See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/302446671

Less invasive hemodynamic monitoring in critically ill patients

Article  in  Intensive Care Medicine · May 2016

DOI: 10.1007/s00134-016-4375-7

CITATIONS READS

108 6,415

13 authors, including:

Maurizio Cecconi Daniel De Backer

Humanitas University Université Libre de Bruxelles
284 PUBLICATIONS   9,125 CITATIONS    613 PUBLICATIONS   26,414 CITATIONS   

SEE PROFILE SEE PROFILE

Christoph K Hofer
Triemli City Hospital
249 PUBLICATIONS   5,759 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

I.S.O.S. View project

Fluid therapy View project

All content following this page was uploaded by Thomas W L Scheeren on 08 October 2017.

The user has requested enhancement of the downloaded file.

Intensive Care Med
DOI 10.1007/s00134-016-4375-7

CONFERENCE REPORTS AND EXPERT PANEL

Less invasive hemodynamic monitoring

in critically ill patients
Jean-Louis Teboul1*, Bernd Saugel2, Maurizio Cecconi3, Daniel De Backer4, Christoph K. Hofer5, Xavier Monnet1,
Azriel Perel6, Michael R. Pinsky7, Daniel A. Reuter2, Andrew Rhodes3, Pierre Squara8, Jean-Louis Vincent9
and Thomas W. Scheeren10

© 2016 Springer-Verlag Berlin Heidelberg and ESICM

Abstract
Over the last decade, the way to monitor hemodynamics at the bedside has evolved considerably in the intensive
care unit as well as in the operating room. The most important evolution has been the declining use of the pulmo-
nary artery catheter along with the growing use of echocardiography and of continuous, real-time, minimally or
totally non-invasive hemodynamic monitoring techniques. This article, which is the result of an agreement between
authors belonging to the Cardiovascular Dynamics Section of the European Society of Intensive Care Medicine, dis-
cusses the advantages and limits of using such techniques with an emphasis on their respective place in the hemody-
namic management of critically ill patients with hemodynamic instability.
Keywords: Hemodynamic monitoring, Pulse contour analysis, Transpulmonary thermodilution, Pulse pressure
variation, Esophageal Doppler, Bioreactance

Introduction the intensive care unit (ICU) [1–3]. Bedside monitoring

Patients with circulatory shock have a high risk of mor- methods have been developed to help clinicians to bet-
tality. Most often, the mechanisms involved in shock are ter assess the hemodynamic situation and to evaluate the
complex and involve more than one of the three major response to therapy.
hemodynamic abnormalities, namely hypovolemia, Over the last decade, hemodynamic monitoring has
myocardial dysfunction, and alteration in vascular tone. evolved considerably in the ICU as well as in the oper-
Sometimes, acute respiratory failure is associated with ating room. The most striking evolution has been the
shock, with risks of lung edema with fluid therapy. It is declining use of the pulmonary artery catheter (PAC)
thus fundamental to accurately assess the respective along with the growing use of either minimally or totally
degree of each of these components to select the most non-invasive hemodynamic monitoring techniques. The
appropriate therapeutic options. Clinical examination reasons for the declining use of the PAC are multiple.
is essential. Although it is of great value in the initial They include not only invasiveness (maintenance of a
phase of shock, it suffers from some limitations in reli- catheter in a pulmonary artery passing through the right
ably identifying the main hemodynamic problem in the ventricle) but also difficulties in appropriately measur-
complex situations that are frequently encountered in ing and interpreting the data [4] and findings from ran-
domized clinical trials showing no outcome benefit of
*Correspondence: jean-louis.teboul@aphp.fr using PAC in ICU patients [5]. Some less invasive tech-
1
Service de réanimation médicale, Hôpital de Bicêtre, Hôpitaux niques such as the transpulmonary thermodilution sys-
universitaires Paris-Sud, AP-HP, 78, rue du Général Leclerc, 94 270 Le
Kremlin-Bicêtre, France
tems still need the placement of a central venous catheter
Full author information is available at the end of the article and a femoral artery catheter, which carry risks of blood-
stream infections [6], although their use by intensivists
On behalf of the Cardiovascular Dynamics Section of the European
Society of Intensive Care Medicine (ESICM).
that have experience with these systems was shown to be
associated with a low rate of complications [7]. One of the or because of modification of therapy. A major interest of
main particularities of minimally and non-invasive tech- the pulse contour analysis systems is the real-time, short-
niques is their ability to provide continuous cardiac out- term tracking of CO changes induced by therapeutic tests
put (CO) and fluid responsiveness variables in real time. such as external (fluid administration) or internal (e.g.,
The importance of the concept of fluid responsiveness, passive leg raising) volume challenges. The pulse con-
