Fellahi et al. Ann.

Intensive Care (2021) 11:58

https://doi.org/10.1186/s13613-021-00845-1

REVIEW Open Access

Perioperative hemodynamic optimization:

from guidelines to implementation—an experts’
opinion paper
Jean‑Luc Fellahi1,2*, Emmanuel Futier3,4, Camille Vaisse5, Olivier Collange6,7, Olivier Huet8,9, Jerôme Loriau10,
Etienne Gayat11,12, Benoit Tavernier13, Matthieu Biais14,15, Karim Asehnoune16,17, Bernard Cholley18,19,20 and
Dan Longrois21,22

Abstract
Despite a large body of evidence, the implementation of guidelines on hemodynamic optimization and goal-directed
therapy remains limited in daily routine practice. To facilitate/accelerate this implementation, a panel of experts in
the field proposes an approach based on six relevant questions/answers that are frequently mentioned by clini‑
cians, using a critical appraisal of the literature and a modified Delphi process. The mean arterial pressure is a major
determinant of organ perfusion, so that the authors unanimously recommend not to tolerate absolute values below
65 mmHg during surgery to reduce the risk of postoperative organ dysfunction. Despite well-identified limitations,
the authors unanimously propose the use of dynamic indices to rationalize fluid therapy in a large number of patients
undergoing non-cardiac surgery, pending the implementation of a “validity criteria checklist” before applying volume
expansion. The authors recommend with a good agreement mini- or non-invasive stroke volume/cardiac output
monitoring in moderate to high-risk surgical patients to optimize fluid therapy on an individual basis and avoid vol‑
ume overload. The authors propose to use fluids and vasoconstrictors in combination to achieve optimal blood flow
and maintain perfusion pressure above the thresholds considered at risk. Although purchase of disposable sensors
and stand-alone monitors will result in additional costs, the authors unanimously acknowledge that there are data
strongly suggesting this may be counterbalanced by a sustained reduction in postoperative morbidity and hospital
lengths of stay. Beside existing guidelines, knowledge and explicit clinical reasoning tools followed by decision algo‑
rithms are mandatory to implement individualized hemodynamic optimization strategies and reduce postoperative
morbidity and duration of hospital stay in high-risk surgical patients.
Keywords: Hemodynamic optimization, Blood pressure, Fluid responsiveness, Vasopressors, Perioperative morbidity,
High-risk surgery, Health costs

Introduction comorbidities, and explains the renewed interest on

Postoperative mortality has been reported to vary strategies that would result into lower postoperative
between 1 and 4% in developed countries [1]. This mor- mortality. Hemodynamic optimization (goal-directed
tality rate is orders of magnitude higher than that of therapy or GDT) has been recommended by several
anesthesia-related deaths, even for patients with severe international guidelines because it has been shown con-
sistently to improve outcomes [2]. However, the imple-
mentation of guidelines on GDT remains limited in
*Correspondence: [email protected] routine practice [3, 4]. The explanations for absent/poor
1
Service D’Anesthésie-Réanimation, Hôpital Louis Pradel, 59 boulevard
Pinel, 69500 Hospices Civils de Lyon, Lyon, France implementation of the guidelines are numerous [5–7] but
Full list of author information is available at the end of the article it is possible that the current guidelines on perioperative

© The Author(s) 2021. This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and
the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material
in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material
is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the
permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​
mmons.​org/​licen​ses/​by/4.​0/.
Fellahi et al. Ann. Intensive Care (2021) 11:58 Page 2 of 10

GDT are on the one hand not explicit enough to allow a To date, there is no universal definition for arterial
reproducible decision-making process, and on the other hypotension during anesthesia. However, evidence from
hand not suited to manage uncertainty [8]. Management several large observational studies suggest that intraoper-
of uncertainty refers to situations where there is no evi- ative mean arterial pressure (MAP) below 60–70 mmHg
dence reported in the guidelines or situations where the may be associated with postoperative acute kidney injury,
clinical reasoning tools (analysis of “abnormality”, criti- myocardial injury, and death [11–14]. Although injury is
cal analysis of the information gathered from the clinical magnified with increasing hypotension magnitude, avail-
situations/monitors, positive and differential diagnoses, able evidence suggests that MAP below 60 mmHg sus-
treatment plan, evaluation of the effectiveness of the tained for 5 min or more may be associated with organ
treatment plan) are not explicit [9]. dysfunction and increased mortality [15, 16]. Elevated
To facilitate/accelerate implementation of guidelines risks or organ injury were also reported with prolonged
on perioperative GDT, a national panel of experts in the exposure (< 10 min) to MAP below 70 mmHg [17]. To
field, representative of the whole French territory and specifically prevent acute kidney injury, French Guide-
who have previously collaborated on guidelines of the lines suggest to maintain MAP between 60 and 70 mmHg
French Society of Anesthesiology and Critical Care is intraoperatively [18]. In patients with chronic arterial
proposing an approach based on questions/answers on hypertension undergoing elective non-cardiac surgery,
issues that are frequently mentioned by clinicians. These targeting MAP values higher than 70 mmHg may be rea-
six relevant questions are: sonable, ideally adapted to the clinical and surgery condi-
tions [19]. Even if systolic, pulse pressure (the difference
1. Which blood pressure goals should be targeted dur- between systolic and diastolic pressures) and MAP were
ing anesthesia and the perioperative period? recently found to have comparable discriminative abil-
2. Intraoperative fluids management: restrictive, stand- ity in evaluating the risk of organ injury [20], MAP is
ard, liberal and beyond? the major determinant of organ perfusion. Moreover,
3. Are dynamic indices and maneuvers useful to predict systolic and diastolic pressures are indirectly calculated
and manage volume expansion? from MAP and may be less reliable when the oscillo-
4. When should we measure stroke volume (SV) and metric method is used. For this reason, MAP should be
cardiac output (CO)? the main monitoring variable on which interventions to
5. Fluids or vasoconstrictors: how to decide? prevent/correct arterial hypotension should be based
6. What is the economic impact of hemodynamic mon- on. Whether or not relative changes in MAP for a given
itoring for GDT? patient rather than absolute values should be preferred
is still a matter of debate. In a simpler approach, anes-
The members of the panel first identified the questions thetic management of arterial hypotension could be
and subsequently organized the answers by summariz- based on absolute values without considering percentage
ing elements of guidelines, knowledge, and explicit clini- changes from (difficult to document) preoperative arte-
cal reasoning tools followed by decision algorithms when rial pressure values [21]. In a more complex approach, a
appropriate. The decision algorithms were adopted fol- multicenter randomized controlled trial performed in
lowing a modified Delphi process and the RAND/UCLA high-risk surgical patients undergoing major abdomi-
Appropriateness Method [10]. We consider that explicit nal surgery (all received SV-guided intraoperative fluid)
clinical reasoning tools will improve the decision-making compared an individualized strategy using low-dose
process and facilitate management of uncertainty. norepinephrine to maintain intraoperative systolic arte-
rial pressure within 10% of the preoperative reference
Q1. Which blood pressure goals should be targeted value to a strategy of standard management. The study
during anesthesia and the perioperative period? revealed a significant reduction in postoperative organ
Monitoring blood pressure is a prerequisite during anes- dysfunction in the individualized strategy group [22]. Of
thesia. The main objective is to prevent postoperative note, the use of intraoperative continuous low-dose nor-
complications resulting from either arterial hypo- or epinephrine on a devoted peripheral intravenous line was
hypertension. The arterial pressure can be monitored at safe and can be recommended for routine practice.
different anatomical sites, using various techniques, non- Finally, because even short cumulative durations of
invasively or invasively, and in a continuous or intermit- arterial hypotension are associated with poor outcome
tent manner. Although all excessive excursions of arterial and because continuous (versus intermittent) measure-
pressure may be detrimental, the association between ment of arterial pressure was associated with higher
intraoperative arterial hypotension and postoperative sensitivity to diagnose arterial hypotension during anes-
complications is the most thoroughly documented. thesia, continuous measurement of arterial pressure
Fellahi et al. Ann. Intensive Care (2021) 11:58 Page 3 of 10

