Review
doi: 10.1111/joim.12326
Fluid resuscitation in acute medicine: what is the current
situation?
J. A. Myburgh
From the 1Department of Intensive Care Medicine, St George Clinical School, University of New South Wales; and 2Division of Critical Care
and Trauma, The George Institute for Global Health, Sydney, Australia
Abstract. Myburgh JA (University of New South
Wales, Sydney; The George Institute for Global
Health, Sydney, Australia). Fluid resuscitation in
acute medicine: what is the current situation?
(Review) J Intern Med 2015; 277: 58–68.
The administration of intravenous fluids for
resuscitation is the most common intervention
in acute medicine. There is increasing evidence
that the type of fluid may directly affect patient-
centred outcomes. There is a lack of evidence that
colloids confer clinical benefit over crystalloids
and they may be associated with harm. Hydroxy-
ethyl starch preparations are associated with
increased mortality and use of renal replacement
therapy in critically ill patients, particularly those
with sepsis; albumin is associated with increased
mortality in patients with severe traumatic brain
injury. Crystalloids, such as saline or balanced
salt solutions, are increasingly recommended as
first-line resuscitation fluids for the majority of
Introduction
‘I keep six honest serving men (they
taught me all I knew),
Their names are What and Where and
When and Why and How and Who’
The Elephant’s Child by Rudyard Kipling
It is an iconic image: a bag of intravenous fluid
hanging above a sick patient. Administration of
intravenous fluid provides a direct link between the
attending clinician and the patient, and is the most
common intervention in acute medicine. Asanguin-
eous intravenous fluids are administered to almost
all patients during and following major surgery,
most patients admitted to the accident and emer-
gency (A&E) department following acute trauma
and burns, the majority of patients treated in
intensive care units (ICUs) and children admitted
58 ª 2014 The Association for the Publication of the Journal of Internal Medicine
patients with hypovolaemia. There is emerging
evidence that saline may be associated with
adverse outcomes due to the development of
hyperchloraemic metabolic acidosis, although
the safety of balanced salt solutions has not been
established. Fluid requirements vary over the
course of critical illness. The excessive use of
fluids during the resuscitative period is associated
with increased cumulative fluid balance and
adverse outcomes in critically ill patients. The
selection of fluid depends on the clinical context
in which it is administered and requires careful
consideration of the dose and potential for toxic-
ity. There is an urgent need to conduct further
high-quality randomized controlled trials of cur-
rently available fluid therapy in patients with
critical illness.
Keywords: albumin, colloids, crystalloids, fluids, hy-
droxyethyl starch, saline.
to hospital with severe gastroenteritis and fever.
They are widely used as first-line resuscitation
fluids in bleeding patients before blood transfusion
is available, to replace excessive loss of bodily
fluids, for hydration in patients who are unable to
drink and as a vehicle for the delivery of drugs.
Despite their established role in clinical practice for
over a century, until recently, none of these fluids
had been evaluated in specific trials to determine
their safety and efficacy. This is because intrave-
nous fluids were introduced into routine clinical
practice at a time when they were not regarded in
the same light as intravenous drugs [1].
The selection and use of intravenous fluids is
based on physiological principles relating to the
constituents of the fluid; however, there is marked
variation between clinicians, institutions, regions
and countries. The selection of fluids is largely
determined by local preferences and custom, avail-
ability and cost as well as the influence of com-
 J. A. Myburgh
mercial marketing [2]. There has been considerable
debate for over 30 years, based more on opinion
and experience rather than objective evidence, on
whether to use crystalloid-based or colloid-based
fluid resuscitation strategies. This debate has
intensified over the last decade following the pub-
lication of the results of high-quality randomized
controlled trials that have demonstrated that both
the type of fluids and the mode of administration
directly affect patient outcomes [3–6].
To use the Kipling’s analogy, the debate has moved
from focussing on the preferred type – the ‘What’ –
to the broader clinical applicability and efficacy of
fluid resuscitation – the ‘Where, When, Why, How
and Who’.
Given the ubiquity of the use of fluids in vulnerable
patient populations, small but quantifiable
increases in the risk of harm or benefit associated
with specific fluids may have a substantial impact
on public health worldwide.
Historical perspective
The administration of an alkalinized salt solution
to humans was first described during the cholera
pandemic in the early 19th century when benefi-
cial effects were observed in dehydrated and
shocked patients. These observations provided
the basis for establishing the life-saving effects
of intravenous fluids in severely hypovolaemic
patients [7].