extensively developed during recent years, is emphasized tour analysis systems also provide automatic calculation
by the two following facts. First, half of ICU patients are of dynamic indices of fluid responsiveness such as pulse
fluid non-responders as their CO does not increase with pressure variation (PPV) and/or stroke volume variation
fluid administration [8]. Second, fluid overload in ICU (SVV). Using PPV and SVV to predict fluid responsive-
patients was shown to be associated with increased mor- ness is based upon the concept of heart–lung interactions
tality [9]. Bedside techniques that provide indices of fluid during mechanical ventilation revealing cardiac preload
responsiveness are helpful to better assess the benefit/ dependence [12]. In  situations where PPV and SVV are
risk ratio of fluid therapy because outcome studies using not valid (e.g., spontaneous breathing activity, arrhyth-
these techniques in ICU patients are still lacking. mias, low tidal volume, low lung compliance), monitor-
In this article, we review the main minimally or non- ing pulse contour CO during internal volume challenges
invasive hemodynamic monitoring techniques. We also such as passive leg raising or end-expiratory occlusion
define their place in the management of ICU patients, can reliably predict fluid responsiveness [13, 14].
because no strong evidence has emerged in spite of the In clinical practice, the reliability of pulse contour CO
high number of articles published over the last decade. and derived variables is decisively dependent on the qual-
Most of them included a single-center evaluation and/or ity of the arterial pressure signal, i.e., over- and under-
a limited number of patients with heterogeneous cardio- damping of the signal, for example induced by bubbles of
vascular derangements. air within the liquid-filled or unnecessary prolonged arte-
A common characteristic of the minimally and non- rial lines.
invasive techniques is to measure and monitor CO, a
macrocirculatory variable which is well known by ICU Calibrated arterial pulse analysis systems
physicians. However, monitoring CO is far from being Transpulmonary thermodilution and lithium dilution can
enough to manage patients with complex hemodynamic serve to externally calibrate the pulse contour analysis.
disorders, since this variable is only one piece of the puz-
zle. Most of the monitoring techniques described in this Transpulmonary thermodilution The transpulmonary
article provide other relevant hemodynamic variables, thermodilution method provides intermittent measure-
which help to better define the macrocirculatory disor- ments of CO and other variables by applying the indica-
ders, to select the best therapy, and to monitor its effects. tor dilution principle based on temperature changes over
time (Fig. 1). The transpulmonary thermodilution devices
Minimally (or less invasive) hemodynamic [PiCCO (Pulsion Medical systems, Germany) and Vol-
technologies umeView (Edwards Lifesciences, USA)] are less invasive
In this section, we first consider the methods that use the than the PAC (no catheter traversing the heart) but still
arterial pulse contour analysis and then the esophageal require insertion of a central venous catheter (for cold
Doppler that uses ultrasound. bolus injection) and a thermistor-tipped (femoral) artery
catheter. This technique is being used in devices that com-
Methods that use arterial pulse contour analysis bine transpulmonary thermodilution and pulse contour
General principles analysis. The mathematical analysis of the thermodilution
All less invasive and non-invasive devices that estimate curve (blood temperature vs. time) allows calculation of
stroke volume from the arterial pressure pulse waveform the following variables: (1) CO; (2) global end-diastolic
are based on the principle of ventriculo-arterial coupling, volume, a volumetric estimate of global preload; (3) car-
in that the arterial pulse pressure and its contour are pri- diac function index and global ejection fraction, indica-
marily determined by left ventricular stroke volume and tors of cardiac systolic function; (4) extravascular lung
arterial impedance. Each device uses different propri- water (EVLW), a quantitative measure of lung edema;
etary algorithms based on slightly different assumptions and (5) pulmonary vascular permeability index, a marker
that make their interoperability questionable [10]. In gen- of lung capillary leak. There is acceptable agreement
eral, devices that are externally calibrated using an inde- between transpulmonary thermodilution and intermit-
pendent estimate of CO tend to be more accurate but tent pulmonary artery thermodilution measures of CO
do require frequent recalibration [11] if vasomotor tone in ICU patients [15]. The measurement of CO is reliable
changes, either spontaneously (e.g., as a result of sepsis) provided that three cold boluses are injected [16]. Moreo-
 –∆T in °C
m0
lung CO =
0.6

