review www.kidney-international.

org

Fluid management in the critically ill

Jean-Louis Vincent1
1
Department of Intensive Care, Erasme Hospital, Université libre de Bruxelles, Brussels, Belgium

F
Fluid therapy, which is provided to restore and maintain luid management in critically ill patients is both a simple
tissue perfusion, is part of routine management for almost and a complex process. It can be considered “simple”
all critically ill patients. However, because either too much because inserting a catheter and allowing a solution to
or too little fluid can have a negative impact on patient run into a vein is really not a very sophisticated procedure.
outcomes, fluid administration must be titrated carefully Fluid administration represents one of the core interventions
for each patient. The “salvage, optimization, stabilization, in the management of acutely ill patients, with about one
de-escalation” (SOSD) mnemonic should be used as a third of patients in the intensive care unit (ICU) receiving
general guide to fluid resuscitation, and fluid resuscitation fluids on a specific day1 and many more
administration should be adapted according to the course receiving maintenance fluids. However, although adminis-
of the disease. In the initial salvage phase, lifesaving fluid tering fluids in critically ill patients is a common intervention,
should be administered generously. Once hemodynamic it is far from simple if it is performed correctly. Determining
monitoring is available, fluid administration should be the optimal type and amount of fluid for each patient at any
optimized by determining the patient’s fluid status and the specific moment in his or her ICU course is actually rather
need for further fluid. This determination can be difficult, complex. Because colloid fluids can be administered in
however; clinical indicators of hypovolemia, such as heart smaller amounts than can crystalloid fluids to produce the
rate, blood pressure, and urine output, may not detect same effect, the amount and type of fluid are related. How-
early hypovolemia, and edema is a late sign of fluid ever, for the sake of simplicity we will consider these 2 aspects
overload. Dynamic tests of fluid responsiveness such as separately.2
pulse pressure or stroke volume variation can be used in
only a small percentage of critically ill patients, and thus a Amount of fluid
fluid challenge technique is most frequently used to assess The amount of fluid required by any one patient depends on
ongoing fluid requirements. Once a patient has been several factors, including the patient’s diagnosis and severity
stabilized, efforts should start to concentrate on removing of disease. We need to find the ideal balance between too
excess fluid. Which fluid should be used remains a matter little and too much fluid in all patients, because both these
of some debate. Crystalloid solutions are cheaper than conditions are associated with worse outcomes3 (Figure 14).
colloid solutions, but colloid solutions remain in the Too little fluid leads to hypovolemia, which can result in
intravascular space for a longer period, making edema less inadequate cardiac output and thus decreased tissue perfu-
likely. Thus crystalloids and colloids should be used sion. Hypovolemia ultimately leads to multiple organ failure
together, especially in patients likely to require large fluid and death. However, too much fluid leads to hypervolemia,
volumes. Human albumin is a natural colloid and may have which can result in edema. Edema can result in altered or-
beneficial effects in patients with sepsis in addition to its gan function. Importantly, altered capillary permeability and
volume effects. Fluids should be prescribed as are other hypo-oncocity can contribute to edema formation even
medications, taking into account individual patient factors, when the effective circulating blood volume is not increased,
disease processes, and other treatments. which is why the term “fluid overload” should be used with
Kidney International (2019) 96, 52–57; https://doi.org/10.1016/ great caution.5 The effects of lung edema are perhaps the
j.kint.2018.11.047 most obvious clinically, with altered gas exchange rapidly
KEYWORDS: albumin; colloid; crystalloid; fluid challenge; hypovolemia apparent. However, edema in other organs can lead to
Copyright ª 2019, International Society of Nephrology. Published by delirium, abdominal compartment syndrome, and impaired
Elsevier Inc. All rights reserved. wound healing, to mention just a few possible consequences.
The kidney, being an encapsulated organ, is also sensitive to
the effects of edema. Increases in venous pressure have been
associated with the development of acute renal failure,6 and
fluid overload is associated with an increased risk of acute
renal failure7 and worse outcomes in patients with acute
Correspondence: Jean-Louis Vincent, Department of Intensive Care, Erasme renal failure.7,8 In the Sepsis Occurrence in Acutely Ill Pa-
University Hospital, Route de Lennik 808, B-1070 Brussels, Belgium. E-mail: tients study, a positive fluid balance was an independent
[email protected] predictor of poor outcome in patients with acute
Received 13 April 2018; revised 20 November 2018; accepted 28 renal failure (hazard ratio: 1.21; 95% confidence interval:
November 2018; published online 4 March 2019 1.13–1.28; P < 0.001).8

