Journal of Critical Care 35 (2016) 161–167

Contents lists available at ScienceDirect

Journal of Critical Care

journal homepage: www.jccjournal.org

Fluid management in sepsis: The potential beneficial effects of albumin☆

Jean Louis Vincent, MD, PhD a,⁎, Daniel De Backer, MD, PhD b, Christian J. Wiedermann, MD c
a
Department of Intensive Care, Erasme Hospital, Université Libre de Bruxelles, 1070 Brussels, Belgium
b
Department of Intensive Care, CHIREC Hospitals, Université Libre de Bruxelles, B-1420 Braine L'Alleud, Brussels, Belgium
c
Department of Internal Medicine, Central Hospital of Bolzano, 39100 Bolzano, Bozen, Italy

a r t i c l e i n f o a b s t r a c t

Keywords: Fluid administration is a key intervention in hemodynamic resuscitation. Timely expansion (or restoration) of
Albumin plasma volume may prevent tissue hypoxia and help to preserve organ function. In septic shock in particular,
Colloids delaying fluid resuscitation may be associated with mitochondrial dysfunction and may promote inflammation.
Crystalloids Ideally, infused fluids should remain in the plasma for a prolonged period. Colloids remain in the intravascular
Microcirculation space for longer periods than do crystalloids, although their hemodynamic effect is affected by the usual metab-
Resuscitation
olism of colloid substances; leakage through the endothelium in conditions with increased permeability, such as
Nephrotoxicity
sepsis; and/or external losses, such as with hemorrhage and burns. Albumin has pleiotropic physiological activ-
ities including antioxidant effects and positive effects on vessel wall integrity. Its administration facilitates
achievement of a negative fluid balance in hypoalbuminemia and in conditions associated with edema. Fluid re-
suscitation with human albumin is less likely to cause nephrotoxicity than with artificial colloids, and albumin
infusion has the potential to preserve renal function in critically ill patients. These properties may be of clinical
relevance in circulatory shock, capillary leak, liver cirrhosis, and de-escalation after volume resuscitation. Sepsis
is a candidate condition in which human albumin infusion to preserve renal function should be substantiated.
© 2016 Elsevier Inc. All rights reserved.

1. Impact of plasma volume expansion on microvascular perfusion
Circulatory failure or shock is associated with a high risk of death and
characterized by hemodynamic alterations that result in impaired tissue
perfusion [1,2]. Fluid resuscitation is one of the most frequent interven-
tions used at the bedside to improve tissue perfusion, especially in
septic shock. In experimental models of sepsis, generous fluid adminis-
tration is associated with improved survival [3,4]. Fluids should also be
given in a timely fashion. Compared to early administration, delay in
fluid administration resulted in more severe microcirculatory alter-
ations [5], an increase in expression of proinflammatory molecules
and excessive mitochondrial dysfunction [6]. Importantly, excessive
fluid administration should be avoided because a positive fluid balance
is associated with poor outcome [7,8]. In an observational study in more
than 9000 septic patients, the amount of fluid given in the first few
hours of resuscitation followed a U-shaped curve, with an optimal
amount comprised between 15 and 45 mL/kg; mortality rates more
than doubled with higher and lower amounts of fluid [9]. Fluids should,
☆ Conflicts of interest: JLV declares no conflict of interest regarding this manuscript. DDB
declares receiving honoraria as a speaker for Grifols. CW declares receiving honoraria as a
speaker for Kedrion, CSL Behring, Baxter, and Grifols and as a consultant for CSL Behring
and Grifols.
⁎ Corresponding author at: Department of Intensive Care, Erasme Hospital, Université
Libre de Bruxelles, Route de Lennik 808, 1070 Brussels, Belgium. Tel./fax: +32 25553380.
E-mail addresses: jlvincent@intensive.org (J.L. Vincent), ddebacke@ulb.ac.be
(D. De Backer), christian.wiedermann@asbz.it (C.J. Wiedermann).
therefore, be given generously in the early phases but restricted at later
stages, with, if possible, achievement of a negative fluid balance [10,11].
There is still intense debate as to which is the best solution for fluid
resuscitation, even after inclusion of thousands of patients in large-scale
randomized trials [12–14]. Differences in fluid composition can impact
the magnitude and the duration of hemodynamic effects, including at
the microcirculatory level. Indeed, recently, we have begun to focus
more on microcirculatory alterations [15], which are key determinants
of tissue perfusion and also of capillary leak [16,17]. Several studies
have shown that the severity of these alterations is associated with
outcome [18], and we have begun to evaluate the impact of fluid resus-
citation on microvascular perfusion and capillary leakage in patients
with sepsis.
1.1. The effects of fluids: basic principles
Fluids distribute not only in the intravascular compartment, the
desired effect being to increase cardiac output, but also in the extravas-
cular and intracellular compartments. The distribution between com-
partments is principally determined not only by the electrolyte
content and oncotic properties of infused fluids but also by the degree
of microvascular permeability.
The equilibrium between intravascular and interstitial compart-
ments is regulated by the endothelium, which is impermeable to
large molecules, such as albumin and other proteins, but can be crossed
freely by water and electrolytes (Fig. 1). Active transfer of albumin

http://dx.doi.org/10.1016/j.jcrc.2016.04.019
0883-9441/© 2016 Elsevier Inc. All rights reserved.
162 J.L. Vincent et al. / Journal of Critical Care 35 (2016) 161–167
Fig. 1. Schematic distribution of fluids in the intravascular, extravascular, and intracellular
compartments. Water and electrolytes cross compartments freely. Isotonic solutions, such
as Ringer's lactate, have little movement across cellular membranes and distribute in the
vascular and interstitial spaces. The endothelium is impermeable to large molecules,
such as colloids, which thus increase intravascular oncotic pressure and promote diffusion
of water from the other compartments to the intravascular space. In conditions where per-
meability is increased, colloids may also distribute in the interstitial space (dotted arrow).
and other proteins occurs at the endothelial level but the oncotic
gradient between the intravascular and extracellular spaces always re-
mains positive because the concentration of oncotic substances is al-
ways higher in the plasma than in the interstitium. This effect helps
maintain an effective intravascular volume and limits development of
interstitial edema.