should be preferred. Non-invasive continuous monitor- similar for all individuals. This “one size fits all” strategy
ing of arterial pressure is not yet considered interchange- is bound to result in inadequate volume management in
able with invasive monitoring and there is insufficient the vast majority of patients [25]. Low-risk patients will
evidence to recommend its use in high-risk surgical usually tolerate the deviation from their adequate volume
patients and/or high-risk surgery. Future efforts should requirements, but high-risk patients will be exposed to
concentrate on trying to verify whether treating hypoten- the above-mentioned complications. The available lit-
sive episodes as detected by those new techniques results erature do not provide evidence-based recommendations
in improvement in patient outcome rather that repeat- regarding continuous fluid infusion. The basal fluid losses
ing validation studies that are bound to provide the same via insensible perspiration are approximately 0.5 ml/kg/h,
results over and over. extending to 1 ml/kg/h during major abdominal surgery
[26]. When continuous fluid infusion is used, it should
Experts’ opinion: 
We propose to prefer continuous be limited to less than 2 ml/kg/h, including drug infusion
invasive arterial pressure moni- [25].
toring in moderate to high-risk Therefore, optimizing tissue perfusion in high-risk
surgical patients, ideally using patients relies on an individualized approach. The MAP
an algorithm-based approach will be maintained above a level close to the usual value
which aims at preventing/manag- of the patient, as suggested above. Minimal value for
ing arterial hypotension. We also hemoglobin concentration and transfusion thresholds
propose to maintain MAP above will vary according to the comorbidities of each patient.
And since the ideal values of organ flow for a given
65 mmHg or within 10–20% of
patient are unknown, it is recommended to titrate flu-
preoperative reference resting
ids using small iterative boluses (100 to 250 ml crystal-
values (agreement 100%).
loids over 5 to 10 min) guided by measurements of SV
variation [27] (Fig. 1). A SV increase > 10–12% (to avoid
Q2. Intraoperative fluids management: restrictive, being confounded by measurement variability) assessed
standard, liberal and beyond? one minute after the end of fluid infusion indicates that
Intravenous fluid administration is the most frequent the patient is able to increase flow and tissue perfusion
therapeutic intervention to maintain or restore tissue in response to fluids. A lack of increase of SV after fluid
perfusion during surgical procedures. However, it is well bolus is the most reliable indication that additional vol-
established that inadequate volume therapy can result in ume therapy may generate congestion and edema and
deleterious effects, especially in frail or high-risk patients. thus become deleterious. The smaller the volume admin-
Insufficient fluid administration will lead to reduced flow istered, the minimal the congestion resulting from the
and potentially inadequate perfusion in some territories unnecessary volume overload. The number of studies
where the conductive vessels are more resistive. These that have demonstrated the reduction in complications
territories may differ from one patient to another and can associated with a SV-guided fluid titration in high-risk
involve every organ. The reduced local blood flow can
induce cellular hypoxia and subsequent organ dysfunc-
tion or failure. On the other hand, if fluid is administered
in excess, venous congestion and edema will ensue. The
consequences of fluids in excess are probably as deleteri-
ous as the consequences of hypovolemia, and many stud-
ies have established a clear relation between positive fluid
balance and postoperative complications [23, 24]. Thus,
targeting the right amount of volume expander required
by each surgical patient during the procedure is a daily
challenge for practitioners. Recommendations are often
blurry with statements indicating to maintain “adequate
volemia” or “optimal volume”, which do not translate into
Fig. 1 Typical intraoperative goal-directed therapy algorithm based
quantitative meaningful information. Even worse, some on an individualized approach. MAP mean arterial pressure, PPV pulse
protocols suggest to administer an identical predefined pressure variation, SVV stroke volume variation. Values for PPV/SVV,
amount of fluids to everyone, taking into account the MAP and cardiac index are indicative and must be adapted on an
body weight and the duration of the surgical procedure, individual basis. The use of vasopressors could also be considered
when diastolic arterial pressure < 40 mmHg
assuming that the requirements and the tolerance are
Fellahi et al. Ann. Intensive Care (2021) 11:58 Page 4 of 10