A number of events led to the development of
‘modern’ intravenous fluids. In the 1920s, an
American paediatrician, Alexis Hartmann, admin-
istered an alkalinized salt solution to treat children
with gastroenteritis; this solution had originally
been used in physiological experiments in isolated
heart muscle in Sydney Ringer’s laboratory at
University College London in the late 19th century.
The original solution was piped water from the New
River Water Company that contained carbonated
sodium chloride, calcium and magnesium and was
modified to become compounded sodium lactate
that is used to this day as Hartmann’s solution in
the UK and Ringer’s lactate solution in the USA [8].
Development of the use of sodium chloride, often
termed ‘normal’ saline, was based on in vitro
studies of red cell lysis conducted in the 1880s
that suggested that the concentration of salt in
human blood was 0.9%; it is now known to be
Review: Fluid resuscitation in acute medicine
0.6%. Isotonic (0.9%) saline is the most commonly
used fluid worldwide [9].
Human serum albumin was developed from frac-
tionation of whole blood in 1941 and was used
extensively as a resuscitation fluid by the Allies in
World War II. The potential benefits of colloidal
solutions for resuscitation subsequently led to the
development of semi-synthetic plasma alterna-
tives.
Physiological considerations
Based on Ernest Starling’s [10] description of the
determinants of transvascular exchange across
semi-permeable membranes, clinicians have pref-
erentially used solutions that exert and maintain
an effective colloid osmotic pressure or tonicity
relative to extracellular fluid. Theoretically, col-
loids, containing suspensions of molecules with a
molecular weight that makes them relatively
impermeable to an intact capillary membrane,
are retained within the intravascular compart-
ment on an isovolumetric basis; crystalloids,
containing solutions of ions, rapidly equilibrate
across the capillary membrane. On this basis,
colloids were considered to be three times more
effective than crystalloids in increasing depleted
intravascular volumes. Consequently, crystalloids
are associated with the development of greater
clinically significant interstitial oedema than col-
loids [11].
Recent descriptions of the presence and function of
the endothelial glycocalyx layer have provided
insights into the effects of inflammatory and path-
ological conditions on tissue permeability, suggest-
ing that the theoretical benefits of colloids in
conditions such as sepsis or trauma may be
substantially less than those of crystalloids, both
in terms of increasing intravascular volume and
the development of interstitial oedema [1, 12].
Does the type of fluid matter?
Pragmatic factors such as ease of availability, long
shelf life, low cost and familiarity have led to
recommendations for the use of crystalloids as
first-line fluids for acute resuscitation in patients
with trauma, burns or sepsis and during surgery.
The wide availability of semi-synthetic colloids,
particularly hydroxyethyl starch (HES) prepara-
tions, has led to these fluids being used exten-
sively as alternative colloids to human albumin,
ª 2014 The Association for the Publication of the Journal of Internal Medicine 59
Journal of Internal Medicine, 2015, 277; 58–68
 J. A. Myburgh
particularly in regions in which the availability of
albumin is restricted due to cost.
Until recently, there was little evidence that the
type of fluid used for resuscitation was associated
with specific benefit or harm to patients. Clinicians
working in A&E departments, operating theatres
and ICUs focused on short-term haemodynamic
effects associated with fluid resuscitation, rather
than considering potential longer-term conse-
quences, such as organ failure, increased mortality
and decreased functional survival. This approach
was based on the assumption that the principal
driver for increasing survival in hypovolaemic
patients was the prompt correction of symptomatic
hypovolaemia to restore vital organ perfusion.
However, this paradigm has been challenged over
the last decade, particular in relation to albumin,
HES and most recently saline.
Albumin
Before 1998, there was no clear evidence to suggest
that resuscitation with colloids was superior to
resuscitation with crystalloids in terms of patient
outcomes such as mortality. Most studies to com-
pare these fluids had investigated the effects on
short-term physiological surrogate outcomes in
animal studies and human volunteers. In 1998, a
systematic review and meta-analysis of these stud-
ies, reported by the Cochrane Injury Group Albu-
min Reviewers, compared the effects of albumin
versus other fluids in patients with hypovolaemia,
burns and hypo-albuminaemia as well as on over-
all mortality [13]. The group concluded that the
administration of albumin was associated with a
statistically significant 6% increase in mortality
[relative risk (RR) 1.68, 95% confidence interval
(CI) 1.28 to 2.23]. The findings of this analysis
caused considerable alarm in regions in which
albumin was widely used for resuscitation, such as
the UK, and resulted in a sharp decline in the use
of albumin [14]. Albumin was also widely used in
Australia where, produced by blood fractionation
by a single manufacturer (CSL Bioplasma, Mel-
bourne, Australia), it is supplied to public hospitals
free of charge through the National Blood Author-
ity.