Of note, the permeability barrier is not solely guaranteed by endo-
thelial cells but also by the glycocalyx. The glycocalyx is a layer of a
few microns at the surface of endothelial cells and is composed of gly-
cosaminoglycans that also contain antioxidant and anticoagulant mole-
cules. However, the glycocalyx does not contain albumin, in part due to
the negative charges of glycosaminoglycans and albumin, which thus
tend to repel each other [19]. As a result, albumin tends to accumulate
at the outer layer of the glycocalyx, generating a second oncotic gradient
that helps to limit fluid leakage [20,21]. When the glycocalyx and/or en-
dothelium are damaged or dysfunctional as in sepsis, vascular perme-
ability is increased, allowing leakage of fluids but also of large
molecules (including albumin) into the interstitium. Importantly, the
negative charges of albumin and the glycocalyx also decrease, limiting
somewhat the formation of the albumin layer at the outer glycocalyx
[19]. The oncotic gradient may thus be minimized but is never reversed,
as active recapture of interstitial proteins occurs, even in sepsis. Never-
theless, this increase in permeability may alter the plasma expansion
properties of fluids.
The equilibrium between interstitial and intracellular compartments
is driven by osmotic forces at the level of the cellular membrane. These
osmotic forces are generated by electrolyte channels and transporters
and remain tightly regulated, even in sepsis.
1.2. Plasma-expanding effects of fluids
Oncotic and osmotic gradients explain the different plasma-
expanding properties of intravenous solutions in normal conditions.
Isotonic solutions, such as normal saline or plasmalyte, and near isotonic
solutions, such as Ringer's lactate and Ringer's acetate, distribute in the
vascular and interstitial space with little movement of water across cel-
lular membranes [22] (Fig. 1). These solutions have a plasma-expanding
capacity that is lower than the infused volume. Colloid solutions have
plasma-expanding capabilities greater than just the effect of the infused
volume [23]. Indeed, their administration increases intravascular
oncotic pressure, promoting diffusion of water from the other
compartments into the intravascular space [24,25] (Fig. 1). This effect
was nicely demonstrated by Margarson and Soni [26], who showed
that administration of 200 mL of 20% albumin was immediately follow-
ed by a decrease in the hematocrit corresponding to the dilutional effect
induced by adding 200 mL to the plasma, and then a further decrease,
twice as large, over the next few minutes, peaking at 60 minutes and re-
maining stable for 240 minutes. The extent and duration of the hemody-
namic effect depend on the ability of the colloid molecules to remain
inside the intravascular space, a condition that is affected by the metab-
olism of these substances as well as the degree of leakage through the
endothelium [25]. External losses of plasma proteins can also occur in
other conditions, such as bleeding, burns, or nephrotic syndrome, but
this is beyond the scope of this review.
1.3. Is the plasma-expanding capacity of albumin affected in conditions of
increased endothelial permeability?
Administration of albumin does not decrease capillary leak [27], and
part of the albumin will escape into the interstitial fluid. One may thus
be concerned that albumin may be relatively ineffective in conditions
associated with significant capillary leak. In a model of experimental
sepsis associated with increased permeability, the expanding capacity
of albumin remained 3 times higher than that of crystalloid, as in normal
conditions [28]. Interestingly, Dubois et al [29] demonstrated that the
administration of albumin was associated with significantly increased
albumin levels in patients with septic shock, indicating that the albumin
remained, at least in part, in the plasma. Similar results were observed
in the ALBIOS trial [14].
In several large randomized trials in sepsis, the plasma-expanding
capacity of colloids was less than in normal conditions [12,30]. In the
SAFE trial, the amount of fluid was only 30% higher in the crystalloid
than in the albumin group [12], and the amounts of fluid administered
were similar in the 2 groups in the ALBIOS trial [14]. The total amount
of fluid administered was also approximately 30% higher in the crystal-
loid than in the hydroxyethyl starch (HES) groups in several large trials
[13,30,31]. These trials indicate that the expanding capacity of colloids is
indeed decreased in sepsis.
1.4. Duration of the plasma-expanding capacity of the various solutions
In addition to the immediate change in plasma volume, the duration
of the plasma expansion effect should also be considered. In patients un-
dergoing cholecystectomy, the duration of plasma expansion was only 2
hours for gelatin and 4 hours for HES [32]. In experimental conditions,
plasma expansion persisted at 6 hours but was reduced by 30% [33].
Similarly, in patients with septic shock, Margarson and Soni [26] report-
ed that, although albumin disappeared from plasma more rapidly than
in normal conditions, at 6 hours its level was still only 20% lower in sep-
tic patients than in healthy control subjects [26].
1.5. Impact of fluids on microvascular perfusion
Sepsis-associated microcirculatory alterations are characterized by
heterogeneity of perfusion, with nonperfused capillaries in close vicinity
to adequately perfused capillaries [15,18,34,35]. Several studies have
demonstrated that fluids can improve microvascular perfusion, increas-
ing the proportion of perfused capillaries and decreasing perfusion het-
erogeneity [36,37]. Of note, these microcirculatory effects are quite
independent from any systemic effects. Interestingly, the earlier
the fluid administration, the greater the effect on microcirculatory
alterations [36], although large amounts of fluid may not be needed.
In 1 study evaluating the effects of repeated fluid boluses, only
the first bolus was associated with an improvement in microvascular
perfusion [37].
The impact of the type of fluid on the microcirculation is debated. In
experimental conditions, colloids and especially albumin had a more
 J.L. Vincent et al. / Journal of Critical Care 35 (2016) 161–167
beneficial impact on the microcirculation than crystalloids [38–41].
These effects were associated with decreased leukocyte and platelet ad-
hesion to the endothelium [39,40] and even decreased cellular damage
[40]. In patients, the evidence is more limited. In 1 trial, colloids were as-
sociated with greater microvascular responses than crystalloids [42],
but this was not shown in another trial [36]. There may be a sensitivity
issue, as the effects of small amounts of fluids may be too limited to be
documented. A final question is whether this effect is sustained over
time, and this has not yet been investigated.