0.4 AUC
0.2

0.0

0 10 20 30 40 50 [s]

Descending aorta
SVC

Bolus RA
injec!on

Arterial
thermistor catheter
Fig. 1 Thermodilution method for intermittent cardiac output (CO) measurements. After injecting a cold indicator (usually saline) into the right
atrium (RA) via a central venous catheter, the resultant thermodilution curve can be derived in the descending aorta (transpulmonary thermodi-
lution). AUC area under thermodilution curve, m0 = amount (or mass) of injected cold at the time of injection (t0) = (blood temperature minus
injectate temperature) × (injectate volume minus dead space volume of catheter), −∆T = decrease in blood temperature, °C = degree Celsius, SVC
superior vena cava, RA right atrium

ver, the transpulmonary thermodilution bolus injection is tion vs. time). This technique has been validated against
being used to calibrate the artery pressure waveform anal- pulmonary artery thermodilution in humans [21]. As for
ysis that provides continuous, real-time calculation of CO transpulmonary thermodilution, three measurements
by using proprietary algorithms based on the relationship should be averaged to achieve a good precision [22]. The
between stroke volume and arterial pressure waveform. major inconvenience of this system is the need for lithium
An acceptable agreement between arterial pressure- injection, which is less safe than saline injection and can-
derived and thermodilution CO was reported in hemo- not be repeated infinitely because of lithium accumulation,
dynamically unstable patients [17]. However, frequent and moreover it is costly. The monitor also contains a pro-
recalibration is required [11]. prietary algorithm that converts an arterial blood pressure
One major advantage of the transpulmonary thermodi- waveform-based signal into an arterial blood flow meas-
lution devices is that they provide EVLW, which can be urement using a pulse power analysis. In addition the lith-
used as a safety parameter during fluid therapy, especially ium bolus injection serves to calibrate the system, which
in capillary leak states [18], where it was shown to have a then provides a beat-to-beat measurement of CO, PPV,
prognostic value [19, 20]. and SVV. The lithium dilution system can be used with
a radial artery catheter but it does not provide advanced
Lithium dilution The lithium dilution method (LiD- hemodynamic and volumetric variables such as EVLW.
COplus, LiDCO, UK) is an indicator dilution technique,
which provides intermittent CO measurements. A small Uncalibrated arterial pressure waveform analysis CO
amount of lithium chloride is injected through a central monitors
venous catheter, and changes in lithium levels are detected Some monitors provide real-time CO measurements by
in the blood drawn from a radial artery catheter over a deriving the stroke volume from the arterial pressure
lithium-selective sensor. The CO is then measured from waveform recorded from an arterial catheter, but they
analysis of the lithium dilution curve (lithium concentra- do so without external calibration. Several devices are
commercialized [FloTrac (Edwards Lifesciences, USA), constant distribution of CO between the upper territo-
LiDCOrapid (LiDCO UK), ProAQT (Pulsion Medical ries and the descending aorta, the CO value is inferred
Systems, Germany)] and use different proprietary algo- from the descending aorta blood flow value. The validity
rithms that analyze the characteristics of the arterial of CO estimation by esophageal Doppler was confirmed
pressure waveform along with patient-specific anthropo- in both critically ill and patients undergoing surgery [31].
metric and demographic data. By nature, these systems However, some limitations must be known. First, the dis-
necessarily use a statistical correction that mandates tribution of CO between the upper and the lower parts
a bias when a specific patient is out of standard range. of the arterial system can be affected by changes in the
These devices can be used with any arterial catheter. sympathetic tone, which occur frequently in patients
Knowing that frequent recalibration of pulse contour with shock and/or receiving vasoactive drugs. Second,
analysis is actually required in hemodynamically unsta- the diameter of the descending aorta is not measured but
ble patients to provide reliable data [11], it is clear that estimated from the patient’s characteristics. However,
the uncalibrated systems must become unreliable when the aorta at this level is compliant enough to change its
major hemodynamic changes are occurring. Hence, these diameter in response to changes in mean arterial pres-
systems should be restricted to hemodynamically stable sure [32]. Thus, currently available esophageal Doppler
patients or when CO monitoring is required for short systems that only estimate the aortic diameter bear a risk
periods of time, e.g., during surgery. In such situations of poorly tracking the real changes in CO during shock
and provided that CO is normal or low, the most recent resuscitation [32]. On the other hand, old models of
versions of uncalibrated CO monitoring devices provide esophageal Doppler probes that measure the aortic diam-
reliable CO measurements [23], as suggested by percent- eter carry some risk of error of measurement of stroke
age errors of less than 30 % [24] found in validation stud- volume, as even a limited error in the diameter can have
ies [23]. However, the upper limit of acceptability of the a significant impact as the radius is dependent on the
percentage error also depends on the reproducibility of square of that value. Finally, movements of the Doppler
the compared methods [25], which was not always pro- probe often occur in less-sedated patients, resulting in
vided in the studies that reported percentage error val- loss of the signal with the necessity of repositioning the
ues. The derived PPV and/or SVV is very suitable for probe. For these reasons, the use of esophageal Doppler
predicting fluid responsiveness in the operating room is more questionable in the ICU than in the operating
setting, where these indices are generally reliable [26] room setting, where its use for goal-directed hemody-
and, as such, used in many goal-directed algorithms for namic management was shown to decrease postsurgical
guiding intraoperative fluid management. Finally, the morbidity [33]. Nevertheless, esophageal Doppler can be
ability of uncalibrated CO monitors to track short-term helpful in sedated ICU patients for assessing short-term
changes in CO following fluid infusion could be accept- changes in CO such as those induced by fluid loading or
able [27], although divergent results were reported [23]. passive leg raising, especially when no other hemody-
The pressure-recording analytical method monitors namic monitoring systems are available.
CO in real time using a proprietary algorithm that takes
into account the area under the systolic part of the arte- Non-invasive techniques
rial pressure curve and the mean arterial pressure [28]. Fully non-invasive techniques providing CO estimation
This technology, implemented in the MostCare device have been introduced recently [34–36].
(Vytech, Italy), does not require any calibration or adjust- Continuous analysis of the arterial pressure waveform is
ments based on user-entered data. When compared to possible by using either the volume clamp method [Clear-
thermodilution, divergent results were reported [29, 30]. sight (Edwards Lifesciences, USA), ex Nexfin (BMYE,
Uncalibrated CO systems do not provide other hemo- NL), CNAP (CNSystems, Austria)] or the radial artery
dynamic variables than CO, PPV, or SVV. This represents applanation tonometry (T-Line, Tensys, USA) [35–37].
an important disadvantage for the complex hemody- As delineated in Fig. 2, the volume clamp method derives
namic situations compared to the advanced monitoring the finger arterial pressure waveform from the cuff pres-
methods such as the PAC or the transpulmonary ther- sure that is needed to keep the blood volume (assessed
modilution systems. by photoplethysmography) in the finger arteries constant
throughout the cardiac cycle [37]. The continuous radial
Esophageal Doppler artery applanation tonometry technique records the
Esophageal Doppler (CardioQ, Deltex Medical, UK) pro- arterial pressure waveform using a sensor that is electro-
vides real-time estimation of blood flow in the descend- mechanically driven over the radial artery [37] (Fig.  2).
ing thoracic aorta from the aortic blood velocity and By applying proprietary algorithms for pulse contour
the aortic diameter. On the basis of the hypothesis of a analysis to the non-invasively obtained arterial pressure
 Radial artery applanation tonometry Volume clamp method

Control System

Pulse contour analysis

proprietary algorithms

Arterial catheter

Fig. 2 Pulse contour analysis-derived cardiac output. Different techniques either non-invasive (radial artery applanation tonometry, volume clamp
method) or invasive (using an arterial catheter) can provide continuous and real-time CO from the pulse contour analysis