surgical patients provides compelling evidence to sup- a transient rise in tidal volume from 6 to 8 ml/kg IBW
port that simple approach to minimize the deleterious could be useful to predict fluid responsiveness with sen-
side effects of intraoperative fluids [28, 29]. sitivity and specificity values > 90% [42]. As well, an abso-
lute decrease in PPV > 2% during a mini-fluid challenge
Experts’ opinion: 
We propose to routinely use a could help to predict fluid responsiveness [43]. Limita-
personalized approach of intra- tions to the interpretation of PPV also include cardiac
operative fluid infusion/volume arrhythmias, right and left ventricular failure, decreased
expansion based on the indi- lung compliance, and spontaneous breathing activity [32,
vidual hemodynamic response 40]. Importantly, the coelioscopic surgery-induced pneu-
to volume titration to reduce the moperitoneum is a frequent situation which decreases
deleterious side effects of fluids thoracic compliance, leading to changes in the interpre-
and improve patients’ outcome tation of PPV (a situation defined as false positive but this
requires a commentary: the patient is preload-dependent
(agreement 100%).
due to decreased venous return secondary to increased
abdominal pressure meaning that volume expansion will
Q3. Are dynamic indices and maneuvers useful
unfrequently correct fluid responsiveness) [44, 45]. Fur-
to predict and manage volume expansion?
ther studies are needed to determine the impact of an
Fluid responsiveness is defined as a significant increase
increased abdominal pressure between 10 and 15 mmHg
in blood flow in response to a fluid bolus [30]. Predict-
on PPV threshold values that should be considered to
ing fluid responsiveness is useful to identify patients who
identify intraoperative fluid responsiveness. Meanwhile,
may benefit from volume expansion and, more impor-
PPV must be interpreted with caution in that specific
tantly, to prevent fluid administration in non-responders
surgical setting or in patients in prone position. Finally,
[31]. Cardiac preload indices, such as central venous pres-
dynamic indices seem to predict fluid responsiveness
sure, have repeatedly been shown to be unreliable predic-
with insufficient accuracy in an open-chest condition
tors of fluid responsiveness. On the contrary, the arterial
during cardiac and/or thoracic surgery [46].
pulse pressure variation (PPV) induced by mechanical
Subsequently, it could be reasonable for routine prac-
ventilation is known as a sensitive and specific marker of
tice to implement a “validity criteria checklist” before
fluid responsiveness [32]. In patients receiving controlled
using PPV or similar approaches to estimate fluid
mechanical ventilation with a tidal volume ≥ 7–8 ml/
responsiveness (Table 2) [47]. When PPV cannot be used,
kg of ideal body weight (IBW), fluid responsiveness is
it remains possible to assess fluid responsiveness in sur-
very likely when PPV is > 13%, very unlikely when PPV
gical patients undergoing general anesthesia by measur-
is < 9% and uncertain when PPV ranges between 9 and
ing changes in SV during an end-expiratory occlusion
13% (grey zone of uncertainty) [33]. Importantly, tidal
test, a lung recruitment maneuver or during a mini-fluid
volumes of 7–8 ml/kg of IBW are consistent with recent
challenge [48, 49]. Out of the operating room, the most
recommendations for perioperative lung-protective ven-
validated maneuver is the passive leg raising test [50].
tilation [34]. Since the sentinel study by Lopes et al. [35],
The main limiting factor to the clinical adoption of those
numerous clinical studies using either PPV or the pulse
methods is the availability of a cardiac output monitor
contour-derived stroke volume variation (SVV) to indi-
to quantify SV changes (Table 1). The pleth variability
vidualize fluid therapy showed a decrease in postopera-
index (PVI), a non-invasive surrogate for PPV, may also
tive complications and hospital lengths of stay [36]. The
be useful to predict fluid responsiveness during sur-
estimation of PPV requires either invasive or non-inva-
gery [51]. Recently, the quantification of changes in the
sive recording of a continuous arterial pressure wave-
peripheral perfusion index (PI), a variable used as a sig-
form [37–39]. There are several limitations to the use of
nal quality indicator by most pulse oximeters, has been
PPV/SVV which have been described in detail elsewhere
proposed in an exploratory study to predict fluid respon-
[40]. For instance, protective mechanical ventilation is a
siveness with acceptable sensitivity and specificity [52].
potential obstacle to the use of PPV when very low tidal
However, monitors using finger cuff technologies have in
volumes are used (e.g. 6 ml/kg IBW or less) [41]. During
common the risk of poor reliability in cases of peripheral
very low tidal volume ventilation, a high PPV still sug-
hypoperfusion.
gests fluid responsiveness whereas a low PPV cannot rule
out fluid responsiveness. Therefore, alternative meth-
ods have been proposed to predict fluid responsiveness. Experts’ opinion: 
We propose to implement a
They include the assessment of PPV changes during a “validity criteria checklist” before
tidal volume challenge or during a mini-fluid challenge using PPV (or similar methods)
(Table 1). Thus, a 3.5% absolute increase in PPV during to estimate fluid responsiveness,
Fellahi et al. Ann. Intensive Care (2021) 11:58 Page 5 of 10

Table 1 Main advantages and limitations of dynamic indices and maneuvers developed to predict fluid responsiveness (chronologic
order from top to bottom)
Methods (year of first validation) Main advantages Main limitations

PPV (2000) Automatically calculated by most bedside Need for general anesthesia, an arterial line and a
monitors tidal volume > 7 ml/kg
SVV (2001) Automatically calculated by most CO monitors Need for a CO monitor, general anesthesia, an
arterial line and a tidal volume > 7 ml/kg
Changes in CO during a PLR maneuver (2006) Useful when Vt < 7 ml/kg IBW Need for a CO monitor, PLR maneuver difficult to
perform during surgery
PVI (2008) Non-invasive from a pulse ox Need for general anesthesia and a tidal vol‑
ume > 7 ml/kg, influenced by peripheral
perfusion
Changes in SV during an EEO test (2009) Useful when Vt < 7 ml/kg IBW Need for a CO monitor, prone to error measure‑
ments (small magnitude of changes in SV)
Changes in SV during a mini-fluid challenge Useful when Vt < 7 ml/kg IBW Need for a CO monitor, prone to error measure‑
(2011) ments (small magnitude of changes in SV)
Changes in PPV during a mini-fluid challenge Useful when Vt < 7 ml/kg IBW Prone to error measurements (small magnitude of
(2015) changes in PPV)
Changes in PPV during a Vt challenge (2017) Useful when Vt < 7 ml/kg IBW Prone to error measurements (small magnitude of
changes in PPV)
Changes in SV during a LRM (2017) Useful when Vt < 7 ml/kg IBW Need for a CO monitor
Changes in PI during a PLR maneuver (2019) Useful when Vt < 7 ml/kg IBW, non-invasive Influenced by peripheral perfusion, PLR maneuver
(pulse ox) difficult to perform during surgery
Low level of scientific validation
Changes in PI during a LRM (2020) Useful when Vt < 7 ml/kg IBW, non-invasive Influenced by peripheral perfusion
(pulse ox) Low level of scientific validation
CO cardiac output, EEO end-expiratory occlusion, IBW ideal body weight, LRM lung recruitment maneuver, PI perfusion index, PLR passive leg raising, PPV pulse
pressure variation, Pulse ox pulse oximetry, PVI pleth variability index, SV stroke volume, SVV stroke volume variation, Vt tidal volume