Recently, it has been proposed that the microcirculation could be
used to guide fluid resuscitation. In septic patients identified as fluid re-
sponders in terms of systemic hemodynamic changes, organ function
improved only in those patients in whom the microcirculation also im-
proved in response to fluids [43]. Interestingly, these patients also had
an impaired microcirculation at baseline, suggesting that evaluation of
the microcirculation may be used to better identify those patients who
may benefit from fluid administration.
1.6. Impact of fluid administration on the glycocalyx
As discussed earlier, the glycocalyx is a complex network of macro-
molecules, including cell-bound proteoglycans and sialoproteins, lining
the apical side of vascular endothelial cells. Maintenance of the glycoca-
lyx is important for microvascular perfusion and permeability [21,44].
Destruction of the glycocalyx is associated with occurrence of microvas-
cular disorders similar to those observed in sepsis [45]. In sepsis, several
investigators have reported an alteration of the endothelial glycocalyx
[46,47]. Sepsis-associated alterations of the glycocalyx induced by in-
flammatory mediators and leukocytes compromise endothelial perme-
ability, resulting in associated interstitial fluid shift and generalized
edema, conditions that result in acute kidney injury (AKI) and other
organ dysfunctions. To what extent the administration of fluids can pro-
tect or restore the glycocalyx is still a matter of research [44], but a few
studies have shown that colloid administration may help to preserve
the glycocalyx [48,49]. Experimentally, albumin protected against the
loss of glycocalyx at the capillary surface in ischemia-reperfusion
models [20,50,51] and reduced the damage to the glycocalyx and facil-
itated its restoration in hemorrhagic shock [52].
2. Potential beneficial effects of albumin in patients
Since 2000, several large randomized controlled trials (RCTs) have
been performed to clarify the efficacy and safety of volume resuscitation
in critically ill patients. The results of these studies have encouraged
new thinking about the role of artificial colloids and paved the way for
the development of lower risk crystalloids. In contrast to results of
small trials using nonvalidated surrogate markers, several large RCTs
have confirmed an increased need for renal replacement therapy and
a trend toward or even a significantly higher mortality in HES groups
[13,30,31,53]. For initial volume substitution in severe sepsis and septic
shock in particular, the international Surviving Sepsis Campaign guide-
lines therefore now recommend that crystalloid preparations should be
used and not HES [54]. Administration of natural albumin infusions may
be considered in critically ill patients with sepsis, if crystalloids alone
appear insufficient [54].
2.1. Lessons learned from studies of albumin administration in sepsis
Fluid replacement with a 4% albumin solution was shown to be safe
compared to a 0.9% NaCl solution in the randomized, double-blind SAFE
study [12,55,56]. Although mortality rates were similar overall in the 2
groups, in the predefined subgroup of patients with severe sepsis, the
mortality rate was lower in the albumin group than in the group receiv-
ing saline solution, particularly in those whose serum albumin levels
were increased to more than 2.5 g/dL [55]. The therapeutic efficacy of al-
bumin administration in terms of morbidity endpoints has also been
163
shown in patients in whom hypoalbuminemia was corrected by in-
creasing serum albumin levels to greater than or equal to 3 g/dL (for re-
view, see Vincent et al [57]).
The effects of keeping the albumin level at 3.0 g/dL for up to 28 days
during the intensive care unit (ICU) stay in patients with sepsis were
studied in the ALBIOS trial [14]. In this open-label RCT, the efficacy of sa-
line with and without administration of 20% albumin solution was stud-
ied in 100 ICUs in Italy with a total of 1818 patients. There were no
significant differences in overall survival after 28 or 90 days. In the
large subgroup of patients with septic shock, including more than
1300 subjects, there was a significant survival advantage after 90 days
in the albumin group (42.6% vs 48.4%; P = .03); however, this observa-
tion needs further study before definitive conclusions can be drawn.
The survival curves from the 3 largest studies of volume therapy
with human albumin in sepsis, namely, SAFE [55], ALBIOS [14], and
EARSS (results not yet fully published [58]), are shown in Fig. 2. A
pooled analysis of primary outcome mortality data confirmed that ad-
ministration of albumin can significantly reduce mortality in patients
with severe sepsis or septic shock [59]. These 3 studies [12,14,58] com-
prise 91% of all patients enrolled in RCTs on albumin in sepsis. Similar ef-
fect sizes of albumin were seen in meta-analyses including small RCTs
[60,61]. The meta-analysis by Patel et al [60] failed to confirm significant
efficacy of human albumin solutions in patients with sepsis, but this
may have been due to the fact that the mortality data in this meta-
analysis included a number of “high-risk-of-bias” studies [62,63]. Use
of albumin in sepsis is supported by a network meta-analysis by
Rochwerg et al [61], who concluded that when volume therapy with
human albumin solutions was performed, sepsis mortality was reduced.
Thus, current best evidence confirms the 2012 Surviving Sepsis
Campaign Guideline recommendation for the administration of
human albumin solutions in patients with sepsis when crystalloids
alone are insufficient.
2.2. Effects of albumin on renal function
In contrast to artificial colloids, such as HES [13,31,53,64], adminis-
tration of human albumin as a colloid has no nephrotoxic effects
[12,14,55,57,64]. None of the large studies of albumin infusion, using
hyperoncotic or hypooncotic solutions, in patients with severe sepsis
or septic shock has given any signal of an adverse effect of albumin on
renal function [14,55,58]. The data from the ALBIOS study showed that
the need for renal replacement therapy was not significantly influenced
by use of albumin, but in a post hoc analysis of patients with septic
shock, treatment with albumin resulted not only in reduced mortality
but also in a favorable negative fluid balance at an earlier stage, more
frequent hemodynamic stabilization in the first 24 hours and reduced
need for vasopressor therapy [14]. These effects may beneficially influ-
ence renal function in patients with septic shock.