waveforms, these uncalibrated techniques provide CO surface electrodes that apply a low-amplitude and high-
estimations in a continuous manner. For the volume frequency electrical current, which traverses the thorax.
clamp method, validation studies showed good agreement Clinical validation studies showed contradicting results
and trending ability compared with reference techniques [48–50]. Bioreactance systems afforded acceptable results
in the perioperative context [38, 39]. However, poorer in cardiac surgery patients [48] but not in non-cardiac
results were reported after cardiac surgery and in ICU surgical ICU patients [49, 50]. CO measurements can be
patients [40–43], maybe as a result of alterations in vaso- disturbed by a variety of factors, such as pleural effusions,
motor tone [35, 36]. The radial applanation tonometry pulmonary edema, arrhythmias, electrical interference,
method is novel and the first clinical data are promising internal or external pacemakers, or movement.
[44, 45], but further confirmatory studies are required. The continuous and real-time estimation of CO based
Though easy to apply, each of the available systems still on the pulse wave transit time method (esCCO, Nihon
has specific limitations in its clinical applicability [35, 46]. Kohden, Japan) requires an electrocardiogram and a
The main limitations of the volume clamp method are pulse oximetry plethysmographical waveform [34, 35]. In
peripheral edema and severe vasoconstriction [35]. The theory, the pulse wave transit time (i.e., the time between
quality of the radial artery applanation tonometry signal the appearance of the R wave and the arrival of the pulse
can also be impaired by movement of the extremity where wave at the finger level) is inversely correlated with the
the sensor is placed [35]. stroke volume [35]. However, most studies comparing the
Other techniques that non-invasively estimate CO in pulse wave transit time-derived CO with reference meth-
real time are electrical bioimpedance and bioreactance as ods in ICU patients showed clinically unacceptable disa-
well as the pulse wave transit time method [34–36]. greement [51–54]. This might be explained by the fact
Bioimpedance [BioZ (Cardiodynamics, USA), Aesculon that CO estimation from pulse wave transit time can be
(Osypka Medical, Germany)] and bioreactance (NICOM, impeded in patients with vasoconstriction, cold extremi-
Cheetah Medical, Israel) systems derive CO from changes ties, and arrhythmias. Administration of vasopressors
in thoracic impedance or phase shift in voltage over the also limits the use of plethysmographic variability indices
cardiac cycle because pulsatile changes in intrathoracic to assess fluid responsiveness in critically ill patients [55,
blood volume induce changes in the electrical conduc- 56], whereas such indices are of great value in the intra-
tivity of the thorax [34, 35, 47]. These systems use skin operative setting [57, 58].
What is the place of less invasive hemodynamic as soon as possible to quickly obtain important informa-
monitoring in the ICU? tion on the systolic and diastolic ventricular functions [55].
There is a wide consensus to recommend insertion of arte- It also allows one to evaluate valvular competency and
rial and central venous catheters and early performance of diagnose/exclude obstructive shock (e.g., pericardial tam-
echocardiography in patients with shock [59]. The pres- ponade), knowing that CO measurements by echocardi-
ence of an arterial catheter allows measurements of sys- ography are not interchangeable with thermodilution CO
tolic arterial pressure (a reflection of the left ventricular measurements [66].
afterload), diastolic arterial pressure (an indicator of the Combination of all the pieces of information drawn
arterial tone), mean arterial pressure (a determinant of early from both clinical examination (mottling score, cap-
organ perfusion pressure used as a major target for hemo- illary refill time, etc.) and basic hemodynamic exploration
dynamic resuscitation), and pulse pressure, which if low is (arterial catheter, central venous catheter, and echocardi-
an indicator of a low stroke volume, especially in patients ography) is of importance to understand the underlying
with stiff arteries. In addition, the arterial catheter provides mechanisms of the shock state and to select the most log-
the value of PPV, which under appropriate conditions of ical initial therapy. If the hemodynamic status improves
interpretation is a good predictor of fluid responsiveness with this therapy, it is reasonable to continue with the
[15, 21]. In addition, the arterial catheter allows one to eas- same monitoring until complete resolution of the shock
ily perform repeated blood sampling for laboratory tests, state (Fig.  3). If, however, the patient does not respond
including arterial blood gas measurements. The presence (or insufficiently responds) to the initial therapy, it is
of a central venous catheter, which is inserted at least when recommended to obtain more information, in particular
vasoactive drugs are required, allows measurements of to measure CO to better evaluate the necessity to apply
central venous pressure (CVP) and central venous oxygen further fluids or inotropes and track the hemodynamic
saturation (ScvO2). It must be stressed that the CVP has response to these therapeutic measures [59]. In such
limited value in predicting fluid responsiveness [60–62], complex situations, the use of advanced hemodynamic
knowing that extreme values, although rarely encountered systems [59, 67] can be considered (Fig.  3). Insertion of
in ICU patients, still keep some value [62]. Nevertheless, a PAC can be indicated in the presence of a severe right
measuring changes in CVP can be helpful to monitor the ventricular dysfunction [59] diagnosed by echocardiog-
response to fluid therapy. In this regard, the CVP could raphy. This approach bears the advantage of monitoring
be used as a stopping rule (safety end-point) but not as a SvO2 and of measuring pulmonary artery pressure and
target for fluid resuscitation [63]. It is also important to pulmonary artery occlusion pressure, knowing that this
know the CVP value for estimating the perfusion pres- pressure shares the same limitations as CVP for assessing
sure of most organs, which is assumed to be reflected bet- fluid responsiveness. Transpulmonary thermodilution
ter by the difference between mean arterial pressure and systems on the other hand can take advantage of meas-
CVP rather than by the sole mean arterial pressure [64]. uring EVLW [18], especially in the context of acute res-
This could be particularly important to take into account piratory distress syndrome (ARDS) [59]. In case of severe
in cases of profound hypotension and high CVP. The ARDS associated with shock, it has been suggested to
ScvO2 is used as a surrogate of mixed venous blood oxy- consider using advanced monitoring devices at an earlier
gen saturation (SvO2), which reflects in real time the bal- phase (Fig. 3), when it is anticipated that the basic hemo-
ance between oxygen consumption and oxygen delivery. dynamic monitoring will not be sufficient to define a
Hence, a low ScvO2 may indicate insufficient global oxy- logical therapeutic approach [59, 67]. It must be stressed
gen delivery in case of shock and incite one to increase it. that a randomized study showed that hemodynamic
However, there are situations where absolute values as well management guided by transpulmonary thermodilution
as dynamic changes of ScvO2 and SvO2 differ [65]. Finally, vs. PAC did not affect outcomes of patients with shock
coupling arterial and central venous blood sampling allows [68], knowing that the use of PAC in ICU patients was
calculation of the venous-to-arterial carbon dioxide pres- never demonstrated to improve outcome [5]. On the
sure difference (PCO2 gap), which could be a good indi- other hand, it was also shown in a randomized trial that
cator of the adequacy of CO relative to the actual global fluid management guided by EVLW vs. pulmonary artery
metabolic conditions and could be helpful in conditions occlusion pressure resulted in a better maintained fluid
where oxygen extraction is altered while ScvO2 is within balance and a shorter duration of mechanical ventilation
the normal range. In this particular case, an abnormally and ICU length of stay in critically ill patients [69]. How-
high PCO2 gap (>6 mmHg) could suggest that CO should ever, results of such randomized studies [68, 69] should
be elevated to improve tissue oxygenation. Echocardiogra- be cautiously interpreted since therapeutic algorithms
phy, which is not a hemodynamic monitoring device but based on measurements with any single device can be
rather a diagnostic tool, is recommended to be performed criticized [70].
 acute circulatory failure

AP, PPV
CVP, ScvO2
Central venous catheter Clinical assessment Lactate Echocardiography Arterial catheter PaCO2
PcvCO2
PaO2, SaO2

associated severe ARDS ?