Q4. When should we measure stroke volume

Table 2 “Validity criteria checklist” before performing volume and cardiac output?
expansion based on pulse pressure variation/similar methods- While guidelines highly recommend the assessment of
provided information. Adapted from [47] CO/SV for perioperative fluid GDT in high-risk surgical
patients [27], less than one-third of patients in Europe
1 Is the patient ventilated with CMV without spontaneous efforts?
and in the United States actually benefit from CO moni-
2 Is the patient ventilated with tidal volume at least 7–8 ml/kg IBW?
toring during the perioperative period [53]. The main
3 Is the patient in closed-chest condition?
reasons reported by physicians are: (i) CO monitoring
4 Is the patient in sinus rhythm?
reference methods are too invasive; (ii) Non-invasive CO
5 Is lung compliance normal?
monitoring is unreliable; (iii) CO monitoring is useless
6 Is the patient unaffected by valvular disease?
to guide fluid optimization. Beyond classical knowledge
7 Is the patient unaffected by right/left ventricular dysfunction?
regarding accuracy and precision of CO/SV measure-
8 Does the patient have normal abdominal pressure?
ments, other criteria should now be considered to imple-
9 Is the HR/RR ratio ≥ 3.6?
ment a strategy of GDT in high-risk patients. These
10 Can you safely assess the efficacy of VE without HR or vasomotor
tone changes?
criteria are the mini-invasive or non-invasive approach
of the monitoring method; instantaneous and continuous
CMV: controlled mechanical ventilation; HR/RR: heart rate/respiratory rate ratio;
IBW: ideal body weight; VE: volume expansion information; automatized assessment without external
calibration; easy-to-use plug and play system; the absence
of both operator-dependence and learning curve; reason-
able cost; the ability to impact decision-making process
then to give iterative small fluid and outcomes. Thus, many devices using various tech-
boluses to maintain intraop- nologies have been commercially developed and sci-
erative PPV (or similar meth- entifically evaluated during the last 20 years (Table 3).
ods) below the threshold value Those devices are undoubtedly not interchangeable with
that defines fluid responsiveness reference methods (namely Fick principle, bolus ther-
(agreement 100%). modilution and echocardiography). However, it remains
Fellahi et al. Ann. Intensive Care (2021) 11:58 Page 6 of 10

Table 3 Invasive, mini-invasive, and non-invasive cardiac output monitoring technologies

Device Technology Properties Calibration

Cardio-Q (Gamida) Esophageal Doppler Mini-invasive, continuous, operator-dependent No

NICO (Novametrix) Fick principle applied to ­CO2 Non-invasive, discontinuous, operator-independent No
PiCCO (Getinge) Transpulmonary thermodilution Invasive, discontinuous, operator-independent Yes
Volume View (Edwards)
Niccomo (Imedex) Bioimpedance Non-invasive, continuous, operator-independent No
Physioflow (Manatec)
ECOM (ConMed)
NICOM (Baxter) Bioreactance Non-invasive, continuous, operator-independent No
Vigileo/Flotrac (Edwards) Pulse contour analysis Mini-invasive, continuous, operator-independent Optional
Pulsioflex/ProAQT (Getinge)
MostCare (Vygon)
LIDCOrapid (LIDCO Ltd)
ClearSight (Edwards) Digital photoplethysmography Non-invasive, continuous, operator-independent No
NICCI (Getinge)
esCCO (Nihon Kohden) Pulse wave transit time Non-invasive, continuous, operator-independent Optional

difficult to give a universal definition of what is actually (agreement 83%).

an acceptable agreement between a new method of CO Two experts pointed out problematic limits regard-
measurement and a reference method [54]. Moreover, ing accuracy of those mini- or non-invasive methods in
numerous studies and meta-analyses suggest that mini- patients with hypothermia or peripheral hypoperfusion or
invasive technologies such as esophageal Doppler [28] or receiving continuous infusion of vasopressors.
pulse contour analysis [55] are useful to guide GDT and
improve outcomes in high-risk surgical patients when
compared with a standard of care. Calibrated or uncali- Q5. Fluids or vasoconstrictors: how to decide?
brated pulse contour analysis methods especially seem Tissue perfusion means delivering enough oxygen, glu-
to have the favor of practitioners, being used in nearly cose and other metabolites to every cell, as well as clear-
75% of cases [56, 57]. Thus, beside reference methods ing the byproducts of cell metabolism. This is achieved
of CO measurement, often invasive, uneasy to imple- when the flow at the level of the microcirculation of
ment and clearly underused in the operating room, new each organ is “adequate”, a condition that is impossi-
mini-invasive and non-invasive technologies, while not ble to assess with our current monitoring capabilities.
interchangeable with the latter, could be useful to facili- Therefore, we concentrate our attention on the major
tate implementation of GDT and should probably be determinants that will most likely result in adequate
employed to improve outcomes in moderate to high-risk microcirculatory blood flow: MAP value that is suf-
surgical patients. Invasive reference methods are, how- ficient to drive blood flow through every organ and
ever, sometimes unavoidable, as in cardiac surgery or a systemic blood flow that is able to cover for the total
shock states for instance. Clinicians should actually be body oxygen requirements. Thus, ideally, both instanta-
aware that whatever the precision of the method used to neous arterial pressure and beat by beat systemic blood
estimate SV, the most important is to stop fluid admin- flow should be monitored in addition to heart rate to
istration as soon as SV no longer increase in response to “optimize” tissue perfusion. Diastolic arterial pressure is
fluid boluses. The majority of monitors used to estimate mainly determined by vascular tone [58]. Low diastolic
SV should be able to provide a clinically relevant answer arterial pressure could be likened to as a sign of vasodila-
to that crucial issue. tion (by lowering vascular tone) and it has been proposed
to introduce vasopressors in ICU patients when dias-
tolic arterial pressure is below 40 mmHg [59]. However,
Experts’ opinion: 
We propose to use mini- or non-
available data are conflicting in the perioperative setting
invasive continuous methods
[20]. When tissue perfusion is deemed inadequate, or
to monitor CO/SV rather than
when its “optimization” is attempted prophylactically in
invasive reference techniques to
a high-risk patient, the first therapeutic approach is usu-
implement perioperative GDT in ally based on intravenous fluid titration, as developed
moderate to high-risk patients above. In the meantime, if arterial pressure values are
undergoing non-cardiac surgery considered too low with respect to the “usual values” of
Fellahi et al. Ann. Intensive Care (2021) 11:58 Page 7 of 10

the patient, vasoconstrictors may be administered intra- Experts’ opinion: 

