al. Preload-dependent variation of the propa-
• gation velocity in patients with congestive heart failure. J Am Soc Echocar-
diogr 2004; 17:432— —438.
In 24 spontaneously breathing patients with congestive heart failure, the authors
showed that the decrease in pulmonary artery occlusion pressure induced by ther-
apy was closely related to a significant drop in early diastolic flow propagation
velocity (Vp) measured using color Doppler M-mode. Significance of Vp and tech-
nical issues regarding its measurement are discussed.
43 Jacques DC, Pinsky MR, Severyn D, et al. Influence of alterations in load-
•• ing on mitral annulus velocity by tissue Doppler echocardiography and its
associated ability to predict filling pressures. Chest 2004; 126:1910— —
1918.
Animal study evaluating the influence of loading conditions and contractility on the
early diastolic velocity of the mitral annulus (Ea) measured by tissue Doppler im-
aging. The authors demonstrated that Ea was significantly influenced by acute var-
iations in left ventricular loading conditions and concluded that an E/Ea ratio was
less predictive of left ventricular filling pressure in the presence of a preserved
systolic function and under-filled ventricle.
44 Bernard GR, Artigas A, Brigham KL, et al. The American-European Consen-
sus Conference on ARDS. Definitions, mechanisms, relevant outcomes,
and clinical trial coordination. Am J Respir Crit Care Med 1994; 149:818—
—
824.
45 Hüttemann E, Schelenz C, Kara F, et al. The use and safety of transesopha-
•• geal echocardiography in the general ICU. A minireview. Acta Anaesthesiol
Scand 2004; 48:827— —836.
A study was performed to assess the use and safety of TEE in a non-cardiac gen-
eral surgical ICU and the pertinent literature was reviewed. Indications and impact
on management according to the type of ICU, as well as safety of TEE are
discussed.
46 Denault AY, Couture P, McKenty S, et al. Perioperative use of transesopha-
geal echocardiography by anesthesiologists: impact in noncardiac surgery
and in the intensive care unit. Can J Anaesth 2002; 49:287——293.
47 Schmidlin D, Schuepbach R, Bernard E, et al. Indications and impact of post-
operative transesophageal echocardiography in cardiac surgical patients.
Crit Care Med 2001; 29:2143— —2148.
48 Colreavy FB, Donovan K, Lee KY, et al. Transesophageal echocardiography in
critically ill patients. Crit Care Med 2002; 30:989—
—996.
49 Wasir H, Mehta Y, Mishra YK, et al. Transesophageal echocardiography in
hypotensive post-coronary bypass patients. Asian Cardiovasc Thorac Ann
2003; 11:139——142.
50 Harris KM, Petrovic O, Davila-Roman VG, et al. Changing patterns of trans-
esophageal echocardiography use in the intensive care unit. Echocardiogra-
phy 1999; 16:559— —565.
51 Benjamin E, Griffin K, Leibowitz AB, et al. Goal-directed transesophageal
echocardiography performed by intensivists to assess left ventricular func-
tion: comparison with pulmonary artery catheterization. J Cardiothorac Vasc
Anesth 1998; 12:10— —15.
52 Costachescu T, Denault A, Guimond JG, et al. The hemodynamically unstable
patient in the intensive care unit: hemodynamic vs. transesophageal echocar-
diographic monitoring. Crit Care Med 2002; 30:1214— —1223.
234 Cardiopulmonary monitoring
53 Wake PJ, Ali M, Carroll J, et al. Clinical and echocardiographic diagnoses dis-
agree in patients with unexplained hemodynamic instability after surgery. Can
J Anaesth 2001; 48:778— —783.
54 Heidenreich PA, Stainback RF, Redberg RF, et al. Transesophageal echocar-
diography predicts mortality in critically ill patients with unexplained hypoten-
sion. J Am Coll Cardiol 1995; 26:152— —158.
55 Vignon P, Spencer KT, Rambaud G, et al. Differential transesophageal echo-
cardiographic diagnosis between linear artifacts and intraluminal flap of aortic
dissection or disruption. Chest 2001; 119:1778— —1790.
56 Swanevelder J, Chin D, Kneeshaw J, et al. Accreditation in transoesophageal
echocardiography: statement from the Association of Cardiothoracic Anaes-
thetists and the British Society of Echocardiography Joint TOE Accreditation
Committee [editorial]. Br J Anaesth 2003; 91:469— —472.
57 Cahalan MK, Abel M, Goldman M, et al. American Society of Echocardiography
and Society of Cardiovascular Anesthesiologists task force guidelines for train-
ing in perioperative echocardiography. Anesth Analg 2002; 94:1384— —1388.
58 Vignon P, Guéret P. Formation des réanimateurs à l’échographie cardiaque.
Réanimation 2004; 13:126— —130.