NO YES

• Transpulmonary thermodilu!on systems

posi!ve response insufficient response or
to ini!al therapy to ini!al therapy • Pulmonary artery catheter

(especially in case of RV dysfunc!on)

Con!nue with same hemodynamic monitoring

un!l shock resolu!on

Fig. 3 Simplified algorithm for the choice of hemodynamic monitoring in patients with acute circulatory failure. AP arterial pressure, ARDS acute
respiratory distress syndrome, CVP central venous pressure, PaCO2 carbon dioxide pressure in the arterial blood, PaO2 oxygen pressure in the arterial
blood, PcvCO2 carbon dioxide pressure in the central venous blood, PPV pulse pressure variation, RV right ventricular, SaO2 arterial blood oxygen
saturation, ScvO2 central venous blood oxygen saturation

The place of devices using uncalibrated arterial pressure functional images, e.g., by electrical impedance tomog-
waveform analysis is more limited in the context of shock, raphy, will also increase the amount of information avail-
as they rapidly become less reliable and cannot provide able at the bedside [75, 76]. Further intelligent visual
other variables than CO, PPV, and/or SVV, which are too postprocessing of hemodynamic information in graphi-
limited in the context of complex shock when different cal displays will potentially facilitate the understanding of
mechanisms may coexist and when associated with ARDS. complex pathophysiology [77]. This will be advanced by an
Esophageal Doppler and less invasive uncalibrated increasing connectivity of different monitoring systems,
devices are predominantly reserved for the perioperative which will maybe further push the development of tools for
setting [71] where goal-directed hemodynamic optimiza- predictive analytics [78]. For sure, telemetric monitoring
tion based on algorithms using variables included these will become available for much more complex physiological
monitoring devices may result in improved outcomes signals, which will offer the opportunity to expand patient
[33], in particular when these devices allow using goal- surveillance beyond the doors of the ICU [74]. For more
directed fluid therapy based on dynamic variables of than one decade, clinical research has been performed in
preload responsiveness [72, 73]. Non-invasive hemody- the field of the monitoring of microcirculation. In spite of
namic monitors are currently not recommended for use abundant literature on the potential interest of such moni-
in patients with shock since these patients need arterial toring to manage patients with shock, in part explained by
catheterization anyway. dissociation between the macrocirculation and the micro-
circulation [79], no bedside monitors are currently available
What could the future of hemodynamic monitoring for clinical practice [59]. It is expected that technological
be? developments in this field will allow one to better select and
It is hard to predict the future, but for hemodynamic adjust therapies for treating patients with shock states.
monitoring, the future will become more non-invasive for
sure. Visualization of complex information, either by cre- Conclusion
ating more detailed real, anatomical images [74], such as During the few last years, hemodynamic monitoring
by pocket-size 2D and (in the future) 3D ultrasound, or has evolved considerably from invasiveness to less or no
invasiveness and from intermittent to continuous and 2. Saugel B, Ringmaier S, Holzapfel K, Schuster T, Phillip V, Schmid RM, Huber
W (2011) Physical examination, central venous pressure, and chest radi-
real-time measurements of hemodynamic variables. New ography for the prediction of transpulmonary thermodilution-derived
parameters such as fluid responsiveness indices (PPV, hemodynamic parameters in critically ill patients: a prospective trial. J Crit
SVV), EVLW, and volumetric measures of preload have Care 26:402–410
3. Perel A, Saugel B, Teboul JL, Malbrain ML, Belda FJ, Fernández-Mondéjar
also been implemented in less invasive hemodynamic E, Kirov M, Wendon J, Lussmann R, Maggiorini M (2015) The effects
monitors making them particularly attractive to manage of advanced monitoring on hemodynamic management in critically
patients with complex shock. Non-invasive monitors are ill patients: a pre and post questionnaire study. J Clin Monit Comput.
doi:10.1007/s10877-015-9811-7
increasingly used in high-risk surgical patients. Contin- 4. Gnaegi A, Feihl F, Perret C (1997) Intensive care physicians insufficient
ual technological refinements will probably make them knowledge of right-heart catheterization at the bedside: time to act? Crit
become the hemodynamic monitoring of the future. Care Med 25:213–220
5. Rajaram SS, Desai NK, Kalra A, Gajera M, Cavanaugh SK, Brampton W,
Young D, Harvey S, Rowan K (2013) Pulmonary artery catheters for adult
patients in intensive care. Cochrane Database Syst Rev 2:CD003408
Author details 6. O’Horo JC, Maki DG, Krupp AE, Safdar N (2014) Arterial catheters as a
1
Service de réanimation médicale, Hôpital de Bicêtre, Hôpitaux universitaires source of bloodstream infection: a systematic review and meta-analysis.
Paris-Sud, AP-HP, 78, rue du Général Leclerc, 94 270 Le Kremlin-Bicêtre, France. Crit Care Med 42:1334–1339
2
Department of Anesthesiology, Center of Anesthesiology and Intensive Care 7. Belda FJ, Aguilar G, Teboul JL, Pestaña D, Redondo FJ, Malbrain M, Luis
Medicine, University Medical Center Hamburg-Eppendorf, 20246 Hamburg, JC, Ramasco F, Umgelter A, Wendon J, Kirov M, Fernández-Mondéjar E,
Germany. 3 Anaesthesia and Intensive Care, St George’s Hospital and Medi- PICS Investigators Group (2011) Complications related to less-invasive
cal School, London, UK. 4 Department of Intensive Care, CHIREC Hospitals haemodynamic monitoring. Br J Anaesth 106:482–486