Development of AKI and occurrence of death in patients with AKI
were studied in a meta-analysis in critically ill patients, including
those with cirrhosis [64]. Hyperoncotic human albumin (20% or 25%)
significantly reduced the incidence of AKI, whereas after hyperoncotic
HES solutions, the onset of AKI was significantly increased, with similar
results for mortality. The conclusion of this meta-analysis that
hyperoncotic colloid solutions do not per se cause AKI but that the
renal effects are colloid specific—human albumin is nephroprotective
and HES is nephrotoxic—was confirmed by the EARSS and ALBIOS stud-
ies [14,58].
2.3. Hypoalbuminemia as a renal risk factor
The relationship between hypoalbuminemia and renal failure was
analyzed in a systematic review of studies describing the association
of serum albumin with the incidence of AKI as well as the impact of
lower serum albumin levels on mortality in patients with AKI [65]. Sev-
enteen trials involving over 3000 patients were included, and
164 J.L. Vincent et al. / Journal of Critical Care 35 (2016) 161–167
Fig. 2. Survival curves of 3 large studies of volume resuscitation with human albumin solutions in severe sepsis with no adverse effects on renal function: SAFE [55] (A), EARSS [58] (B), and
ALBIOS [14] (C). Copyright permission obtained from original sources.
hypoalbuminemia was confirmed as an independent predictor for de-
veloping AKI and for death after development of AKI. In patients who
developed AKI, with each 1.0 g/dL decrease in serum albumin, mortality
increased by 147%. Prospective RCTs on renal protection with exoge-
nous human albumin in patients with hypoalbuminemia therefore
seem justified.
2.4. Renal protection of albumin in patients with cirrhosis
Convincing evidence for renal protection by administration of
human albumin comes from studies in patients with cirrhosis compli-
cated by tense ascites, hepatorenal syndrome, or spontaneous bacterial
peritonitis. Cardiovascular disorders after large-volume paracentesis in
patients with liver cirrhosis can be prevented better by human albumin
infusions than by infusions of other solutions, likely as a result of the he-
modynamic and renal characteristics of human albumin administration.
After paracentesis, morbidity and mortality differed when patients re-
ceived human albumin rather than alternative infusions [66], including
artificial colloids that could have independently exerted adverse renal
effects, in particular hyperoncotic solutions.
Worsening of renal function in patients with cirrhosis and spontane-
ous bacterial peritonitis augments mortality independently of the
correct administration of antibiotics. Whether human albumin adminis-
tration to patients with spontaneous bacterial peritonitis could improve
kidney function and reduce mortality was also tested [67]. In 4 studies
with a total of 288 patients receiving antibiotics, the incidence of renal
dysfunction was significantly reduced from 31% in the control group
to 8% in groups treated with human albumin; there was also a signifi-
cant reduction in mortality.
2.5. Mechanisms involved in renal protection by albumin
The physiological functions of albumin include the transport of
water-insoluble substances in plasma, buffering acid-base properties,
and anti-inflammatory and antioxidative effects [68,69], all of which
likely contribute to its favorable effects on cell and organ function in se-
verely ill patients in addition to its hemodynamic actions related to pre-
serving the colloid osmotic pressure. In addition, endogenous albumin is
involved, through various signal transduction pathways, in maintaining
the integrity and function of the proximal tubule cells through a
proliferation-regulating effect [70].
2.6. Reduced drug nephrotoxicity with albumin
The concentration of serum albumin is the most powerful predictor
of amikacin-induced nephrotoxicity [71], suggesting that the potential
nephrotoxicity of other drugs may also be associated with plasma
serum albumin concentrations. In the amikacin study, in which patients
with other risk factors for AKI were excluded, serum creatinine was se-
quentially determined and plasma concentrations of amikacin mea-
sured in 104 participants treated for at least 36 hours with
intravenous amikacin [71]. Nephrotoxicity developed in 9.6% (10/104)
of patients, and the lower the serum albumin concentration, the higher
was the risk of toxicity; the serum albumin level was the most powerful
predictor of nephrotoxicity of aminoglycoside therapy in logistic regres-
sion analysis. Low serum albumin concentrations were also associated
with increased amikacin plasma trough levels, independent of initial
renal function, dose of amikacin administered, and other less important
risk factors [71].
2.7. Restoration of optimal net fluid balance by albumin
Volume replacement therapy by infusion of large volumes of fluid
gives rise to fluid retention. Edema formation is usually worsened by
hypoalbuminemia and can be responsible for delayed weaning from
ventilation and associated with various organ complications and in-
creased mortality. Even a fluid overload of 5% to 10% body weight in crit-
ically ill patients is associated with prolonged mechanical ventilation,
impaired gas exchange, and poorer postoperative wound healing and
blood clotting (for review, see Glassford and Bellomo [72]). Administra-
tion of diuretics to restore an optimal net fluid balance is considered to
be the standard treatment [73].
Nephrotic syndrome is characterized by macroalbuminuria and also
edema formation. As a typical observation in such patients, the loop di-
uretic, furosemide, partly loses its natriuretic effect. The administration
of human albumin may potentiate the effect of furosemide in patients
with nephrotic syndrome due to albumin infusion-induced changes in
renal hemodynamics that have been demonstrated in patients with
acute decompensation of cirrhosis and AKI [74]. In a double-blind,
placebo-controlled study, 9 nephrotic patients receiving a standardized
intake of sodium chloride were randomized to receive 60 mg furose-
mide in 200 mL of a 20% solution of human albumin, 60 mg furosemide
in 200 mL placebo solution, or placebo in 200 mL of a 20% solution of
human albumin. Administration of furosemide alone was associated
 J.L. Vincent et al. / Journal of Critical Care 35 (2016) 161–167
with a significantly greater urinary volume and urine sodium excretion
than when only human albumin was administered. The simultaneous
administration of furosemide and human albumin caused a significant
increase in the furosemide effect and plasma levels of atrial natriuretic
factor, and serum and albumin concentrations increased [75].