Given the potential public health impact of the
Cochrane review, clinician researchers from the
Australian and New Zealand Intensive Care Society
Clinical Trials Group and the George Institute
for Global Health designed and conducted a
60 ª 2014 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2015, 277; 58–68
Review: Fluid resuscitation in acute medicine
large-scale, prospective, blinded, randomized
controlled trial [the Saline vs. Albumin Evaluation
(SAFE)] to determine whether resuscitation with
4% albumin (in 0.9% saline) was associated with
increased mortality at 28 days compared to 0.9%
saline alone in a heterogeneous population of ICU
patients [3]. The study was powered to determine
half the treatment effect observed in the Cochrane
review (3%) yielding a study population of 7000
patients. The SAFE study demonstrated no differ-
ence in 28-day mortality between patients resusci-
tated in the ICU with albumin or saline (RR 0.99,
95% CI 0.91 to 1.09, P = 0.87), thereby refuting the
conclusions of the 1998 Cochrane review [3].
A number of new findings emerged from the SAFE
study. No differences in resuscitation end-points
such as mean arterial pressure, heart rate and
urine output were observed between patients who
received albumin compared to those who received
saline, although there was a statistically significant
but clinically small increase in central venous
pressure in the former group. However, this
increase in intravascular volume was achieved
with similar volumes of study fluid; the observed
albumin : saline ratio in this blinded trial was
1 : 1.4 rather than the traditionally accepted col-
loid : crystalloid ratio of 1 : 3.
The baseline serum albumin concentration in ICU
patients was approximately 28 g/L, consistent
with an acute-phase reduction in serum albumin.
Subsequent analyses demonstrated that there was
no difference in 28-day mortality in patients with
hypo-albuminaemia (<28 g/L) who received albu-
min or saline for resuscitation compared to those
with albumin concentrations >28 g/L [15].
A statistically significant increase in mortality
2 years after randomization was observed in
patients with severe traumatic brain injury who
received albumin (RR 1.63, 95% CI 1.17 to 2.26,
P = 0.003) [16]. The increase in the rate of death
was attributed to the development of intracranial
hypertension during the first 7 days after random-
ization, possibly due to extravasation of albumin
across the damaged blood–brain barrier and/or
due to the relative hypotonicity of the 4% albumin
preparation used in the study (Albumex; CSL
Bioplasma) [17].
Amongst patients with an admission diagnosis of
sepsis, a significant reduction in the adjusted risk
of death at 28 days was demonstrated in those who
 J. A. Myburgh
received albumin compared to those treated with
saline (odds ratio 0.71, 95% CI 0.52 to 0.97,
P = 0.03). This finding suggests that there may be
a beneficial effect of albumin in this patient pop-
ulation [18].
In the Albumin Italian Outcome Sepsis (ALBIOS)
study, the administration of concentrated albumin
as a drug infusion rather than as a resuscitation
fluid to maintain a serum albumin level >30 g/L
was compared to the use of crystalloids alone in
patients with severe sepsis. No statistically signif-
icant difference in 28-day mortality between the
two groups was demonstrated (RR 0.94; 95% CI
0.85 to 1.05, P = 0.29), although there was a
significant reduction in mortality in the subgroup
of patients with septic shock at enrolment (RR
0.87, 95% CI 0.77 to 0.99, P = 0.03) [19].
The findings of the SAFE and ALBIOS studies
suggest a potential beneficial effect of albumin for
resuscitation of patients with sepsis during the
initial admission period to the ICU. However, at
present, this remains the only evidence-based
indication for albumin.
Hydroxyethyl starch
The relative expense, limited availability and logis-
tical difficulties associated with the mass produc-
tion and distribution of albumin as a resuscitation
fluid provided a major impetus for the commercial
development of semi-synthetic plasma alterna-
tives. Of these, HES preparations are the most
commonly used semi-synthetic colloids, particu-
larly in Western Europe where they have been in
use for the last 40 years. Other semi-synthetic
colloids include gelatin–polygeline preparations
and dextran solutions, although the use of the
latter has decreased substantially in the last
decade.
Hydroxyethyl starch solutions are produced by
hydroxyethyl substitution of amylopectin obtained
from naturally occurring starch products such as
sorghum, maize and potatoes. They are categorized
according to their concentration, molecular weight
and the degree of molar substitution on glucose
molecules which determines the rate of enzymatic
hydrolysis. This substitution on glucose molecules
results in expansion of the intravascular volume in
human volunteers for a period of time that is
shorter than with equivalent volumes of albumin,
but longer than that produced by crystalloids; it is
Review: Fluid resuscitation in acute medicine
also a determinant of the degree of accumulation
within the reticulo-endothelial tissues of skin, liver
and kidney.