52 Kidney International (2019) 96, 52–57

J-L Vincent: Fluid management in the critically ill review

Important interactions between organs also can occur.9 Risk of

For example, altered cardiac function can further reduce complications
tolerance to fluids, with increases in cardiac filling pressures.
In intensive care medicine, the interaction between the lungs
and the kidneys is often an important consideration
Hypovolemia Hypervolemia
(Figure 2). Generally it is preferable to keep the lungs dry to
limit lung edema and preserve gas exchange, but the kidneys Altered tissue perfusion Edema
need to be well perfused to maintain function. Whereas the Renal failure Intra-abdominal hypertension
Anastomosis leakage Respiratory failure
lungs do not suffer much from fluid restriction, the kidneys Confusion, risk of CVA Impaired healing
do suffer from fluid restriction. Splanchnic ischemia Altered mobilization
Unfortunately one cannot use global formulas to decide MOF MOF
how much fluid a patient requires; this process requires
Volume
individualization guided by hemodynamic monitoring. status
Moreover, the time factor is of great importance and varies Figure 1 | The relationship between blood volume and the risk
according to the phase of resuscitation. Four phases can be of complications. Both hypo- and hypervolemia should be
recognized: salvage, optimization, stabilization, and de- avoided because these 2 extremes are associated with a greater
escalation.10 risk of complications. CVA, cerebrovascular accident; MOF,
multiple organ failure. Modified from Vincent JL, Pelosi P, Pearse R,
Salvage. Salvage is the initial phase during which a pa- et al. Perioperative cardiovascular monitoring of high-risk patients:
tient needs rapid resuscitation for profound alterations in a consensus of 12. Crit Care. 2015;19:224.4 Copyright ª 2015
tissue perfusion as a result of multiple conditions, including Vincent et al.; licensee BioMed Central. 2015. This is an Open
sepsis, hemorrhage, and profound dehydration. During this Access article distributed under the terms of the Creative Com-
mons Attribution License (http://creativecommons.org/licenses/
very early phase, a plentiful amount of fluids must be by/4.0), which permits unrestricted use, distribution, and repro-
administered rapidly to facilitate effective resuscitation. Often duction in any medium, provided the original work is properly
the need to act quickly means there is no time to position credited.
complex monitoring systems to guide fluid administration.
The Surviving Sepsis Campaign guidelines were strongly
criticized for recommending administration of 30 ml/kg of changes in patients and thus to establish who may benefit
fluid over 3 hours, which is too long a period. The updated 1- from these therapies; the techniques of weighing, measuring
hour Surviving Sepsis Campaign bundle11 is still suboptimal, fluid balance using in-out charts, and biomarkers are all
because it only states that one should “begin” administration unreliable. When vasopressors are used, norepinephrine is
of fluid within the first hour. One should rather consider the vasopressor of choice in critically ill patients.17
giving 500 ml to 1 L of fluid over a short period, definitely less Optimization. As soon as is possible, in addition to per-
than 1 hour. Authors of a recent study reported increased forming a reliable assessment of heart rate and arterial pres-
mortality rates in patients with sepsis who received initial sure, an echocardiogram should be obtained to assess the
fluid resuscitation more than 2 hours after diagnosis.12 As degree of filling and function of the cardiac chambers,
additional monitoring devices are utilized and we move into because clinical assessment of ventricular function is often
the next phase of optimization, the amount of fluid admin- inaccurate.18 Echography should be repeated periodically to
istered can quickly be individualized to the patient’s needs. monitor the response to treatment. Every physician involved
Importantly, in patients with severe hypotension, vaso- in the management of acutely ill patients should be able to
pressor therapy should be started along with fluid admin- perform an echocardiogram, and mobile point-of-care de-
istration because a clear relation exists between the vices are now widely available.19 Blood lactate levels, which
magnitude and duration of hypotension and patient out- are increased in patients with shock and reflect tissue
comes.13,14 Moreover, in patients with septic shock, hypo- oxygenation, should be measured serially (at least every hour)
tension is in large part the result of decreased vascular tone, as an indicator of the effectiveness of therapy and resolution
which is not corrected by fluid administration alone.15 of shock.20
Indeed, it is often difficult to determine the prevalent In terms of fluid administration, the optimization phase
pathophysiologic process underlying the hemodynamic consists of the administration of a limited amount of i.v. fluid,
failure or shock, which can include some degree of increased with careful evaluation of the patient’s response. Signs of fluid
endothelial permeability (in which case fluids are required) responsiveness before any additional fluid is given can be very
and/or vasodilation (requiring vasopressor agents). Thera- valuable to avoid giving fluids to patients who may not benefit
peutic strategies that can decrease permeability alterations from them. Importantly, such tests of fluid responsiveness
by protecting the endothelium are being studied, but should be used only if it is unclear whether a patient needs
currently fluid administration is the only effective option. fluids. If a patient clearly needs fluid—for example, if he or
Vasopressin (and vasopressin derivatives), interferon-b, and she has acute bleeding or hypotension that recurs as soon as
thrombomodulin are potential candidates for endothelial the fluid administration rate is slowed—then time should not
cell protection,16 but it is difficult to evaluate permeability be wasted in assessing fluid responsiveness. However, in