(Université Libre de Bruxelles), Brussels, Belgium. 5 Department of Transversal 8. Michard F, Teboul JL (2002) Predicting fluid responsiveness in ICU
Medicine, Institute of Anesthesiology and Intensive Care Medicine, Triemli City patients: a critical analysis of the evidence. Chest 121:2000–2008
Hospital, Zurich, Switzerland. 6 Department of Anesthesiology and Intensive 9. Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K, Gerlach H, Moreno
Care, Sheba Medical Center, Tel Aviv University, Tel Aviv, Israel. 7 Department R, Carlet J, Le Gall JR, Payen D, Sepsis Occurrence in Acutely Ill Patients
of Critical Care Medicine, University of Pittsburgh, Pittsburgh, PA 15261, USA. (2006) Sepsis in European intensive care units: results of the SOAP study.
8
Clinique Ambroise Paré, 92200 Neuilly-Sur-Seine, France. 9 Department Crit Care Med 34:344–353
of Intensive Care, Erasme University Hospital, Université Libre de Bruxelles, 10. Hadian M, Kim H, Severyn DA, Pinsky MR (2010) Cross-comparison of
Brussels, Belgium. 10 Department of Anesthesiology, University of Groningen, cardiac output trending accuracy of LiDCO, PiCCO FloTrac and pulmonary
University Medical Center Groningen, Groningen, The Netherlands. artery catheters. Crit Care 14:R212
11. Hamzaoui O, Monnet X, Richard C, Osman D, Chemla D, Teboul JL (2008)
Compliance with ethical standards Effects of changes in vascular tone on the agreement between pulse
contour and transpulmonary thermodilution cardiac output measure-
Conflicts of interest ments within an up to 6-hour calibration-free period. Crit Care Med
JLT is a member of the medical advisory board of Pulsion Medical Systems 36:434–440
and received honoraria from Edwards Lifesciences and Masimo Inc. for 12. Michard F, Boussat S, Chemla D, Anguel N, Mercat A, Lecarpentier Y,
consulting. BS is a member of the medical advisory board of Pulsion Medical Richard C, Pinsky MR, Teboul JL (2000) Relation between respiratory
Systems and a received institutional research grants, unrestricted research changes in arterial pulse pressure and fluid responsiveness in septic
grants, and refunds of travel expenses from Tensys Medical Inc. BS received patients with acute circulatory failure. Am J Respir Crit Care Med
honoraria for giving lectures for CNSystems Medizintechnik AG. MC consulted 162:134–138
and lectured for Edwards Lifesciences and LiDCO. He received support from 13. Marik PE, Monnet X, Teboul JL (2011) Hemodynamic parameters to guide
Edwards Lifesciences, LiDCO, Deltex Medical, Applied Physiology, Masimo, fluid therapy. Ann Intensive Care 1:1
Bmeye, Cheetah Medical, Imacor (travel expenses, honoraria, advisory board, 14. Monnet X, Osman D, Ridel C, Lamia B, Richard C, Teboul JL (2009)
unrestricted educational grant, and research material). DDB received honoraria Predicting volume responsiveness by using the end-expiratory occlusion
for lectures for Edwards Lifesciences and Nihon Kohden. DDB received grant/ in mechanically ventilated intensive care unit patients. Crit Care Med
material for studies for Edwards Lifesciences, Maquet, Vytech, Cheetah, Imacor, 37:951–956
and Nihon Kohden. XM is a member of the medical advisory board of Pulsion 15. Sakka SG, Reinhart K, Meier-Hellmann A (1999) Comparison of pulmonary
Medical systems and received honoraria from Cheetah Medical for consulting. artery and arterial thermodilution cardiac output in critically ill patients.
AP is a member of the medical advisory board of Pulsion Medical Systems and Intensive Care Med 25:843–846
is a consultant for Masimo Inc. MRP is a consultant for Edwards Lifesciences, 16. Monnet X, Persichini R, Ktari M, Jozwiak M, Richard C, Teboul JL (2011)
Masimo Inc., and LiDCO and has stock options in LiDCO and Cheetah Medical Precision of the transpulmonary thermodilution measurements. Crit Care
companies. DAR is a member of the medical advisory board of Pulsion Medi- 15:R204
cal Systems and gave lectures for Edwards Lifesciences. AR has no conflict of 17. Gödje O, Höke K, Goetz AE, Felbinger TW, Reuter DA, Reichart B, Friedl R,
interest to declare. PS was a consultant for Cheetah Medical and for Edwards Hannekum A, Pfeiffer UJ (2002) Reliability of a new algorithm for continu-
Lifesciences. TS received honoraria from Edwards Lifesciences and Masimo Inc. ous cardiac output determination by pulse-contour analysis during
for consulting. TS received honoraria from Pulsion Medical Systems for lectur- hemodynamic instability. Crit Care Med 30:52–58
ing. JLV has no conflict of interest to declare. 18. Jozwiak M, Teboul JL, Monnet X (2015) Extravascular lung water in critical
care: recent advances and clinical applications. Ann Intensive Care 5:38
Received: 2 February 2016 Accepted: 28 April 2016 19. Cordemans C, De Laet I, Van Regenmortel N, Schoonheydt K, Dits H,
Huber W, Malbrain ML (2012) Fluid management in critically ill patients:
the role of extravascular lung water, abdominal hypertension, capillary
leak, and fluid balance. Ann Intensive Care 5:2
20. Jozwiak M, Silva S, Persichini R, Anguel N, Osman D, Richard C, Teboul JL,
Monnet X (2013) Extravascular lung water is an independent prognostic
References factor in patients with acute respiratory distress syndrome. Crit Care Med
1. Connors AF Jr, McCaffree DR, Gray BA (1983) Evaluation of right-heart 41:472–480
catheterization in the critically ill patient without acute myocardialinfarc- 21. Linton RA, Band DM, Haire KM (1993) A new method of measuring
tion. N Engl J Med 308:263–267 cardiac output in man using lithium dilution. Br J Anaesth 71:262–266
22. Cecconi M, Dawson D, Grounds R, Rhodes A (2009) Lithium dilution 44. Saugel B, Meidert AS, Langwieser N, Wagner JY, Fassio F, Hapfelmeier
cardiac output measurement in the critically ill patient: determination of A, Prechtl LM, Huber W, Schmid RM, Godje O (2014) An autocalibrating