Tissue accumulation associated with HES was
described histologically as hydrops lysosomalis
generalisatus in skin, liver and kidney, and has
been suggested to be a potential mechanism of
toxicity presenting as pruritus, hyperbilirubina-
emia, coagulopathy and acute kidney injury [20,
21]. Concerns about the safety of hyperoncotic
(>6%), high molecular weight (>200 kD), highly
substituted (>0.5) HES preparations were raised
following the report of an increased incidence of
postgraft azotaemia and requirement for dialysis in
transplant recipients who received kidneys from
brain-dead organ donors resuscitated with 6%
HES (200/0.62) compared to 4% gelatin [22].
Similar results, particularly an increase in renal
dysfunction and requirement for renal replacement
therapy, were obtained in patients with severe
sepsis resuscitated with 6% HES (200/0.6–0.66)
compared to 3% gelatin [23] and in patients
resuscitated with high doses (70 mL/kg) of 10%
HES (200/0.5) compared to compound sodium
lactate [24].
New-generation isotonic (6%) HES preparations
(tetrastarches) with lower molecular weight (<130
kD) and a lower degree of molar substitution (<0.5)
were produced with a reputed improved safety
profile. Many of the studies attesting to the
improved safety of tetrastarches compared the
effects of these preparations to crystalloids, albu-
min or historical controls [25, 26]. These studies
were underpowered and used surrogate or physi-
ological outcomes in selected patient populations
or in post hoc subgroups defined by postrandom-
ization variables.
Tetrastarches, predominantly 6% HES (130/0.4) in
0.9% saline, were rapidly introduced into clinical
practice and became the most commonly pre-
scribed colloids worldwide, particularly in patients
undergoing anaesthesia for major surgery, for
resuscitation in military trauma and in the ICU.
Tetrastarches were recommended for use in doses
of <50 mL/kg/day. However, clinician uncertainty
about the overall safety of HES remained, based on
systematic reviews that consistently demonstrated
that their administration was associated with
increased mortality [27] and because the new-
generation tetrastarches had not been evaluated
in high-quality, investigator-initiated, randomized
ª 2014 The Association for the Publication of the Journal of Internal Medicine 61
Journal of Internal Medicine, 2015, 277; 58–68
 J. A. Myburgh
controlled trials. This was particularly relevant in
Australia, where HES, specifically 6% HES (130/
0.4), was licensed for the first time by the Thera-
peutic Goods Administration of Australia in 2006.
Consequently, a large-scale (n = 7000), blinded,
randomized controlled trial, modelled on the SAFE
study [Crystalloid vs. Hydroxyethyl Starch Trial
(CHEST)], was designed and conducted in Austra-
lia and New Zealand between 2009 and 2012.
CHEST compared the effects of resuscitation with
either 6% HES (130/0.4) in 0.9% saline (Voluven,
Fresenius Kabi, Bad Homburg, Germany) or saline
alone on 90-day mortality and the use of renal
replacement therapy in ICU patients [4]. At the
same time, a blinded, randomized controlled trial
[the Scandinavian Starch for Severe Sepsis/Septic
Shock study (6S)] compared the effects of resusci-
tation with either 6% HES (130/0.42) in Ringer’s
acetate (Tetraspan, B Braun, Melsungen, Ger-
many) or Ringer’s acetate alone on 90-day mortal-
ity and use of renal replacement therapy in 804
ICU patients with septic shock [5].
In 2010, during the recruitment period of CHEST
and 6S, the first of 87 reports published by
Joachim Boldt was retracted due to scientific fraud
[28]. The retracted studies included 11 reports on
the safety and efficacy of 6% HES (130/0.4); the
results of some of these studies had been used in
pharmaceutical industry product information
sheets, clinical trial protocols and submissions
for licensing by medical regulatory authorities [29].
This scandal highlighted the importance of the
results of CHEST and 6S in determining the safety
and efficacy of HES for resuscitation [30].
In CHEST, no significant difference in 90-day
mortality between patients who received either
6% HES (130/0.4) or saline for fluid resuscitation
was found (18% vs. 17%; RR 1.06; 95% CI 0.96 to
1.18; P = 0.26). However, the use of HES was
associated with a significant increase in the
requirement for renal replacement therapy (RR
1.21; 95% CI 1.00 to 1.45, P = 0.04). The median
cumulative dose of HES was 17 mL/kg [4]. The 6S
study demonstrated a significant increase in 90-
day mortality in the 6% HES group compared with
the group treated with Ringer’s acetate alone (51%
vs. 43%; RR 1.17; 95% CI 1.01 to 1.30; P = 0.03)
and a significant increase in the use of renal
replacement therapy (RR 1.35; 95% CI 1.01 to
1.80; P = 0.04). The median cumulative dose of
HES was 44 mL/kg [5].