Fluid accumulation in the lungs of patients with acute lung injury
tends to increase when serum protein levels are low. Such patients
may respond better to a combination of a diuretic with albumin, than
to furosemide alone. In a double-blind, placebo-controlled multicenter
RCT, 40 mechanically ventilated patients with acute respiratory distress
syndrome (ARDS) and serum protein concentrations less than 6 g/dL
were studied [76]. Patients who received albumin plus furosemide
had better oxygenation and better serum protein concentrations than
the control group who received furosemide alone and showed signifi-
cantly less fluid retention (more negative net fluid balance); control pa-
tients more frequently developed hypotension and had fewer
vasopressor-free ICU days. These effects were reflected in correspond-
ing differences in the extent of organ failure at the end of the study
[76]. Thus, in this clinical situation also, a protective negative fluid bal-
ance can be more easily achieved with human albumin infusions than
without. This finding was confirmed in a recent systematic review and
meta-analysis [77].
High central venous pressure was associated with AKI in a retrospec-
tive study in surgical ICU patients with severe sepsis even after adjust-
ment for fluid balance and positive end-expiratory pressure level [78].
The observation that an increase in renal venous pressure may directly
cause sodium retention in the kidney has been described before [79],
and this may further increase venous pressure because of expansion of
the plasma volume. A cumulative net positive fluid balance after initial
generous fluid resuscitation in septic shock may be an ICU scenario in
which this pathophysiological mechanism could play a role. The poten-
tially negative impact of excessive fluid therapy has received increasing
attention recently [80]. In these patients, excretory function is already
impaired. As a consequence, additional interstitial edema delays the
functional recovery of the kidney. Conservative strategies in volume
therapy are, therefore, increasingly being advocated [81]. Targets of vol-
ume therapy need to be carefully defined and fluid status accurately
assessed [82], not only for early ARDS but also for fluid resuscitation in
severe sepsis and septic shock [83].
Observational data of intraoperative fluid management with albu-
min infusion in renal transplantation have shown that among the
most important factors influencing the frequency of delayed graft func-
tion, 1-year graft survival and overall mortality were the cold ischemia
time of the cadaver transplant, duration of surgery, age of the recipient,
and the intraoperative administration of human albumin; higher doses
of albumin were associated with more and earlier urine production [84].
Additional evidence of the beneficial effects of albumin infusion on the
restoration of an optimal net fluid balance was seen in a study investi-
gating whether administration of human albumin to correct hypoalbu-
minemia could have a positive impact on organ function in a mixed
medical-surgical population of critically ill patients [29]. Adult patients
with a serum albumin concentration of 3.0 g/dL or less were randomly
assigned to receive 300 mL of 20% albumin solution on day 1 followed
by 200 mL/d as long as the serum albumin concentration remained
less than 3.1 g/dL or no albumin (control group). The 2 groups were
comparable at baseline with respect to age, sex, serum albumin concen-
tration, and Acute Physiology and Chronic Health Evaluation II score. In
the control group, fluid retention was almost 3 times as high as in the
albumin group [29]. Although comparable diuretics were used, correc-
tion of hypoalbuminemia by administration of human albumin solu-
tions resulted in less pronounced fluid retention.
3. Conclusions and future directions
The plasma-expanding effects of crystalloids and colloids are de-
creased in conditions of increased permeability, but the extent and
165
duration of plasma expansion remain higher with colloids than with
crystalloids. Both types of fluid can increase microvascular perfusion,
but there is marked individual variability in this effect. The impact of
the type of fluid on microvascular perfusion and function needs to be
better defined.
Albumin has pleiotropic physiological activities, including antioxi-
dant effects and positive effects on vessel wall integrity. Although not
indicated in all critically ill patients, there are data to support albumin
administration in patients with sepsis. Its primary role on plasma colloid
osmotic pressure has direct importance for the maintenance of normal
renal function. Infusion of albumin solution facilitates achievement of
a negative fluid balance in hypoalbuminemia and in conditions associat-
ed with edema. Administration of human albumin helps maintain glo-
merular filtration via hemodynamic and oncotic mechanisms. By
contrast with artificial colloids, fluid resuscitation with human albumin
is not nephrotoxic. Reduction of renal morbidity in liver cirrhosis and
freedom from nephrotoxicity support a renoprotective effect of
human albumin, in addition to its unique pharmacodynamic properties.
Further studies must show in which clinical indications, beyond
hepatology, this property may be particularly useful. Sepsis is a candi-
date condition in which the therapeutic potential of human albumin in-
fusion on renal function should be substantiated.
Acknowledgments
This review is based on the lectures presented at the satellite sympo-
sium organized by Grifols, manufacturer of albumin, held on February 6,
2015, in Barcelona (Spain), in the context of the 20th International Sym-
posium on Infections in the Critically Ill Patient.
Jordi Bozzo, PhD, CMPP, and Eva Medina (Grifols) are acknowledged
for their editorial assistance in the preparation of the manuscript.
References
[1] Vincent JL, De Backer D. Circulatory shock. N Engl J Med 2013;369:1726–34.
[2] Cecconi M, De Backer D, Antonelli M, Beale R, Bakker J, Hofer C, et al. Consensus on
circulatory shock and hemodynamic monitoring. Task force of the European Society
of Intensive Care Medicine. Intensive Care Med 2014;40:1795–815.
[3] Zanotti-Cavazzoni SL, Guglielmi M, Parrillo JE, Walker T, Dellinger RP, Hollenberg
SM. Fluid resuscitation influences cardiovascular performance and mortality in a
murine model of sepsis. Intensive Care Med 2009;35:748–54.
[4] Hollenberg SM, Dumasius A, Easington C, Colilla SA, Neumann A, Parrillo JE. Charac-
terization of a hyperdynamic murine model of resuscitated sepsis using echocardi-
ography. Am J Respir Crit Care Med 2001;164:891–5.
[5] Legrand M, Bezemer R, Kandil A, Demirci C, Payen D, Ince C. The role of renal hypo-
perfusion in development of renal microcirculatory dysfunction in endotoxemic rats.
Intensive Care Med 2011;37:1534–42.
[6] Correa TD, Vuda M, Blaser AR, Takala J, Djafarzadeh S, Dunser MW, et al. Effect of
treatment delay on disease severity and need for resuscitation in porcine fecal peri-
tonitis. Crit Care Med 2012;40:2841–9.