Kidney International (2019) 96, 52–57 53

review J-L Vincent: Fluid management in the critically ill

Renal dysfunction
Table 1 | Requirements for use of pulse pressure variation and
stroke volume variation to assess fluid responsiveness
Pre-renal Renal
 Mechanical ventilation
Maximize fluids Maximize fluids  Relatively large tidal volume
+ vasoactive agents? + vasoactive agents?  No spontaneous breathing activity
 No major arrhythmia
 No significant tachypnea
 No right ventricular failure
 No intra-abdominal hypertension
Hemodynamic Nonhemodynamic
Minimize fluids Minimize fluids
+ vasoactive agents? + vasoactive agents?
changes, close monitoring is required and the patient must
Lung edema
receive no other interventions during the procedure.
Figure 2 | The lung/kidney dilemma during fluid resuscitation. Fluid responsiveness also can be assessed by an “internal” or
Restricting fluid administration may reduce the risk of lung “endogenous” fluid challenge using a passive leg raising test,
edema but also can reduce renal perfusion, leading to renal
dysfunction. However, providing sufficient fluid to maintain renal which results in transfer of blood from the legs to the intra-
function can result in pulmonary edema with reduced gas thoracic compartment. This technique is more complex than
exchange. simply “raising the legs.”25 Importantly, one cannot use an
arterial pressure measurement alone to monitor the response
patients for whom the added benefit of fluid is unclear, to passive leg raising; stroke volume needs to be monitored
assessment of fluid responsiveness can be useful. Several closely in a quiet environment, because the result is dependent
techniques have been proposed for this purpose. In patients on detecting a transient increase in stroke volume.
undergoing mechanical ventilation, changes in the arterial Whichever technique is used, it must be remembered that
pulse pressure (pulse pressure variation) during the respira- the presence of fluid responsiveness does not necessarily
tory cycle can reflect the sensitivity of cardiac filling. Likewise, mean that fluids should be administered (indeed, healthy
some hemodynamic monitoring systems assess the variation persons are fluid responsive), but lack of fluid responsiveness
in stroke volume during the respiratory cycle. However, these suggests that fluid administration should be stopped, similar
measures have several limitations, which are listed in Table 1. to the situation in which there is no response to a fluid
The most important limitation is that these measures are only challenge. A simple algorithm to evaluate the need for fluid is
reliable during controlled mechanical ventilation, when the presented in Figure 3.
patient does not trigger the ventilator. This restriction may be Stabilization. Importantly, the optimization methods
possible when anesthesia is induced in the operating room previously described better serve to identify when to stop
but not under other conditions, because we try to avoid use of fluids than when to start them. Once the patient is stable, or
sedative agents as much as possible in ICUs today.21 In when the patient no longer responds to fluid, aggressive fluid
addition, the use of low tidal volumes is recommended today administration must be stopped, and only minimal, if any,
not only in patients with acute respiratory distress syndrome maintenance fluid is required. Too often, clinicians continue
but in all patients undergoing mechanical ventilation, but the to administer fluids when they are no longer indicated. Per-
use of low tidal volumes decreases the magnitude of the sisting positive fluid balance is associated with worse out-
signal, making it less reliable. comes,26 more so than a positive fluid balance early during
Given the limitations of these signs of fluid responsiveness, resuscitation. In the large Intensive Care Over Nations study
central venous pressure (CVP) measurement can still be of 1808 patients with sepsis in the ICU, it was observed that a
helpful. Importantly, a single measurement of CVP may not higher cumulative fluid balance at 72 hours after ICU
provide much information about the patient’s response to admission but not during the first 24 hours was indepen-
fluid, and changes in CVP values should be measured during dently associated with an increase in the hazard of death.26 If
a fluid challenge to evaluate the dynamic response to fluid.22 fluid removal can be accomplished without needing to in-
The Surviving Sepsis Campaign guidelines23 recommend that crease doses of vasopressor agents, this may be attempted, but
fluid challenge be the first method used to guide fluid withdrawing fluid while increasing the dose of vasopressors
administration in patients with sepsis. Nevertheless, fluid can be dangerous.
challenges must be conducted correctly, with a predefined De-escalation. When the patient is well stabilized, he or
amount of fluid (100 to 200 ml) given as a slow bolus while she enters the de-escalation phase and should not receive
monitoring the response to ensure that preset safety limits, much fluid; rather, excess fluids should be eliminated to
such as a specific CVP value, are not exceeded. In a patient remove any edema that has formed. During this phase, di-
with hypovolemia, the fluid challenge will result in an in- uretics can be administered if a negative fluid balance is not
crease in cardiac output with a small increase in CVP; in a achieved spontaneously. Loop diuretics, such as furosemide,
patient with adequate or excessive filling, however, the CVP are most widely used but can lead to hypernatremia if sodium
will increase considerably while the cardiac output remains excretion is not adequate.27 Although diuretics have been
unchanged.24 Importantly, to correctly interpret these associated with worse outcomes in patients with acute kidney

54 Kidney International (2019) 96, 52–57

J-L Vincent: Fluid management in the critically ill review

Problem
Crystalloids
that could be alleviated by fluid Crystalloids
+ colloids
administration (hypotension,
tachycardia, or oliguria) 1.5

Very low
cardiac filling Intermediate/high
(pressures or volumes) cardiac filling 1
and/or (pressures or volumes)
context of hypovolemia
(bleeding, diarrhea, ...)

Controlled mechanical ventilation

(deep sedation — no triggering) No 0.5
No major arrhythmia
No RV failure Fluid
Fluid challenge
loading Yes