62 ª 2014 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2015, 277; 58–68
Review: Fluid resuscitation in acute medicine
No significant differences in haemodynamic resus-
citation end-points were demonstrated in either the
CHEST or 6S study between patients receiving HES
and the respective crystalloid, other than a tran-
sient increase in central venous pressure.
Although the use of HES was associated with lower
volumes of resuscitation fluid compared to the use
of the crystalloid, the HES : crystalloid ratio was
1 : 1.3, consistent with the findings of the SAFE
study and other blinded randomized controlled
trials [25]. In CHEST, HES was associated with an
increase in urine output compared to saline at low
levels of acute kidney injury but this was not
associated with a decrease in serum creatinine or
in the use of renal replacement therapy, suggesting
a direct nephrotoxic effect. In both studies, HES
was also associated with increased use of blood
products and adverse events, particularly pruritus.
Following the publication of the results of the
CHEST and 6S studies, a number of independent
systematic reviews were published that consis-
tently demonstrated a significant increase in mor-
tality and use of renal replacement therapy
associated with HES [31–34]. These effects appear
to be common to all preparations of HES, dose
dependent and applicable to all patient popula-
tions in which no safe dose of HES can be recom-
mended.
In June 2013, the Pharmacovigilance Risk Assess-
ment Committee (PRAC) of the European Medicines
Agency suspended marketing of HES preparations
across the European Union, the US Food and Drug
Administration issued a ‘black box’ warning
against the use of HES in high-risk patients and
the Medicines and Healthcare Products Regulatory
Agency withdrew the registration of HES across the
UK and recalled all unused stock [35]. In October
2013, HES marketing authorization holders
appealed against the PRAC decision based on
industry-commissioned expert reports, unpub-
lished industry-sponsored registry data and inves-
tigator-initiated trials published after the original
PRAC decision, in particular the Colloids versus
Crystalloids for the Resuscitation of the Critically
Ill (CRISTAL) trial [36]. The unblinded CRISTAL
trial compared the effects of resuscitation with any
colloid and with any crystalloid on 28-day mortal-
ity in hypovolaemic, hypotensive ICU patients
(n = 2857). No difference in 28-day mortality
between colloid and crystalloid treatment was
observed (25% vs. 27%; RR 0.96; 95% CI 0.88 to
1.04; P = 0.26); however, colloids were associated
 J. A. Myburgh
with decreased 90-day mortality compared to
crystalloids (31% vs. 34%, RR 0. 92; 95% CI 0.86
to 0.99; P = 0.03) although this was not a prespec-
ified secondary outcome measure. Despite the fact
that HES was used as the predominant open-label
colloid in this trial, the efficacy and safety of HES in
this population cannot be conclusively determined
due to lack of randomization and baseline imbal-
ance.
The PRAC revised its original recommendation and
allowed continued use of HES for resuscitation in
patients with severe hypovolaemia and mandated
extended monitoring of kidney function for 90 days
following the administration of HES. Consequently,
the use of HES has decreased substantially across
the European Union, although it continues to be
marketed heavily and used widely in Asia.
The ‘starch story’ over the last 5 years and is an
example of the impact of high-quality trials on
clinical practice and the reaction from industry
when results conflict with marketing expectations
[37]. Based on the currently available high-quality
evidence, the use of HES for resuscitation provides
no clinical benefit and is associated with the
development of general, dose-dependent nephro-
toxicity, adverse events and increased costs. It is
therefore difficult to justify its use in any patient
population.
Crystalloids
As a consequence of high-quality evidence showing
a lack of benefit associated with colloids, clinical
practice guidelines increasingly recommend cryst-
alloids as first-line resuscitation fluids in high-risk
patients including for the treatment of sepsis [38],
trauma [39], burns [40], dehydration and diabetic
emergencies [41] as well as during major surgery
[42]. There is insufficient evidence to support these
recommendations and there is increasing concern
about the safety and efficacy of crystalloids for
resuscitation, both in terms of the type of crystal-
loid used and in the volumes administered.
Of the crystalloids, isotonic (0.9%) saline is the
most commonly used fluid worldwide, particularly
in the USA [2]. Despite widespread use, there is
increasing concern about the effects of hyperchlo-
raemic metabolic acidosis induced by large vol-
umes of saline. The clinical consequences of this
phenomenon are unclear but the findings of both
animal and human volunteer studies have sug-
Review: Fluid resuscitation in acute medicine
gested that they may be associated with the devel-
opment of acute kidney injury, particularly
reduction in glomerular filtration [43] and immune
dysfunction [44].