[7] Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K, Gerlach H, et al. Sepsis in
European intensive care units: results of the SOAP study. Crit Care Med 2006;34:
344–53.
[8] Boyd JH, Forbes J, Nakada TA, Walley KR, Russell JA. Fluid resuscitation in septic
shock: a positive fluid balance and elevated central venous pressure are associated
with increased mortality. Crit Care Med 2011;39:259–65.
[9] Liu V, Morehouse JW, Soule J, Whippy A, Escobar GJ. Fluid volume, lactate values,
and mortality in sepsis patients with intermediate lactate values. Ann Am Thorac
Soc 2013;10:466–73.
[10] Murphy CV, Schramm GE, Doherty JA, Reichley RM, Gajic O, Afessa B, et al. The im-
portance of fluid management in acute lung injury secondary to septic shock. Chest
2009;136:102–9.
[11] Hoste EA, Maitland K, Brudney CS, Mehta R, Vincent JL, Yates D, et al. Four phases of
intravenous fluid therapy: a conceptual model. Br J Anaesth 2014;113:740–7.
[12] Finfer S, Bellomo R, Boyce N, French J, Myburgh J, Norton R. A comparison of albumin
and saline for fluid resuscitation in the intensive care unit. N Engl J Med 2004;350:
2247–56.
[13] Perner A, Haase N, Guttormsen AB, Tenhunen J, Klemenzson G, Aneman A, et al.
Hydroxyethyl starch 130/0.42 versus Ringer's acetate in severe sepsis. N Engl J
Med 2012;367:124–34.
[14] Caironi P, Tognoni G, Masson S, Fumagalli R, Pesenti A, Romero M, et al. Albumin re-
placement in patients with severe sepsis or septic shock. N Engl J Med 2014;370:
1412–21.
[15] De Backer D, Creteur J, Preiser JC, Dubois MJ, Vincent JL. Microvascular blood flow is
altered in patients with sepsis. Am J Respir Crit Care Med 2002;166:98–104.
166 J.L. Vincent et al. / Journal of Critical Care 35 (2016) 161–167
[16] Chappell D, Jacob M, Paul O, Mehringer L, Newman W, Becker BF. Impaired glycoca-
lyx barrier properties and increased capillary tube haematocrit. J Physiol 2008;586:
4585–6.
[17] Wang Z, Holthoff JH, Seely KA, Pathak E, Spencer III HJ, Gokden N, et al. Development
of oxidative stress in the peritubular capillary microenvironment mediates sepsis-
induced renal microcirculatory failure and acute kidney injury. Am J Pathol 2012;
180:505–16.
[18] De Backer D, Donadello K, Sakr Y, Ospina-Tascon G, Salgado D, Scolletta S, et al. Mi-
crocirculatory alterations in patients with severe sepsis: impact of time of assess-
ment and relationship with outcome. Crit Care Med 2013;41:791–9.
[19] Bansch P, Nelson A, Ohlsson T, Bentzer P. Effect of charge on microvascular perme-
ability in early experimental sepsis in the rat. Microvasc Res 2011;82:339–45.
[20] Woodcock TE, Woodcock TM. Revised Starling equation and the glycocalyx model of
transvascular fluid exchange: an improved paradigm for prescribing intravenous
fluid therapy. Br J Anaesth 2012;108:384–94.
[21] Jacob M, Chappell D. Reappraising Starling: the physiology of the microcirculation.
Curr Opin Crit Care 2013;19:282–9.
[22] Koomans HA, Geers AB, Boer P, Roos JC, Dorhout Mees EJ. A study on the distribution
of body fluids after rapid saline expansion in normal subjects and in patients with
renal insufficiency: preferential intravascular deposition in renal failure. Clin Sci
(Lond) 1983;64:153–60.
[23] Orbegozo Cortes D, Gamarano Barros T, Njimi H, Vincent JL. Crystalloids versus col-
loids: exploring differences in fluid requirements by systematic review and meta-
regression. Anesth Analg 2015;120:389–402.
[24] Ernest D, Belzberg AS, Dodek PM. Distribution of normal saline and 5% albumin in-
fusions in cardiac surgical patients. Crit Care Med 2001;29:2299–302.
[25] Ernest D, Belzberg AS, Dodek PM. Distribution of normal saline and 5% albumin in-
fusions in septic patients. Crit Care Med 1999;27:46–50.
[26] Margarson MP, Soni NC. Changes in serum albumin concentration and volume
expanding effects following a bolus of albumin 20% in septic patients. Br J Anaesth
2004;92:821–6.
[27] Margarson MP, Soni NC. Effects of albumin supplementation on microvascular per-
meability in septic patients. J Appl Physiol 2002;92:2139–45.
[28] Bansch P, Statkevicius S, Bentzer P. Plasma volume expansion with 5% albumin com-
pared to Ringer's acetate during normal and increased microvascular permeability in
the rat. Anesthesiology 2014;121:817–24.
[29] Dubois MJ, Orellana-Jimenez C, Melot C, De Backer D, Berre J, Leeman M, et al. Albu-
min administration improves organ function in critically ill hypoalbuminemic pa-
tients: a prospective, randomized, controlled, pilot study. Crit Care Med 2006;34:
2536–40.
[30] Brunkhorst FM, Engel C, Bloos F, Meier-Hellmann A, Ragaller M, Weiler N, et al. In-
tensive insulin therapy and pentastarch resuscitation in severe sepsis. N Engl J
Med 2008;358:125–39.
[31] Guidet B, Martinet O, Boulain T, Philippart F, Poussel JF, Maizel J, et al. Assessment of
hemodynamic efficacy and safety of 6% hydroxyethylstarch 130/0.4 vs. 0.9% NaCl
fluid replacement in patients with severe sepsis: the CRYSTMAS study. Crit Care
2012;16:R94.
[32] Awad S, Dharmavaram S, Wearn CS, Dube MG, Lobo DN. Effects of an intraoperative
infusion of 4% succinylated gelatine (Gelofusine(R)) and 6% hydroxyethyl starch
(Voluven(R)) on blood volume. Br J Anaesth 2012;109:168–76.