Pulse pressure variation (PPV) Internal 0

Stroke volume variation (SVV) External
(passive leg
End-expiratory pause raising) Figure 4 | Differences in the amount of fluid required if using
crystalloid solutions alone or when combining crystalloid
solution with some colloid solution.
Figure 3 | A simple algorithm for fluid administration in acutely
ill patients. RV, right ventricular. meta-analysis of prospective randomized controlled trials
(RCTs) in various patient populations indicated that the
injury, in patients with excess fluid and acute kidney injury, amount of fluid could be reduced by one third if colloids were
diuretic use may still be beneficial by creating a negative fluid given in addition to crystalloids, instead of giving crystalloids
balance.28 Nevertheless, in patients with renal dysfunction, alone.2 In a large pragmatic clinical trial comparing colloids
diuretics are less effective and renal replacement therapy with crystalloids, the same ratio of 1/1.5 was reported.31
(RRT) will likely be necessary. In a recent survey of Australia, These differences are illustrated in Figure 4 and are clini-
New Zealand, and United Kingdom intensivists, RRT was cally relevant. As just one example, abdominal compartment
preferred over diuretic therapy if a fluid-overloaded critically syndrome is almost always associated with administration of
ill patient was oliguric or anuric or if he or she had a sig- large amounts of crystalloid solutions.32 For many years, the
nificant degree of acute kidney injury.29 Few published data debate was considered to be about “crystalloids versus col-
support the use of pharmacologic or mechanical methods of loids,” but the real debate is actually “crystalloids alone versus
fluid removal in critically ill patients. In patients treated with crystalloids plus colloids.” Indeed, colloids are rarely admin-
RRT, ultrafiltration can be used but should be performed istered alone, and an excessive increase in colloid osmotic
prudently to allow equilibration between the intravascular pressure may alter renal function, supporting the additional
and extravascular volumes. If performed too aggressively, the use of crystalloid solutions when colloids are administered.33
ultrafiltration can result in relative hypovolemia with the Numerous studies have attempted to compare outcomes in
associated risk of altered tissue perfusion. Importantly, the patients treated with crystalloids versus outcomes in patients
impact on perfusion may not be easily recognized clinically receiving colloids. A meta-analysis of 59 RCTs indicated that
because hypotension is a relatively late sign of hypovolemia there was no increased risk of mortality in patients who
and cardiac output or central venous oxygen saturation may received colloids compared with crystalloids.34 In patients with
not be regularly monitored in these already stable patients. It trauma, colloids seemed to decrease the incidence of acute
is difficult to monitor changes in blood volume during RRT; renal failure by about 50% (odds ratio: 0.46 [95% confidence
some systems are available during hemodialysis, but they are interval: 0.23–0.90]), whereas in critically ill patients or pa-
not reliable in critically ill patients. Unfortunately, no tients with sepsis, colloids were associated with an increased
chemical biomarkers apart from blood lactate (discussed risk of acute renal failure compared with crystalloids (odds
earlier) are available that can reliably assess tissue perfusion or ratio: 1.24 [95% confidence interval: 1.09–1.41]).34
fluid status. B-type natriuretic protein has been proposed but Administration of albumin solution, the only natural
is unreliable in the critically ill patient population.30 colloid fluid, can have a place in fluid management and may
even decrease mortality in patients with septic shock.35,36 In
Type of fluid the Albumin Italian Outcome Sepsis open-label randomized
With their large molecular weight, colloids persist longer in trial, the efficacy of saline solution with and without
the intravascular space, and thus the addition of colloids to administration of 20% albumin solution was studied in 1818
fluid management can decrease the total amount of fluid patients with sepsis. Although there were no significant dif-
required and reduce the risk and amount of edema. A ferences in overall survival after 28 or 90 days, in the