We propose to use fluids and
venously as small iterative boluses or as a continuous vasoconstrictors in combination
infusion to achieve the personalized pressure value con- via a GDT algorithm to simul-
sidered adequate [22, 60] (Fig. 1). taneously achieve perioperative
Despite the fact that anesthesia induction produces optimal blood flow and perfusion
vasodilation and reduces venous return via a reduction in pressure, avoid volume overload
effective volemia or stressed volume (as a consequence of and improve outcome in high-
the increased venous capacitance without any reduction risk surgery patients (agreement
in intravascular volume), physicians restore systemic flow 100%).
by giving intravenous fluids rather than vasoconstrictors.
This approach is empirical and does not seem to respond Q6. What is the economic impact of hemodynamic
to the primary determinant of reduced venous return, monitoring for GDT?
which is increased venous compliance. Rather, it is now Implementation of GDT and its additional hemody-
established that optimization of both flow and pressure is namic monitoring has a cost. This increased cost could
associated with improved outcomes in high-risk patients be perceived as an obstacle to hospital adoption. The
(Table 4). Fluids and vasoconstrictors must be used in cost of hemodynamic monitoring is highly variable, not
combination to achieve optimal blood flow and main- only from one technique to another, but also from one
tain pressure above the thresholds considered “at risk”. country to the other, depending on reimbursement poli-
Achieving optimal flow while avoiding venous congestion cies. Whatever such country-dependent pricing policies
cannot be done without the guidance of SV monitoring. of monitoring devices, there is a common denominator:
Continuous measurement of arterial pressure is the man- postoperative complications dramatically increase hospi-
datory counterpart for high-risk patients in whom even tal costs. Studies have shown that the average cost differ-
brief transient arterial hypotension can generate adverse ence between a patient with one or more complications
events. The incorporation in our monitoring devices of and a patient without any complication ranges between
artificial intelligence and machine-learning algorithms 10,000€ and 30,000€ [69, 70]. We personally found in 227
trained to detect changes in hemodynamics that precede patients undergoing colorectal surgery and scheduled
clinically apparent hypotension offers interesting per- for an ERAS program including GDT that the occur-
spectives [65]. Trials are currently ongoing to confirm rence of a complication had a mean extra-cost of 3,167€
that this technology can effectively reduce hypotension (additional treatment cost ≈1,030€ and prolonged length
during surgery and further improve patient outcome [66]. of stay ≈2,058€) which was not covered by the average
The main vasopressors used in the operating theatre 2,777€ additional public health system complication
are phenylephrine, ephedrine, and norepinephrine. Their reimbursement. The largest economic evaluation pub-
pharmacological effects are somewhat different but there lished so far (> 700 high-risk patients undergoing major
are no data on the superiority of one vasopressor over the gastrointestinal surgery), done in the UK with a pulse
others. Phenylephrine could, however, be responsible for contour technique, showed net savings around 400£/
a significant decrease in cardiac output and organ per- patient [71]. The reduction in postoperative morbid-
fusion, especially the brain [67] and in patients without ity is also often associated with a reduction in hospital
preload dependence [68]. length of stay [72]. In that case, the increase in the num-
ber of free beds may allow a boost in surgical activity, a
decreased wait times for patients, and increased revenue.
The « MERCI» equation [73] enables an easy estima-
Table 4 The impact of hemodynamic optimization on
tion of the possible Investment (I) to implement hemo-
postoperative morbidity: results from five meta-analyses in high-
dynamic monitoring at no net costs. It takes into account
risk surgical patients
the current morbidity rate (M), the expected reduction
References Number of studies Reduction in (ER) in postoperative morbidity, and the current cost (C)
(patients) postoperative
morbidity (%) of complications:

Hamilton [61] 29 (4,805) 57 M × ER × C = I.

Grocott [62] 31 (5,292) 32 As an example, if the morbidity rate after colorectal
Pearse [63] 22 (3,024) 23 surgery is 25% (M = 25%), the relative reduction in post-
Michard [55] a 19 (2,159) 54 operative morbidity is 23% (ER = 23%), and the average
Chong [64] 95 (11,659) 34 cost of complications per patient is 15,000€, the invest-
a
Non-calibrated pulse contour analysis only ment to implement hemodynamic monitoring at no net
Fellahi et al. Ann. Intensive Care (2021) 11:58 Page 8 of 10

cost is 863€/patient (0.25 × 0.23 × €15,000 = 863€). If Declarations

the cost of monitoring is less than 863€/patient (which
Ethics approval and consent to participate
is most often the case), the difference would be savings Not applicable.
to the health system. The higher the risk of postoperative
complications and the cost of those complications, the Consent for publication
Not applicable.
more favorable the economic impact of GDT will be. Sev-
eral studies have used the MERCI equation to predict the Competing interests