Most of the evidence for the potential adverse
effects of saline comes from observational studies
comparing the effects of chloride-rich and chloride-
restricted solutions, such as ‘buffered’ or ‘balanced’
salt solutions, in cohorts of acutely ill patients.
Saline is associated with a significant increase in
mortality and use of renal replacement therapy in
patients in ICU [45] and in those with sepsis [46],
as well as a significant increase in major compli-
cations in patients undergoing surgery [47].
‘Balanced’ salt solutions, based on the prototype
compounded sodium lactate, are increasingly
being recommended on the basis that these solu-
tions are physicochemically approximate the con-
stituents of extracellular fluid [42]. However, none
of the current proprietary solutions is truly ‘bal-
anced’ or ‘buffered’ due to the requirement to
maintain electrochemical neutrality by substitut-
ing anions such as lactate, acetate or gluconate
instead of chloride and bicarbonate in these solu-
tions [48]. The administration of large volumes of
these solutions is associated with the development
of metabolic alkalosis and hypotonicity, the clinical
consequences of which are unclear.
Although crystalloids are increasingly being rec-
ommended and used in daily clinical practice,
there is a need to conduct a large-scale, high-
quality, randomized controlled trial to determine
the efficacy, safety and effects on patient outcomes
of saline compared to proprietary preparations of
balanced salt solutions.
Is the dose and timing of fluid important?
Fluid resuscitation is a single component of a
complex resuscitation process with changing
requirements over the duration of acute illness
[1]. Fluids should primarily be used at all times to
correct symptomatic or clinically significant hypo-
volaemia. Estimating hypovolaemia or evaluating
the response to fluid resuscitation is complex and
there is no single clinical or biochemical parameter
that reflects the complexity of the circulation,
particularly under rapidly changing pathological
conditions. Careful assessment of the history of
the presenting problem, an understanding of the
trajectory of the illness and a clinical assessment
ª 2014 The Association for the Publication of the Journal of Internal Medicine 63
Journal of Internal Medicine, 2015, 277; 58–68
 J. A. Myburgh
that integrates trends in the levels of physiological
and biochemical markers are all required.
Resuscitation of patients in shock has been
described in four conceptual phases: an initial
‘salvage’ phase in which the priority is life-saving
measures to restore vital organ perfusion; an
‘optimization’ phase to maintain the restored cir-
culation; a ‘stabilization’ phase to prevent organ
dysfunction following haemodynamic stabilization;
and a ‘de-escalation’ phase in which support is
weaned and intrinsic haemodynamic function is
restored [49].
Salvage
During this phase, typically between 0 and 24 h,
there is a high likelihood of symptomatic hypovol-
aemia, particularly following trauma or in patients
with severe sepsis. Most resuscitation fluid should
be administered during the salvage phase.
Crystalloids, such as saline or balanced salt solu-
tions, are recommended as first-line fluids for
almost all patients. The exception is blood trans-
fusion for bleeding patients where fresh blood and
blood products should be administered as soon as
practical following careful crystalloid resuscitation.
Albumin may have a role for resuscitation in sepsis
during this phase [18, 19], although it is con-
tradicted in patients with traumatic brain injury
[16].
An initial fluid ‘challenge’ or bolus of crystalloid is
recommended in doses of 20–30 mL/kg, primarily
as a treatment for hypovolaemia. Evidence to
support the use of this dose is limited, and is
based on consensus statements and treatment
guidelines [38]. There are emerging data to suggest
that lower doses may be equally effective, for
example as demonstrated in a recent randomized
controlled trial of early goal-directed therapy in
patients with sepsis [50].
The efficacy and safety of bolus fluid resuscitation
has been challenged following publication of the
results of the Fluid Expansion as Supportive
Therapy (FEAST) study [6]. The FEAST study,
conducted in febrile children with compensated
hypotension in resource-limited settings, demon-
strated that bolus resuscitation with albumin or
saline was not associated with a difference in death
at 48 h, but that bolus resuscitation was associ-
ated with a significant increase in death at 48 h
64 ª 2014 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2015, 277; 58–68
Review: Fluid resuscitation in acute medicine
compared to no fluid administration (RR 1.45, 95%
CI 1.13 to 1.86, P = 0.003) [51]. The principal
cause of death in these patients was due to
cardiovascular collapse rather than fluid overload
or neurological factors, suggesting an adverse
interaction between bolus fluid administration
and compensatory autonomic responses [52]. The
generalizability of these results to adult patients in
high-income regions has been questioned, but the
validity of the trial and the findings question the
safety and efficacy of bolus fluid resuscitation in
patients with compensated hypotension and at
least suggest the need for a high degree of caution
before and during administration.