Kidney International (2019) 96, 52–57 55

review J-L Vincent: Fluid management in the critically ill

subgroup of 1121 patients with septic shock, albumin was hypotonic and should be avoided, especially in patients with
associated with a significant survival advantage after 90 days neurologic injury. Plasma-Lyte solutions contain acetate,
(relative risk of death: 0.87; 95% confidence interval: 0.77– which may have its own adverse effects. Davies et al.47 showed
0.99).37 In addition to its volume-expanding qualities, albu- that the use of Plasma-Lyte as the circuit priming fluid during
min has other attributes, including antioxidant effects and cardiopulmonary bypass was associated with supra-
drug-carrying properties. Hydroxyethyl starch (HES) solu- physiological plasma concentrations of acetate. My colleagues
tions may represent a cheaper alternative, although the and I showed that the use of baths containing acetate during
financial advantage is quite limited in the United States. The dialysis was associated with a decrease in myocardial contrac-
potentially harmful renal effects of HES solutions have been tility,48 and acetate was replaced by bicarbonate in dialysis baths
discussed,38 but the data remain controversial, especially a long time ago. However, it is unclear whether the smaller
because the largest study, the Crystalloid versus Hydroxyethyl amounts of acetate received when Plasma-Lyte is used as a
Starch Trial, showed that although patients who received HES resuscitation fluid have similar potentially harmful effects.49
solutions were more likely to need RRT than were those who Hence i.v. fluids should be considered as drugs. Each fluid
received saline solution, the incidence of renal dysfunction as can have adverse effects, and there is no optimal fluid solution
expressed by the “Risk, Injury, Failure, Loss of kidney func- for all circumstances. Choices of fluid combinations in indi-
tion, and End-stage kidney disease” criteria was significantly vidual patients should be made according to various factors,
lower in the HES group than in the saline solution group.39 including likely fluid requirements, underlying diagnoses,
Intriguingly, against the advice of the jury of experts that hemodynamic status, and laboratory results. It is important to
they had themselves constituted, the safety committee have a variety of solutions available and to be cautious before
(Pharmacovigilance Risk Assessment Committee) of the Eu- banning the use of any specific type of solution.
ropean Medicines Agency proposed that HES solutions
should no longer be available in Europe.40 The discussion CONCLUSION
about revising the package insert of HES solutions is still Optimal fluid management in the acutely ill patient remains a
ongoing. Because of their smaller molecular weight, gelatin challenge. Although fluid administration initially may appear
solutions are less effective and are not available in the United simple, it is actually quite complex and dependent on the
States. An ongoing multicenter RCT in Europe is comparing pathophysiology underlying the hemodynamic instability.
gelatin-based resuscitation with balanced crystalloid-based Intravenous fluids should be treated and prescribed as drugs,
resuscitation (ClincialTrials.gov identifier NCT02715466). with the type and dosage adapted to each patient’s individual