economic impact of hemodynamic monitoring in large JLF, EF, CV, OC, OH, JL, KA, BC, and DL are members of an advisory board work‑
ing for Edwards Lifesciences and have received honoraria from the Company
populations of patients undergoing major non-cardiac for their participation to the board. JLF and DL have received honoraria from
surgery [69, 74, 75]. They reported possible investments Masimo for participation to hemodynamic master classes. EF reported receiv‑
of around 500–1000€/patient. Since it may end up being ing consulting fees from Drager Medical, GE Healthcare, and Orion Pharma
and lecture fees from Fresenius Kabi, Baxter, and Fisher and Paykel Healthcare.
the main driver for adoption by hospital administration, EG has received lecture fees from Edwards Lifesciences and research support
the impact of hemodynamic monitoring on hospital prof- from Philips and Radiometer. MB has received lecture fees from Edwards Lifes‑
itability should be well known by physicians arguing for ciences and Maquet Pulsion. KA has received lecture fees from LFB, Fisher and
Paykel Healthcare, and Baxter. BT has no conflicts of interest to declare.
GDT in their institutions.
Author details
1
Service D’Anesthésie-Réanimation, Hôpital Louis Pradel, 59 boulevard Pinel,
Experts’ opinion: 
We propose that clinicians 69500 Hospices Civils de Lyon, Lyon, France. 2 Laboratoire CarMeN, Univer‑
explain to national/hospital sité Claude Bernard Lyon 1, Inserm U1060, Lyon, France. 3 Département de
decision-makers that the extra- Médecine Périopératoire, Anesthésie‑Réanimation, CHU de Clermont-Ferrand,
Clermont‑Ferrand, France. 4 Université Clermont Auvergne, CNRS; Inserm
cost due to hemodynamic moni- U1103, 63000 Clermont‑Ferrand, France. 5 Service D’Anesthésie-Réanimation,
toring when implementing a Hôpital Timone, AP-HM, Marseille, France. 6 Service D’Anesthésie-Réanimation,
Nouvel Hôpital Civil, Hôpitaux Universitaires de Strasbourg, Strasbourg,
perioperative GDT strategy is
France. 7 Université de Strasbourg, Strasbourg, France. 8 Département
counterbalanced by the reduc- D’Anesthésie‑Réanimation, CHRU de La Cavale Blanche, Brest, France.
tion in postoperative complica- 9
Université de Bretagne Occidentale, Brest, France. 10 Service de Chirurgie
Digestive, Groupe Hospitalier Paris Saint-Joseph, Paris, France. 11 Département
tions and hospital length of stay d’Anesthésie‑Réanimation, Hôpital Lariboisière, DMU PARABOL, AP-HP Nord
in high-risk surgery. This could et Université de Paris, Paris, France. 12 UMR-S 942, Inserm, Paris, France. 13 Pôle
be estimated at each institution d’Anesthésie-Réanimation, CHU Lille, Univ. Lille, ULR 2694–METRICS, Lille,
France. 14 Pôle d’Anesthésie-Réanimation, Hôpital Pellegrin, CHU de Bordeaux,
level using the MERCI equation Bordeaux, France. 15 Université de Bordeaux, France, Inserm 1034, Pessac,
(agreement 100%). France. 16 Service d’Anesthésie-Réanimation Chirurgicale, Pôle Anesthésie
Réanimations, Hôtel-Dieu, CHU de Nantes, Nantes, France. 17 Université de
Nantes, Nantes, France. 18 Service d’Anesthésie-Réanimation, Hôpital Européen
Georges Pompidou, AP-HP, Paris, France. 19 Université de Paris, Paris, France.
Conclusion 20
Inserm UMR S1140, Paris, France. 21 Département d’Anesthésie‑Réanimation,
Hôpital Bichat Claude Bernard, AP-HP Nord, Paris, France. 22 Université de Paris,
Because the current guidelines on perioperative hemo- Paris, France.
dynamic optimization are not explicit enough to allow
a reproducible decision-making process and also not Received: 5 October 2020 Accepted: 29 March 2021
suited to manage uncertainty, their implementation
remains limited in routine practice. Numerous relevant
questions are frequently asked by practitioners who need
References
explicit clinical reasoning tools and treatment plans to
1. Pearse RM, Moreno RP, Bauer P, Pelosi P, Metnitz P, Spies C, et al. European
develop strategies of blood pressure and cardiac output Surgical Outcomes Study (EuSOS) group for the Trials groups of the
monitoring and optimization on an individual basis and European Society of Intensive Care Medicine and the European Society of
at a reasonable cost. Anaesthesiology. Mortality after surgery in Europe: a 7 day cohort study.
Lancet. 2012;380:1059–65.
2. Som A, Maitra S, Bhattacharjee S, Baidya DK. Goal directed fluid therapy
decreases postoperative morbidity but not mortality in major non-car‑
Authors’ contributions diac surgery: a meta-analysis and trial sequential analysis of randomized
JLF, EF, CV, OC, OH, JL, EG, BT, MB, KA, BC, DL contributed to the content of this controlled trials. J Anesth. 2017;31:66–81.
paper and critically reviewed the final manuscript. The corresponding author 3. McNally SJ, MacKinnon M, Hawkins M. Practical barriers to the implemen‑
(JLF) edited individual contributions and finalized the manuscript. All authors tation of early goal directed therapy in the UK: trainee skills and aware‑
read and approved the final manuscript. ness. Scott Med J. 2009;54:22–4.
4. Messina A, Robba C, Calabro L, Zambelli D, Lannuzzi F, Molinari E, et al.
Funding Association between perioperative fluid administration and postop‑
None. erative outcomes: a 20-year systematic review and a meta-analysis of
randomized goal-directed trials in major visceral/noncardiac surgery. Crit
Availability of data and materials Care. 2021;25:43.
Not applicable.
Fellahi et al. Ann. Intensive Care (2021) 11:58 Page 9 of 10