[33] Bark BP, Persson J, Grande PO. Importance of the infusion rate for the plasma
expanding effect of 5% albumin, 6% HES 130/0.4, 4% gelatin, and 0.9% NaCl in the
septic rat. Crit Care Med 2013;41:857–66.
[34] Sakr Y, Dubois MJ, De Backer D, Creteur J, Vincent JL. Persistent microcirculatory al-
terations are associated with organ failure and death in patients with septic shock.
Crit Care Med 2004;32:1825–31.
[35] Edul VS, Enrico C, Laviolle B, Vazquez AR, Ince C, Dubin A. Quantitative assessment of
the microcirculation in healthy volunteers and in patients with septic shock. Crit
Care Med 2012;40:1443–8.
[36] Ospina-Tascon G, Neves AP, Occhipinti G, Donadello K, Buchele G, Simion D, et al. Ef-
fects of fluids on microvascular perfusion in patients with severe sepsis. Intensive
Care Med 2010;36:949–55.
[37] Pottecher J, Deruddre S, Teboul JL, Georger JF, Laplace C, Benhamou D, et al. Both pas-
sive leg raising and intravascular volume expansion improve sublingual microcircu-
latory perfusion in severe sepsis and septic shock patients. Intensive Care Med 2010;
36:1867–74.
[38] de Carvalho H, Dorigo D, Bouskela E. Effects of Ringer-acetate and Ringer-dextran so-
lutions on the microcirculation after LPS challenge: observations in the hamster
cheek pouch. Shock 2001;15:157–62.
[39] Horstick G, Lauterbach M, Kempf T, Bhakdi S, Heimann A, Horstick M, et al. Early al-
bumin infusion improves global and local hemodynamics and reduces inflammatory
response in hemorrhagic shock. Crit Care Med 2002;30:851–5.
[40] Steinbauer M, Guba M, Buchner M, Farkas S, Anthuber M, Jauch KW. Impact of
polynitroxylated albumin (PNA) and tempol on ischemia/reperfusion injury: intra-
vital microscopic study in the dorsal skinfold chamber of the Syrian golden hamster.
Shock 2000;14:163–8.
[41] Hoffmann JN, Vollmar B, Laschke MW, Inthorn D, Schildberg FW, Menger MD.
Hydroxyethyl starch (130 kD), but not crystalloid volume support, improves micro-
circulation during normotensive endotoxemia. Anesthesiology 2002;97:460–70.
[42] Dubin A, Pozo MO, Casabella CA, Murias G, Palizas Jr F, Moseinco MC, et al. Compar-
ison of 6% hydroxyethyl starch 130/0.4 and saline solution for resuscitation of the
microcirculation during the early goal-directed therapy of septic patients. J Crit
Care 2010;25:659.e1-8.
[43] Pranskunas A, Koopmans M, Koetsier PM, Pilvinis V, Boerma EC. Microcirculatory
blood flow as a tool to select ICU patients eligible for fluid therapy. Intensive Care
Med 2013;39:612–9.
[44] Chappell D, Jacob M. Role of the glycocalyx in fluid management: small things mat-
ter. Best Pract Res Clin Anaesthesiol 2014;28:227–34.
[45] Cabrales P, Vazquez BY, Tsai AG, Intaglietta M. Microvascular and capillary perfusion
following glycocalyx degradation. J Appl Physiol 2007;102:2251–9.
[46] Marechal X, Favory R, Joulin O, Montaigne D, Hassoun S, Decoster B, et al. Endothelial
glycocalyx damage during endotoxemia coincides with microcirculatory dysfunc-
tion and vascular oxidative stress. Shock 2008;29:572–6.
[47] Chappell D, Hofmann-Kiefer K, Jacob M, Rehm M, Briegel J, Welsch U, et al. TNF-
alpha induced shedding of the endothelial glycocalyx is prevented by hydrocorti-
sone and antithrombin. Basic Res Cardiol 2009;104:78–89.
[48] Chappell D, Brettner F, Doerfler N, Jacob M, Rehm M, Bruegger D, et al. Protection of
glycocalyx decreases platelet adhesion after ischaemia/reperfusion: an animal study.
Eur J Anaesthesiol 2014;31:474–81.
[49] Torres LN, Sondeen JL, Ji L, Dubick MA, Filho I Torres. Evaluation of resuscitation
fluids on endothelial glycocalyx, venular blood flow, and coagulation function
after hemorrhagic shock in rats. J Trauma Acute Care Surg 2013;75:759–66.
[50] Jacob M, Bruegger D, Rehm M, Welsch U, Conzen P, Becker BF. Contrasting effects of
colloid and crystalloid resuscitation fluids on cardiac vascular permeability. Anesthe-
siology 2006;104:1223–31.
[51] Jacob M, Paul O, Mehringer L, Chappell D, Rehm M, Welsch U, et al. Albumin aug-
mentation improves condition of guinea pig hearts after 4 hr of cold ischemia. Trans-
plantation 2009;87:956–65.
[52] Kozar RA, Peng Z, Zhang R, Holcomb JB, Pati S, Park P, et al. Plasma restoration of en-
dothelial glycocalyx in a rodent model of hemorrhagic shock. Anesth Analg 2011;
112:1289–95.
[53] Myburgh JA, Finfer S, Bellomo R, Billot L, Cass A, Gattas D, et al. Hydroxyethyl
starch or saline for fluid resuscitation in intensive care. N Engl J Med 2012;367:
1901–11.
[54] Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving
sepsis campaign: international guidelines for management of severe sepsis and sep-
tic shock: 2012. Crit Care Med 2013;41:580–637.
[55] Finfer S, McEvoy S, Bellomo R, McArthur C, Myburgh J, Norton R. Impact of albumin
compared to saline on organ function and mortality of patients with severe sepsis.
Intensive Care Med 2011;37:86–96.
[56] Delaney AP, Dan A, McCaffrey J, Finfer S. The role of albumin as a resuscitation fluid
for patients with sepsis: a systematic review and meta-analysis. Crit Care Med 2011;
39:386–91.