Dextran solutions have been almost entirely abandoned. ongoing needs and tolerance.
Among the crystalloid solutions, saline solutions, which
include large amounts of chloride (154 mEq/l), induce
DISCLOSURE
hyperchloremic acidosis, and evidence shows that hyper-
The author declared no competing interests.
chloremia may alter renal function by impairing internal
hemodynamics. This finding was first demonstrated in the
classic study by Wilcox in 1983.41 Hence chloride levels must REFERENCES
1. Finfer S, Liu B, Taylor C, et al. Resuscitation fluid use in critically ill adults:
be monitored during infusion and infusion of saline solutions an international cross-sectional study in 391 intensive care units. Crit
must be stopped if hyperchloremia develops. I believe it is Care. 2010;14:R185.
unfortunate that large RCTs are being conducted to compare 2. Orbegozo Cortes D, Gamarano Barros T, Njimi H, Vincent JL. Crystalloids
versus colloids: exploring differences in fluid requirements by systematic
saline solution with balanced solutions,42 because they may review and meta-regression. Anesth Analg. 2015;120:389–402.
expose patients in the saline solution group to excessive risk. 3. Balakumar V, Murugan R, Sileanu FE, et al. Both positive and negative
More specifically, in such a trial, if the saline solution is fluid balance may be associated with reduced long-term survival in the
critically ill. Crit Care Med. 2017;45:e749–e757.
stopped when hyperchloremia develops, no differences will be 4. Vincent JL, Pelosi P, Pearse R, et al. Perioperative cardiovascular
found in outcomes between study arms, and if it is continued, monitoring of high-risk patients: a consensus of 12. Crit Care. 2015;19:224.
there is almost no doubt that saline solution will be found to 5. Vincent JL, Pinsky MR. We should avoid the term "fluid overload.". Crit
Care. 2018;22:214.
be harmful. Indeed, there is no evidence that hyperchloremia
6. Legrand M, Dupuis C, Simon C, et al. Association between systemic
is associated with any beneficial effect; rather, it has been hemodynamics and septic acute kidney injury in critically ill patients: a
shown to be associated with worse outcomes.43 The recent retrospective observational study. Crit Care. 2013;17:R278.
Isotonic Solutions and Major Adverse Renal Events Trial 7. Wang N, Jiang L, Zhu B, et al. Fluid balance and mortality in critically ill
patients with acute kidney injury: a multicenter prospective
(SMART) and Saline against Lactated Ringer’s or Plasma-Lyte epidemiological study. Crit Care. 2015;19:371.
in the Emergency Department (SALT-ED) RCTs indicated an 8. Payen D, de Pont AC, Sakr Y, et al. A positive fluid balance is associated
increase in major adverse kidney events—a composite of with a worse outcome in patients with acute renal failure. Crit Care.
2008;12:R74.
death, new RRT, or persistent renal dysfunction—with saline 9. Lelubre C, Vincent JL. Mechanisms and treatment of organ failure in
solutions.44,45 This type of RCT thus raises ethical questions sepsis. Nat Rev Nephrol. 2018;14:417–427.
because there is no equipoise.46 10. Vincent JL, De Backer D. Circulatory shock. N Engl J Med. 2013;369:1726–
1734.
Use of balanced solutions also can be criticized. Ringer’s 11. Levy MM, Evans LE, Rhodes A. The Surviving Sepsis Campaign bundle:
lactate solutions (also called Hartmann’s solutions) are 2018 update. Crit Care Med. 2018;46:997–1000.