5. Cabana MD, Rand CS, Powe NR, Wu AW, Wilson MH, Abboud PAC, et al. 24. Oh TK, Song IA, Do SH, Jheon S, Lim C. Association of perioperative
Why don’t physicians follow clinical practice guidelines? A framework for weight-based fluid balance with 30-day mortality and acute kidney
improvement. JAMA. 1999;282:1458–65. injury among patients in the surgical intensive care unit. J Anesth.
6. Barth JH, Misra S, Aakre KM, Langlois MR, Watine J, Twomey PJ, et al. 2019;33:354–63.
Why are clinical practice guidelines not followed? Clin Chem Lab Med. 25. Mythen MG, Swart M, Acheson N, Crawford R, Jones K, Kuper M, et al.
2016;54:1133–9. Perioperative fluid management: consensus statement from the
7. Sharma S, Pandit A, Tabassum F. Potential facilitators and barriers to enhanced recovery partnership. Perioper Med. 2012;1:1–4.
adopting standard treatment guidelines in clinical practice. Int J Health 26. Lamke LO, Nilsson GE, Reithner HL. Water loss by evaporation from the
Care Qual. 2017;30:285–98. abdominal cavity during surgery. Acta Chir Scand. 1977;143:279–84.
8. De Robertis E, Longrois D. To streamline the guideline challenge: The 27. Vallet B, Blanloeil Y, Cholley B, Orliaguet G, Pierre S, Tavernier B. Guide‑
European Society of Anaesthesiology policy on guidelines development. lines for perioperative haemodynamic optimization. Ann Fr Anesth
Eur J Anaesthesiol. 2016;33:794–9. Reanim. 2013;32:e151–8.
9. Hewson MG, Kindy PJ, Van Kirk J, Gennis VA, Day RP. Strategies for manag‑ 28. Phan K, Ismail H, Heriot AG, Ho KM. Improving perioperative outcomes:
ing uncertainty and complexity. J Gen Intern Med. 1996;11:481–5. fluid optimization with the esophageal Doppler monitor, a metaanaly‑
10. Fitch K, Bernstein SJ, Aguila MD. The RAND/UCLA Appropriateness sis and review. J Am Coll Surg. 2008;207:935–41.
Method User’s Manual. https://​www.​rand.​org/​pubs/​monog​raph_​repor​ 29. Calvo-Vecino JM, Ripolles-Melchor J, Mythen MG, Casans-Frances R,
ts/​MR1269.​html. Also available in print form. RAND Corporation. Santa Balik A, Artacho JP, et al. Effect of goal-directed haemodynamic therapy
Monica, CA; 2001. on postoperative complications in low-moderate risk surgical patients:
11. Bijker JB, Persoon S, Peelen LM, Moons KGM, Kalkman CJ, Kappelle LJ, a multicentre randomised controlled trial (FEDORA trial). Br J Anaesth.
et al. Intraoperative hypotension and perioperative ischemic stroke 2018;120:734–44.
after general surgery: a nested case-control study. Anesthesiology. 30. Michard F, Teboul JL. Predicting fluid responsiveness in ICU patients: a
2012;116:658–64. critical analysis of the evidence. Chest. 2002;121:2000–8.
12. Mathis MR, Naik BI, Freundlich RE, et al. Multicenter Perioperative 31. Michard F, Biais M. Rational fluid management: dissecting facts from
Outcomes Group Investigators. Preoperative Risk and the association fiction. Br J Anaesth. 2012;108:369–71.
between hypotension and postoperative acute kidney injury. Anesthesi‑ 32. Michard F. Changes in arterial pressure during mechanical ventilation.
ology. 2019;131:1046–62. Anesthesiology. 2005;103:419–28.
13. Walsh M, Devereaux PJ, Garg AX, Kurz A, Turan A, Rodseth RN, et al. 33. Cannesson M, Le Manach Y, Hofer CK, Goarin JP, Lehot JJ, Vallet B, et al.
Relationship between intraoperative mean arterial pressure and clinical Assessing the diagnostic accuracy of pulse pressure variations for the
outcomes after noncardiac surgery: toward an empirical definition of prediction of fluid responsiveness: a “gray zone” approach. Anesthesiol‑
hypotension. Anesthesiology. 2013;119:507–15. ogy. 2011;115:231–41.
14. Sessler DI, Sigl JC, Kelley SD, Chamoun NG, Manberg PJ, Saager L, et al. 34. Young CC, Harris EM, Vacchiano C, Bodnar S, Bukowy B, Elliott RRD,
Hospital stay and mortality are increased in patients having a “triple low” et al. Lung-protective ventilation for the surgical patient: international
of low blood pressure, low bispectral index, and low minimum alveolar expert panel-based consensus recommendations. Br J Anaesth.
concentration of volatile anesthesia. Anesthesiology. 2012;116:1195–203. 2019;123:898–913.
15. Abbott TEF, Pearse RM, Archbold RA, Ahmad T, Niebrzegowska E, Wragg 35. Lopes MR, Oliveira MA, Pereira VO, Lemos IPB, Auler JOC, Michard F.
A, et al. A prospective international multicenter cohort study of intra‑ Goal-directed fluid management based on pulse pressure variation
operative heart rate and systolic blood pressure and myocardial injury monitoring during high-risk surgery: a pilot randomized controlled
after noncardiac surgery: Results of the VISION Study. Anesth Analg. trial. Crit Care. 2007;11:R100.
2018;126:1936–46. 36. Benes J, Giglio M, Brienza N, Michard F. The effects of goal-directed
16. Sun LY, Wijeysundera DN, Tait GA, Beattie WS. Association of intraopera‑ fluid therapy based on dynamic parameters on post-surgical outcome:
tive hypotension with acute kidney injury after elective noncardiac a meta-analysis of randomized controlled trials. Crit Care. 2014;18:584.
surgery. Anesthesiology. 2015;123:515–23. 37. Biais M, Stecken L, Ottolenghi L, Roullet S, Quinart A, Masson F, et al.
17. Wesselink EM, Kappen TH, Torn HM, Slooter AJC, van Klei WA. Intraop‑ The ability of pulse pressure variation obtained with CNAP device to
erative hypotension and the risk of postoperative adverse outcomes: a predict fluid responsiveness in the operating room. Anesth Analg.
systematic review. Br J Anaesth. 2018;121:706–21. 2011;113:523–8.
18. Ichai C, Vinsonneau C, Souweine B, Armando F, Canet E, Clech C, et al. 38. Monnet X, Dres M, Ferré A, Le Teuff G, Jozwiak M, Bleibtreu A, et al.
Acute kidney injury in the perioperative period and in intensive care units Prediction of fluid responsiveness by a continuous non-invasive assess‑
(excluding renal replacement therapies). Anaesth Crit Care Pain Med. ment of arterial pressure in critically ill patients: comparison with four
2016;35:151–65. other dynamic indices. Brit J Anaesth. 2012;109:330–8.
19. Sessler DI, Bloomstone JA, Aronson S, Berry C, Gan TJ, Kellum JA, et al. 39. Stens J, Oeben J, Van Dusseldorp AA, Boer C. Non-invasive measure‑
Perioperative quality initiative consensus statement on intraoperative ments of pulse pressure variation nd stroke volume variation in anes‑
blood pressure, risk and outcomes for elective surgery. Br J Anaesth. thetized patients using the Nexfin blood pressure monitor. J Clin Monit
2019;122:563–74. Comput. 2016;30:587–94.
20. Ahuja S, Mascha EJ, Yang D, Maheshwari K, Cohen B, Khanna AK, et al. 40. Michard F, Chemla D, Teboul JL. Applicability of pulse pressure varia‑
Associations of intraoperative radial arterial systolic, diastolic, mean, tion: how many shades of grey? Crit Care. 2015;19:144.
and pulse pressures with myocardial and acute kidney injury after 41. De Backer D, Heenen S, Piagnerelli M, Koch M, Vincent JL. Pulse pres‑
noncardiac surgery: a retrospective cohort analysis. Anesthesiology. sure variation to predict fluid responsivenss: influence of tidal volume.
2020;132:291–306. Intensive Care Med. 2005;31:517–23.
21. Salmasi V, Maheshwari K, Yang D, Mascha EJ, Singh A, Sessler DI, et al. 42. Myatra SN, Prabu NR, Divatia JV, Monnet X, Kulkarni AP, Teboul JL. The
Relationship between intraoperative hypotension, defined by either changes in pulse pressure variation or stroke volume variation after a
reduction from baseline or absolute thresholds, and acute kidney and “tidal volume challenge” reliably predict fluid responsiveness during
myocardial injury after noncardiac surgery: a retrospective cohort analy‑ low tidal volume ventilation. Crit Care Med. 2017;45:415–21.
sis. Anesthesiology. 2017;126:47–65. 43. Mallat J, Meddour M, Durville E, Lemyze M, Pepy F, Temime J, et al.
22. Futier E, Lefrant JY, Guinot PG, Godet T, Lorne E, Cuvillon P, et al. INPRESS Decrease in pulse pressure and stroke volume variations after mini-
Study Group. Effect of individualized vs standard blood pressure manage‑ fluid challenge accurately predicts fluid responsiveness. Br J Anaesth.
ment strategies on postoperative organ dysfunction among high-risk 2015;115:449–56.
patients undergoing major surgery: a randomized clinical trial. JAMA. 44. Tokics L, Hedenstierna G, Brismar B, Strandberg A, Lundquist H. Thora‑
2017;318:1346–57. coabdominal restriction in supine men: CT and lung function measure‑
23. Chappell D, Jacob M, Hofmann-Kiefer K, Conzen P, Rehm M. A rational ments. J Appl Physiol. 1988;64:599–604.
approach to perioperative fluid management. Anesthesiology.
2008;109:723–40.
Fellahi et al. Ann. Intensive Care (2021) 11:58 Page 10 of 10