Increasingly, the early use of vasopressors, such as
noradrenaline or adrenaline, is recommended as
an adjunctive resuscitation strategy during the
salvage phase, mainly to reduce the volume of
resuscitation fluid and to improve vital organ
perfusion through augmentation of venous return,
mean arterial pressure and cardiac output.
Optimization
During the optimization phase, typically between
24 and 72 h, the incidence of hypovolaemia is
substantially reduced. Smaller volumes of resus-
citation fluid (5–15 mL/kg) are commonly admin-
istered during this phase as part of the diagnostic
evaluation for suspected hypovolaemia, which is
frequently triggered by isolated physiological
causes such as oliguria or low central venous
pressure.
There is insufficient evidence to show that the
administration of fluids in this context improves
vital organ function, such as in patients with
acute kidney injury, or is effective in improving
systemic perfusion or mean arterial pressure [53].
Most of the administered fluid, particularly in the
case of crystalloids, accumulates in interstitial
tissues under conditions of increased tissue per-
meability. The overall net effect of the adminis-
tration of unnecessary and ineffective volumes is
not improvement in systemic haemodynamic
function, but an increase in cumulative fluid
balance and pathological, iatrogenic interstitial
oedema.
The adverse effects of increased fluid balance and
the long-term outcomes, particularly increased
mortality and prolongation of mechanical ventila-
tion, have been demonstrated in patients with
 J. A. Myburgh Review: Fluid resuscitation in acute medicine

Table 1 Pragmatic evidence-based recommendations for fluid resuscitation (with apologies to Rudyard Kipling)
What Use isotonic, buffered salt solutions as first-line resuscitation fluids
Consider isotonic saline in hypovolaemic, alkalotic patients
Use fresh whole blood or blood components in actively bleeding patients
Consider colloids in severely hypovolaemic patients as second-line resuscitation fluids
Where Use crystalloids for ease of access and stability in prehospital, military and low-income settings
Use blood products for acute haemorrhage in surgical and trauma resuscitation settings
When Fluid requirements change over time
Resuscitation fluids are a key intervention during the salvage phase of resuscitation (0–24 h)
Fluid requirements decrease during the optimization and stabilization phases of resuscitation (24–96 h)
Fluids should be reduced or restricted during the de-escalation phase (>96 h)
Why Resuscitation fluids should primarily be used to treat symptomatic hypovolaemia
Maintenance fluids should only be used in patients unable to maintain adequate enteral hydration
How Consider the history, cause and illness trajectory when selecting and administering resuscitation fluids
No single clinical or physiological parameter accurately measures intravascular volume
Identify and replace the fluid that is most likely to be lost with equivalent volumes
Consider serum osmolality and acid–base status when selecting a resuscitation fluid
Consider the early use of vasopressors with fluids, particularly during salvage and optimization
Consider cumulative fluid balance when selecting the dose of resuscitation fluid
Do not use fluids to treat isolated physiological perturbations such as oliguria
Use fluid boluses with caution in patients with compensated shock, particularly children
Use the smallest bolus volume necessary to treat hypovolaemia, particularly when salvage is completed
Use fluid challenges to diagnose fluid responsiveness with caution, particularly when salvage is completed
All fluids cause interstitial oedema; avoid excessive volumes of crystalloids
Who Albumin has a potential beneficial role as an adjuvant fluid in patients with sepsis and septic shock
Albumin is contraindicated in patients with traumatic brain injury
The safety of other colloids has not been established in patients with acute brain injury
Saline is the fluid of choice in patients with traumatic brain injury
Buffered salt solutions are recommended in patients undergoing major surgery and those with burns
The safety and efficacy of semi-synthetic colloids has not been established in any patient population
Hydroxyethyl starch is contraindicated in patients with severe sepsis
Hydroxyethyl starch is contraindicated in patients at risk of developing acute kidney injury
The safety and efficacy of hypertonic crystalloids has not been established in any patient population

sepsis [54] and the acute respiratory distress
syndrome (ARDS) [55].
Stabilization
The role and use of resuscitation fluids during the
stabilization phase, which normally occurs between
72 and 96 h, are similar during the optimization
phase. The incidence of hypovolaemia is low and
therefore fluid challenges should be used infre-
quently if at all, unless there are objective and mea-
sureable losses of fluids that result in hypovolaemia.
It is during this period that unnecessary fluid
administration, particularly ‘maintenance’ fluids
and administration for drug infusions, substan-
tially contributes to cumulative fluid balance.