56 Kidney International (2019) 96, 52–57

J-L Vincent: Fluid management in the critically ill review

12. Leisman DE, Doerfler ME, Schneider SM, et al. Predictors, prevalence, and 31. Annane D, Siami S, Jaber S, et al. Effects of fluid resuscitation with
outcomes of early crystalloid responsiveness among initially hypotensive colloids vs crystalloids on mortality in critically ill patients presenting
patients with sepsis and septic shock. Crit Care Med. 2018;46:189–198. with hypovolemic shock: the CRISTAL randomized trial. JAMA. 2013;310:
13. Bai X, Yu W, Ji W, et al. Early versus delayed administration of 1809–1817.
norepinephrine in patients with septic shock. Crit Care. 2014;18:532. 32. Balogh ZJ, Lumsdaine W, Moore EE, Moore FA. Postinjury abdominal
14. Vincent JL, Nielsen ND, Shapiro NI, et al. Mean arterial pressure and compartment syndrome: from recognition to prevention. Lancet.
mortality in patients with distributive shock: a retrospective analysis of 2014;384:1466–1475.
the MIMIC-III database. Ann Intensive Care. 2018;8:107. 33. Roche AM, James MF. Colloids and crystalloids: does it matter to the
15. Hamzaoui O, Scheeren TWL, Teboul JL. Norepinephrine in septic shock: kidney? Curr Opin Crit Care. 2009;15:520–524.
when and how much? Curr Opin Crit Care. 2017;23:342–347. 34. Qureshi SH, Rizvi SI, Patel NN, Murphy GJ. Meta-analysis of colloids versus
16. Vincent JL, Grimaldi D. Novel interventions—what’s new and the future. crystalloids in critically ill, trauma and surgical patients. Br J Surg.
Crit Care Clin. 2018;34:161–173. 2016;103:14–26.
17. De Backer D, Biston P, Devriendt J, et al. Comparison of dopamine and 35. Vincent JL, De Backer D, Wiedermann CJ. Fluid management in
norepinephrine in the treatment of shock. N Engl J Med. 2010;362:779– sepsis: the potential beneficial effects of albumin. J Crit Care.
789. 2016;35:161–167.
18. Sekiguchi H, Harada Y, Villarraga HR, et al. Focused cardiac ultrasound in 36. Rochwerg B, Alhazzani W, Sindi A, et al. Fluid resuscitation in sepsis: a
the early resuscitation of severe sepsis and septic shock: a prospective systematic review and network meta-analysis. Ann Intern Med. 2014;161:
pilot study. J Anesth. 2017;31:487–493. 347–355.
19. Michard F. A sneak peek into digital innovations and wearable sensors 37. Caironi P, Tognoni G, Masson S, et al. Albumin replacement in patients
for cardiac monitoring. J Clin Monit Comput. 2017;31:253–259. with severe sepsis or septic shock. N Engl J Med. 2014;370:1412–1421.
20. Vincent JL, Quintairos E Silva A, Couto L Jr, Taccone FS. The value of 38. Schetz M, Shaw AD, Vincent JL. Is the literature inconclusive about the
blood lactate kinetics in critically ill patients: a systematic review. Crit harm of HES? We are not sure. Intensive Care Med. 2017;43:1526–1528.
Care. 2016;20:257. 39. Myburgh JA, Finfer S, Bellomo R, et al. Hydroxyethyl starch or saline for
21. Vincent JL, Shehabi Y, Walsh TS, et al. Comfort and patient-centred care fluid resuscitation in intensive care. N Engl J Med. 2012;367:1901–1911.