45. Joris J, Cigarini I, Legrand M, Jacquet N, De Groote D, Franchimont P, et al. 62. Grocott MP, Dushianthan A, Hamilton MA, Mythen MG, Harrison D, Rowan
Metabolic and respiratory changes after cholecystectomy performed via K. Optimisation Systematic Review Steering Group. Perioperative increase
laparotomy or laparoscopy. Br J Anaesth. 1992;69:341–5. in global blood flow to explicit defined goals and outcomes after surgery:
46. Piccioni F, Bernasconi F, Tramontano G, Langer M. A systematic review a Cochrane systematic review. Br J Anaesth. 2013;111:535–48.
of pulse pressure variation and stroke volume variation to predict fluid 63. Pearse RM, Harrison DA, MacDonald N, Gillies MA, Blunt M, Ackland G,
responsiveness during cardiac and thoracic surgery. J Clin Monit Comput. et al. Effect of a perioperative, cardiac output-guided hemodynamic ther‑
2017;31:677–84. apy algorithm on outcomes following major gastrointestinal surgery: a
47. Sondergaard S. Pavane for a pulse pressure variation defunct. Crit Care. randomized clinical trial and systematic review. JAMA. 2014;311:2181–90.
2013;17:327. 64. Chong MA, Wang Y, Berbenetz NM, McConachie I. Does goal-directed
48. Muller L, Toumi M, Bousquet PJ, Riu-Poulenc B, Louart G, Candela D, et al. haemodynamic and fluid therapy improve perioperative outcomes? A
An increase in aortic blood flow after an infusion of 100 ml colloid over 1 systematic review and meta-analysis. Eur J Anaesthesiol. 2018;35:469–83.
minute can predict fluid responsiveness: the mini-fluid challenge study. 65. Saugel B, Kouz K, Hoppe P, Maheshwari K, Scheeren TWL. Predicting
Anesthesiology. 2011;115:541–7. hypotension in perioperative and intensive care medicine. Best Pract Res
49. Guinot PG, Bernard E, Defrancq F, Petiot S, Majoub Y, Dupont H, et al. Clin Anaesthesiol. 2019;33:189–97.
Mini-fluid challenge predicts fluid responsiveness during spontaneous 66. Wijnberge M, Geerts BF, Hol L, Lemmers N, Mulder MP, Berge P, et al. Effect
breathing under spinal anaesthesia: an observational study. Eur J Anaes‑ of a machine learning-derived early warning system for intraoperative
thesiol. 2015;32:645–9. hypotension vs standard care on depth and duration of intraoperative
50. Monnet X, Cipriani F, Camous L, Sentenac P, Dres M, Krastinova E, et al. hypotension during elective noncardiac surgery: the HYPE Randomized
The passive leg raising test to guide fluid removal in critically ill patients. Clinical Trial. JAMA. 2020;323:1052–60.
Ann Intensive Care. 2016;6:46. 67. Larson S, Anderson L, Thomson S. Effect of phenylephrine on cer‑
51. Chu H, Wang Y, Sun Y, Wang G. Accuracy of pleth variability index to pre‑ ebral oxygen saturation and cardiac output in adults when used to
dict fluid responsiveness in mechanically ventilated patients: a systematic treat intraoperative hypotension: a systematic review. JBI Evid Synth.
review and meta-analysis. J Clin Monit Comput. 2016;30:265–74. 2020;19:34–58.
52. De Courson H, Biais M, Chavignier C. Do changes in perfusion index 68. Rebet O, Andremont O, Gérard JL, Fellahi JL, Hanouz JL, Fischer MO.
reflect changes in stroke volume during preload-modifying maneuvers? J Preload dependency determines the effects of phenylephrine on cardiac
Clin Monit Comput. 2020;25:65. output in anaesthetised patients: a prospective observational study. Eur J
53. Cannesson M, Pestel G, Ricks C, Hoeft A, Perel A. Hemodynamic monitor‑ Anaesthesiol. 2016;33:638–44.
ing and management in patients undergoing high risk surgery: a survey 69. Michard F, Mountford WK, Krukas MR, Ernst FR, Fogel SL. Potential return
among North American and European anesthesiologists. Crit Care. on investment for implementation of perioperative goal-directed fluid
2011;15:R197. therapy in major surgery: a nationwide database study. Perioper Med.
54. Saugel B, Cecconi M, Wagner JY, Reuter DA. Noninvasive continuous 2015;4:11.
cardiac output monitoring in perioperative and intensive care medicine. 70. Vonlanthen R, Slankamenac K, Breitenstein S, Puhan MA, Muller MK,
Br J Anaesth. 2015;114:562–75. Hahnloser D, et al. The impact of complications on costs of major surgical
55. Michard F, Giglio MT, Brienza M. Perioperative goal-directed therapy with procedures: a cost analysis of 1200 patients. Ann Surg. 2011;254:907–13.
uncalibrated pulse contour methods: impact on fluid management and 71. Sadique Z, Harrison DA, Grieve R, Rowan KM, Pearse RM. OPTIMISE study
postoperative outcome. Br J Anaesth. 2017;119:22–30. group. Cost-effectiveness of a cardiac output-guided haemodynamic
56. Ahmad T, Beilstein CM, Aldecoa C, Moreno RP, Molnar Z, Novak-Jankovic therapy algorithm in high-risk patients undergoing major gastrointestinal
V, et al. Variation in haemodynamic monitoring for major surgery in Euro‑ surgery. Perioper Med. 2015;4:13.
pean nations: secondary analysis of the EuSOS dataset. Perioper Med. 72. Flynn DN, Speck RM, Mahmoud NN, David G, Fleisher LA. The impact of
2015;4:8. complications following open colectomy on hospital finances: a retro‑
57. Myles PS, Bellomo R, Corcoran T, Forbes A, Peyton P, Story D, et al. Restric‑ spective cohort study. Perioper Med. 2014;3:1.
tive versus liberal fluid therapy for major abdominal surgery. N Engl J 73. Michard F. MERCI for improving quality of surgical care at no cost. World J
Med. 2018;378:2263–74. Surg. 2016;40:3095–6.
58. Lamia B, Chemla D, Richard C, Teboul JL. Clinical review: interpretation of 74. Landais A, Morel M, Goldstein J, Loriau J, Fresnel A, Chevalier C, et al.
arterial pressure wave in shock states. Crit Care. 2005;9:601–6. Evaluation of financial burdens following complications after major sur‑
59. Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, gery in France: potential returns after perioperative goal directed therapy.
et al. Surviving sepsis campaign: international guidelines for manage‑ Anaesth Crit Care Pain Med. 2017;36:151–5.
ment of severe sepsis and septic shock, 2012. Intensive Care Med. 75. Manecke G, Asemota A, Michard F. Tackling the economic burden of
2013;39:165–228. postsurgical complications: would goal directed fluid therapy help? Crit
60. Ackland GL, Brudney CS, Cecconi M, Ince C, Irwin MG, Lacey J, et al. Care. 2014;18:566.
Perioperative Quality Initiative consensus statement on the physiology
of arterial blood pressure control in perioperative medicine. Br J Anaesth.
2019;122:542–51. Publisher’s Note
61. Hamilton MA, Cecconi M, Rhodes A. A systematic review and meta- Springer Nature remains neutral with regard to jurisdictional claims in pub‑
analysis on the use of preemptive hemodynamic intervention to improve lished maps and institutional affiliations.
postoperative outcomes in moderate and high-risk surgical patients.
Anesth Analg. 2011;112:1392–402.