precision of the technique. Intensive Care Med 35:498–504 algorithm for non-invasive cardiac output determination based on the
23. Slagt C, Malagon I, Groeneveld AB (2014) Systematic review of uncali- analysis of an arterial pressure waveform recorded with radial artery
brated arterial pressure waveform analysis to determine cardiac output applanation tonometry: a proof of concept pilot analysis. J Clin Monit
and stroke volume variation. Br J Anaesth 112:626–637 Comput 28:357–362
24. Critchley LA, Critchley JA (1999) A meta-analysis of studies using bias and 45. Wagner JY, Sarwari H, Schon G, Kubik M, Kluge S, Reichenspurner H,
precision statistics to compare cardiac output measurement techniques. Reuter DA, Saugel B (2015) Radial artery applanation tonometry for
J Clin Monit Comput 15:85–91 continuous noninvasive cardiac output measurement: a comparison with
25. Hapfelmeier A, Cecconi M, Saugel B (2016) Cardiac output method intermittent pulmonary artery thermodilution in patients after cardiotho-
comparison studies: the relation of the precision of agreement and the racic surgery. Crit Care Med 43:1423–1428
precision of method. J Clin Monit Comput 30:149–155 46. Saugel B, Reuter DA (2014) Are we ready for the age of non-invasive
26. Yang X, Du B (2014) Does pulse pressure variation predict fluid respon- haemodynamic monitoring? Br J Anaesth 113:340–343
siveness in critically ill patients? A systematic review and meta-analysis. 47. Hofer CK, Rex S, Ganter MT (2014) Update on minimally invasive hemo-
Crit Care 18:650 dynamic monitoring in thoracic anesthesia. Curr Opin Anaesthesiol
27. Monnet X, Vaquer S, Anguel N, Jozwiak M, Cipriani F, Richard C, Teboul JL 27:28–35
(2015) Comparison of pulse contour analysis by Pulsioflex and Vigileo to 48. Squara P, Denjean D, Estagnasie P, Brusset A, Dib JC, Dubois C (2007)
measure and track changes of cardiac output in critically ill patients. Br J Noninvasive cardiac output monitoring (NICOM): a clinical validation.
Anaesth 114:235–243 Intensive Care Med 33:1191–1194
28. Romano SM, Pistolesi M (2002) Assessment of cardiac output from 49. Kupersztych-Hagege E, Teboul JL, Artigas A, Talbot A, Sabatier C, Richard
systemic arterial pressure in humans. Crit Care Med 30:1834–1841 C, Monnet X (2014) Bioreactance is not reliable for estimating cardiac
29. Franchi F, Silvestri R, Cubattoli L, Taccone FS, Donadello K, Romano SM, output and the effects of passive leg raising in critically ill patients. Br J
Giomarelli P, McBride WT, Scolletta S (2011) Comparison between an Anaesth 111:961–966
uncalibrated pulse contour method and thermodilution technique for 50. Fagnoul D, Vincent JL, de Backer D (2012) Cardiac output measurements
cardiac output estimation in septic patients. Br J Anaesth 107:202–208 using the bioreactance technique in critically ill patients. Crit Care 16:460
30. Gopal S, Do T, Pooni JS, Martinelli G (2014) Validation of cardiac output 51. Yamada T, Tsutsui M, Sugo Y, Sato T, Akazawa T, Sato N, Yamashita K,
studies from the Mostcare compared to a pulmonary artery catheter in Ishihara H, Takeda J (2012) Multicenter study verifying a method of
septic patients. Minerva Anestesiol 80:314–323 noninvasive continuous cardiac output measurement using pulse wave
31. Dark PM, Singer M (2004) The validity of trans-esophageal Doppler ultra- transit time: a comparison with intermittent bolus thermodilution cardiac
sonography as a measure of cardiac output in critically ill adults. Intensive output. Anesth Analg 115:82–87
Care Med 30:2060–2066 52. Ball TR, Tricinella AP, Kimbrough BA, Luna S, Gloyna DF, Villamaria FJ, Culp
32. Monnet X, Chemla D, Osman D, Anguel N, Richard C, Pinsky MR, Teboul WC Jr (2013) Accuracy of noninvasive estimated continuous cardiac
JL (2007) Measuring aortic diameter improves accuracy of esophageal output (esCCO) compared to thermodilution cardiac output: a pilot study
Doppler in assessing fluid responsiveness. Crit Care Med 35:477–482 in cardiac patients. J Cardiothorac Vasc Anesth 27:1128–1132
33. Hamilton MA, Cecconi M, Rhodes A (2011) A systematic review and meta- 53. Biais M, Berthezene R, Petit L, Cottenceau V, Sztark F (2015) Ability of
analysis on the use of preemptive hemodynamic intervention to improve esCCO to track changes in cardiac output. Br J Anaesth 115:403–410
postoperative outcomes in moderate and high-risk surgical patients. 54. Thonnerieux M, Alexander B, Binet C, Obadia JF, Bastien O, Desebbe O
Anesth Analg 112:1392–1402 (2015) The ability of esCCO and ECOM monitors to measure trends in car-
34. Marik PE (2013) Noninvasive cardiac output monitors: a state-of the-art diac output during alveolar recruitment maneuver after cardiac surgery:
review. J Cardiothorac Vasc Anesth 27:121–134 a comparison with the pulmonary thermodilution method. Anesth Analg
35. Saugel B, Cecconi M, Wagner JY, Reuter DA (2015) Noninvasive continu- 121:383–391
ous cardiac output monitoring in perioperative and intensive care 55. Biais M, Cottenceau V, Petit L, Masson F, Cochard JF, Sztark F (2011) Impact
medicine. Br J Anaesth 114:562–575 of norepinephrine on the relationship between pleth variability index
36. Thiele RH, Bartels K, Gan TJ (2015) Cardiac output monitoring: a contem- and pulse pressure variations in ICU adult patients. Crit Care 15:R168
porary assessment and review. Crit Care Med 43:177–185 56. Monnet X, Guérin L, Jozwiak M, Bataille A, Julien F, Richard C, Teboul JL
37. Saugel B, Dueck R, Wagner JY (2014) Measurement of blood pressure. (2013) Pleth variability index is a weak predictor of fluid responsiveness in
Best Pract Res Clin Anaesthesiol 28:309–322 patients receiving norepinephrine. Br J Anaesth 110:207–213
38. Broch O, Renner J, Gruenewald M, Meybohm P, Schottler J, Caliebe A, 57. Cannesson M, Desebbe O, Rosamel P, Delannoy B, Robin J, Bastien O,