without excessive sedation: the eCASH concept. Intensive Care Med. 40. Annane D, Fuchs-Buder T, Zoellner C, et al. EMA recommendation to
2016;42:962–971. suspend HES is hazardous. Lancet. 2018;391:736–738.
22. De Backer D, Vincent JL. Should we measure the central venous pressure 41. Wilcox CS. Regulation of renal blood flow by plasma chloride. J Clin
to guide fluid management? Ten answers to 10 questions. Crit Care. Invest. 1983;71:726–735.
2018;22:43. 42. Vincent JL, De Backer D. Saline versus balanced solutions: are clinical
23. Rhodes A, Evans LE, Alhazzani W, et al. Surviving Sepsis Campaign: trials comparing two crystalloid solutions really needed? Crit Care.
international guidelines for management of sepsis and septic shock: 2016;20:250.
2016. Crit Care Med. 2017;45:486–552. 43. Stenson EK, Cvijanovich NZ, Anas N, et al. Hyperchloremia is associated
24. Vincent JL, Weil MH. Fluid challenge revisited. Crit Care Med. 2006;34: with complicated course and mortality in pediatric patients with septic
1333–1337. shock. Pediatr Crit Care Med. 2018;19:155–160.
25. Monnet X, Teboul JL. Passive leg raising: five rules, not a drop of fluid! 44. Semler MW, Self WH, Wanderer JP, et al. Balanced crystalloids versus
Crit Care. 2015;19:18. saline in critically ill adults. N Engl J Med. 2018;378:829–839.
26. Sakr Y, Rubatto Birri PN, Kotfis K, et al. Higher fluid balance increases the 45. Self WH, Semler MW, Wanderer JP, et al. Balanced crystalloids versus
risk of death from sepsis: results from a large international audit. Crit Care saline in noncritically ill adults. N Engl J Med. 2018;378:819–828.
Med. 2017;45:386–394. 46. Vincent JL, De Backer D. We do not appreciate SALT. Am J Respir Crit Care
27. Bellomo R. Issue and challenges of fluid removal in the critically ill. Br J Med. 2018;197:1361.
Anaesth. 2014;113:734–735. 47. Davies PG, Venkatesh B, Morgan TJ, et al. Plasma acetate, gluconate and
28. Grams ME, Estrella MM, Coresh J, et al. Fluid balance, diuretic use, and interleukin-6 profiles during and after cardiopulmonary bypass: a
mortality in acute kidney injury. Clin J Am Soc Nephrol. 2011;6:966–973. comparison of Plasma-Lyte 148 with a bicarbonate-balanced solution.
29. O’Connor ME, Jones SL, Glassford NJ, et al. Defining fluid removal in the Crit Care. 2011;15:R21.
intensive care unit: a national and international survey of critical care 48. Vincent JL, Vanherweghem JL, Degaute JP, et al. Acetate-induced
practice. J Intensive Care Soc. 2017;18:282–288. myocardial depression during hemodialysis for acute renal failure. Kidney
30. Levitt JE, Vinayak AG, Gehlbach BK, et al. Diagnostic utility of B-type Int. 1982;22:653–657.
natriuretic peptide in critically ill patients with pulmonary edema: a 49. Semler MW, Kellum JA. Balanced crystalloid solutions. Am J Respir Crit
prospective cohort study. Crit Care. 2008;12:R3. Care Med. 2019;199:952–960.

Kidney International (2019) 96, 52–57 57