[57] Vincent JL, Navickis RJ, Wilkes MM. Morbidity in hospitalized patients receiving
human albumin: a meta-analysis of randomized, controlled trials. Crit Care Med
2004;32:2029–38.
[58] Charpentier J, Mira JP, EARSS study group. Efficacy and tolerance of hyperoncotic al-
bumin administration in septic shock patients: the EARSS study. Intensive Care Med
2011;37(Suppl. 1):S115–0438 [abst].
[59] Wiedermann CJ, Joannidis M. Albumin replacement in severe sepsis or septic shock.
N Engl J Med 2014;371:83.
[60] Patel A, Laffan MA, Waheed U, Brett SJ. Randomised trials of human albumin for
adults with sepsis: systematic review and meta-analysis with trial sequential analy-
sis of all-cause mortality. BMJ 2014;349:g4561.
[61] Rochwerg B, Alhazzani W, Sindi A, Heels-Ansdell D, Thabane L, Fox-Robichaud A,
et al. Fluid resuscitation in sepsis: a systematic review and network meta-analysis.
Ann Intern Med 2014;161:347–55.
[62] Shafer SL, Wilkes MM. Re: Randomised trials of human albumin for adults with sep-
sis: systematic review and meta-analysis with trial sequential analysis of all-cause
mortality. Br Med J 2014;349:g4561.
[63] Wiedermann CJ. Double-counted mortality data bias in the 2014 meta-analysis of
Patel and coworkers. Br Med J 2014;349:g4561.
[64] Wiedermann CJ, Dunzendorfer S, Gaioni LU, Zaraca F, Joannidis M. Hyperoncotic col-
loids and acute kidney injury: a meta-analysis of randomized trials. Crit Care 2010;
14:R191.
[65] Wiedermann CJ, Wiedermann W, Joannidis M. Hypoalbuminemia and acute kidney
injury: a meta-analysis of observational clinical studies. Intensive Care Med 2010;
36:1657–65.
[66] Bernardi M, Caraceni P, Navickis RJ, Wilkes MM. Albumin infusion in patients under-
going large-volume paracentesis: a meta-analysis of randomized trials. Hepatology
2012;55:1172–81.
[67] Salerno F, Navickis RJ, Wilkes MM. Albumin infusion improves outcomes of patients
with spontaneous bacterial peritonitis: a meta-analysis of randomized trials. Clin
Gastroenterol Hepatol 2013;11:123–30.
[68] Ha CE, Bhagavan NV. Novel insights into the pleiotropic effects of human serum al-
bumin in health and disease. Biochim Biophys Acta 1830;2013:5486–93.
[69] Quinlan GJ, Martin GS, Evans TW. Albumin: biochemical properties and therapeutic
potential. Hepatology 2005;41:1211–9.
[70] Dixon R, Brunskill NJ. Activation of mitogenic pathways by albumin in kidney prox-
imal tubule epithelial cells: implications for the pathophysiology of proteinuric
states. J Am Soc Nephrol 1999;10:1487–97.
[71] Contreras AM, Ramirez M, Cueva L, Alvarez S, de Loza R, Gamba G. Low serum albu-
min and the increased risk of amikacin nephrotoxicity. Rev Investig Clin 1994;46:
37–43.
[72] Glassford NJ, Bellomo R. Acute kidney injury: how can we facilitate recovery? Curr
Opin Crit Care 2011;17:562–8.
[73] McDermid RC, Raghunathan K, Romanovsky A, Shaw AD, Bagshaw SM. Contro-
versies in fluid therapy: type, dose and toxicity. World J Crit Care Med 2014;3:
24–33.
[74] Garcia-Martinez R, Noiret L, Sen S, Mookerjee R, Jalan R. Albumin infusion improves
renal blood flow autoregulation in patients with acute decompensation of cirrhosis
and acute kidney injury. Liver Int 2015;35:335–43.
 J.L. Vincent et al. / Journal of Critical Care 35 (2016) 161–167
[75] Fliser D, Zurbruggen I, Mutschler E, Bischoff I, Nussberger J, Franek E, et al. Coadmin-
istration of albumin and furosemide in patients with the nephrotic syndrome. Kid-
ney Int 1999;55:629–34.
[76] Martin GS, Moss M, Wheeler AP, Mealer M, Morris JA, Bernard GR. A randomized,
controlled trial of furosemide with or without albumin in hypoproteinemic patients
with acute lung injury. Crit Care Med 2005;33:1681–7.
[77] Uhlig C, Silva PL, Deckert S, Schmitt J, de Abreu MG. Albumin versus crystalloid solu-
tions in patients with the acute respiratory distress syndrome: a systematic review
and meta-analysis. Crit Care 2014;18:R10.
[78] Legrand M, Dupuis C, Simon C, Gayat E, Mateo J, Lukaszewicz AC, et al.
Association between systemic hemodynamics and septic acute kidney injury
in critically ill patients: a retrospective observational study. Crit Care 2013;17:
R278.
167
[79] Firth JD, Raine AE, Ledingham JG. Raised venous pressure: a direct cause of renal so-
dium retention in oedema? Lancet 1988;1:1033–5.
[80] Marik PE. Iatrogenic salt water drowning and the hazards of a high central venous
pressure. Ann Intensive Care 2014;4:21.
[81] Prowle JR, Echeverri JE, Ligabo EV, Ronco C, Bellomo R. Fluid balance and acute kid-
ney injury. Nat Rev Nephrol 2010;6:107–15.
[82] Prowle JR, Kirwan CJ, Bellomo R. Fluid management for the prevention and attenu-
ation of acute kidney injury. Nat Rev Nephrol 2014;10:37–47.
[83] Gattinoni L, Cressoni M, Brazzi L. Fluids in ARDS: from onset through recovery. Curr
Opin Crit Care 2014;20:373–7.
[84] Dawidson IJ, Sandor ZF, Coorpender L, Palmer B, Peters P, Lu C, et al. Intraoperative
albumin administration affects the outcome of cadaver renal transplantation. Trans-
plantation 1992;53:774–82.