Steinfath M, Malbrain M, Bein B (2012) A comparison of the Nexfin(R) and Lehot JJ (2008) Pleth variability index to monitor the respiratory vari-
transcardiopulmonary thermodilution to estimate cardiac output during ations in the pulse oximeter plethysmographic waveform amplitude
coronary artery surgery. Anaesthesia 67:377–383 and predict fluid responsiveness in the operating theatre. Br J Anaesth
39. Chen G, Meng L, Alexander B, Tran NP, Kain ZN, Cannesson M (2012) 101:200–206
Comparison of noninvasive cardiac output measurements using the 58. Forget P, Lois F, de Kock M (2010) Goal-directed fluid management based
Nexfin monitoring device and the esophageal Doppler. J Clin Anesth on the pulse oximeter-derived pleth variability index reduces lactate
24:275–283 levels and improves fluid management. Anesth Analg 111:910–914
40. Fischer MO, Avram R, Cârjaliu I, Massetti M, Gérard JL, Hanouz JL, Fellahi 59. Cecconi M, De Backer D, Antonelli M, Beale R, Bakker J, Hofer C, Jaeschke
JL (2012) Non-invasive continuous arterial pressure and cardiac index R, Mebazaa A, Pinsky MR, Teboul JL, Vincent JL, Rhodes A (2014) Con-
monitoring with Nexfin after cardiac surgery. Br J Anaesth 109:514–521 sensus on circulatory shock and hemodynamic monitoring. Task force
41. Monnet X, Picard F, Lidzborski E, Mesnil M, Duranteau J, Richard C, Teboul of the European Society of Intensive Care Medicine. Intensive Care Med
JL (2012) The estimation of cardiac output by the Nexfin device is of poor 40:1785–1815
reliability for tracking the effects of a fluid challenge. Crit Care 16:R212 60. Marik PE, Cavallazzi R (2013) Does the central venous pressure predict
42. Taton O, Fagnoul D, De Backer D, Vincent JL (2013) Evaluation of cardiac fluid responsiveness? An updated meta-analysis and a plea for some
output in intensive care using a non-invasive arterial pulse contour common sense. Crit Care Med 41:1774–1781
technique (Nexfin((R))) compared with echocardiography. Anaesthesia 61. Marik PE (2014) Iatrogenic salt water drowning and the hazards of a high
68:917–923 central venous pressure. Ann Intensive Care 4:21
43. Wagner JY, Grond J, Fortin J, Negulescu I, Schofthaler M, Saugel B (2016) 62. Eskesen TG, Wetterslev M, Perner A (2016) Systematic review including
Continuous noninvasive cardiac output determination using the CNAP re-analyses of 1148 individual data sets of central venous pressure as a
system: evaluation of a cardiac output algorithm for the analysis of predictor of fluid responsiveness. Intensive Care Med 42:324–332
volume clamp method-derived pulse contour. J Clin Monit Comput. 63. Pinsky MR, Kellum JA, Bellomo R (2014) Central venous pressure is a stop-
doi:10.1007/s10877-015-9744-1 ping rule, not a target of fluid resuscitation. Crit Care Resus 16:245–246
 64. Wong BT, Chan MJ, Glassford NJ, Mårtensson J, Bion V, Chai SY, Oughton 72. Scheeren TW, Wiesenack C, Gerlach H, Marx G (2013) Goal-directed
C, Tsuji IY, Candal CL, Bellomo R (2015) Mean arterial pressure and intraoperative fluid therapy guided by stroke volume and its variation in
mean perfusion pressure deficit in septic acute kidney injury. J Crit Care high-risk surgical patients: a prospective randomized multicentre study. J
30:975–981 Clin Monit Comput 27:225–233
65. Squara P (2014) Central venous oxygenation: when physiology explains 73. Benes J, Giglio M, Brienza N, Michard F (2014) The effects of goal-directed
apparent discrepancies. Crit Care 18:579 fluid therapy based on dynamic parameters on post-surgical outcome: a
66. Wetterslev M, Møller-Sørensen H, Johansen RR, Perner A (2016) System- meta-analysis of randomized controlled trials. Crit Care 18:584
atic review of cardiac output measurements by echocardiography vs. 74. Michard F (2016) Hemodynamic monitoring in the era of digital health.
thermodilution: the techniques are not interchangeable. Intensive Care Ann Intensive Care 6:15
Med. doi:10.1007/s00134-016-4258-y 75. Maisch S, Bohm SH, Solà J, Goepfert MS, Kubitz JC, Richter HP, Ridder J,
67. Jozwiak M, Monnet X, Teboul JL (2015) Monitoring: from cardiac output Goetz AE, Reuter DA (2011) Heart-lung interactions measured by electri-
monitoring to echocardiography. Curr Opin Crit Care 21:395–401 cal impedance tomography. Crit Care Med 39:2173–2176
68. Trof RJ, Beishuizen A, Cornet AD, de Wit RJ, Girbes AR, Groeneveld AB 76. Biais M, Carrié C, Delaunay F, Morel N, Revel P, Janvier G (2012) Evaluation
(2012) Volume-limited versus pressure-limited hemodynamic manage- of a new echoscopic device for focused cardiac ultrasonography in an
ment in septic and nonseptic shock. Crit Care Med 40:1177–1185 emergency setting. Crit Care 16:R82
69. Mitchell JP, Schuller D, Calandrino FS, Schuster DP (1992) Improved 77. Drews FA, Westenskow DR (2006) The right picture is worth a thousand
outcome based on fluid management in critically ill patients requiring numbers: data displays in anesthesia. Hum Factors 48(1):59–71
pulmonary artery catheterization. Am Rev Respir Dis 145:990–998 78. Pinsky MR, Dubrawski A (2014) Gleaning knowledge from data in the ICU.
70. Teboul JL, Monnet X, Perel A (2012) Results of questionable management Am J Respir Crit Care Med 190:606–610
protocols are inherently questionable. Crit Care Med 40:2536 79. De Backer D, Donadello K, Sakr Y, Ospina-Tascon G, Salgado D, Scolletta
71. Vincent JL, Pelosi P, Pearse R, Payen D, Perel A, Hoeft A, Romagnoli S, S, Vincent JL (2013) Microcirculatory alterations in patients with severe
Ranieri VM, Ichai C, Forget P, Della Rocca G, Rhodes A (2015) Perioperative sepsis: impact of time of assessment and relationship with outcome. Crit
cardiovascular monitoring of high-risk patients: a consensus of 12. Crit Care Med 41:791–799
Care 19:224

View publication stats