There is increasing debate about introducing
‘restrictive’ fluid strategies to reduce and minimise
cumulative fluid balance, with some evidence that
these approaches are effective in reducing morbid-
ity in surgical patients [56]. However, there is no
consensus or evidence-based guidelines to define
such strategies.

ª 2014 The Association for the Publication of the Journal of Internal Medicine 65
Journal of Internal Medicine, 2015, 277; 58–68
 J. A. Myburgh
De-escalation
The priority during the de-escalation or ‘de-resus-
citation’ phase, usually after 96 h or when haemo-
dynamic stability has returned, is to achieve a
negative fluid balance either by restricting intrave-
nous fluid administration or by increasing fluid
removal through spontaneous or induced diuresis.
There is insufficient evidence to recommend the
use of haemofiltration to achieve this in the
absence of acute kidney injury [53].
Whilst it may be intuitive to begin an active
strategy to remove excess fluid that has accumu-
lated over the preceding periods, restricting the use
of excessive and unnecessary volumes of fluids
would seem to be a more prudent and effective
strategy. The adverse effects of interstitial oedema
may not be prevented by the simple removal of
excess fluid after haemodynamic stability has
returned, particularly if adverse effects are associ-
ated with specific, dose-dependent toxicity, as seen
with the administration of HES. In addition, fluid
restrictive strategies have been associated with
adverse long-term neuropsychological outcomes in
patients with ARDS [57].
There appears to be a changing paradigm to the
use of fluids in a more restrictive manner. Such a
strategy needs to be carefully evaluated and a
definitive randomized controlled trial of restrictive
versus liberal fluid strategies should be conducted
in critically ill patients to address this fundamental
issue.
Conclusion
The success of haemodynamic resuscitation
depends on an integrated and comprehensive
strategy with the aim of identifying and treating
the primary cause of shock, careful assessment
and re-assessment and minimising iatrogenic
harm. Ultimately, the resolution of shock and
survival will depend on the patient’s innate func-
tional reserve and the efficiency of resuscitation.
Pragmatic, evidence-based recommendations for
fluid resuscitation are summarized in Table 1.
There is increasing evidence that the types of fluid
and the manner in which they are administered
independently affect patient outcomes. There is
little doubt that resuscitation fluids are life-saving
in patients with severe, symptomatic hypovola-
emia; however, subsequent use of these fluids
66 ª 2014 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2015, 277; 58–68
Review: Fluid resuscitation in acute medicine
during the course of critical illness requires the
same consideration and care that would be
employed for the use of a potentially toxic drug.
Conflict of interest
Professor Myburgh’s institution, the George Insti-
tute for Global Health, has received reimbursement
for travel expenses, speaker’s fees and unrestricted
grant funding and logistic support from Fresenius
Kabi in relation to the Crystalloid versus Hydroxy-
ethyl Starch Trial and reimbursement for travel
expenses and consultancy fees from Baxter Health-
care.
References
1 Myburgh JA, Mythen MG. Resuscitation fluids. N Engl J Med
2013; 369: 1243–51.
2 Finfer S, Liu B, Taylor C et al. Resuscitation fluid use in
critically ill adults: an international cross-sectional study in
391 intensive care units. Crit Care 2010; 14: R185.
3 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.
4 Myburgh JA, Finfer S, Bellomo R et al. Hydroxyethyl starch or
saline for fluid resuscitation in intensive care. N Engl J Med
2012; 367: 1901–11.
5 Perner A, Haase N, Guttormsen AB et al. Hydroxyethyl starch
130/0.42 versus Ringer’s acetate in severe sepsis. N Engl J
Med 2012; 367: 124–34.
6 Maitland K, Kiguli S, Opoka R et al. Mortality after fluid bolus
in African children with shock. N Engl J Med 2011; 364:
2483–95.
7 Lewins R. Saline injections into the veins. London Medical
Gazette 1832; 257–68.
8 Lee JA. Sydney Ringer (1834-1910) and Alexis Hartmann
(1898-1964). Anaesthesia 1981; 36: 1115–21.
9 Awad S, Allison SP, Lobo DN. The history of 0.9% saline. Clin
Nutr 2008; 27: 179–88.
10 Starling EH. On the absorption of fluids from connective
tissue spaces. J Physiol 1896; 19: 312–26.
11 Krogh A, Landis EM, Turner AH. The movement of fluid
through the human capillary wall in relation to venous
pressure and to the colloid osmotic pressure of the blood. J
Clin Invest 1932; 11: 63–95.
12 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.
13 Cochrane Injuries Group Albumin Reviewers. Human albu-
min administration in critically ill patients: systematic
review of randomised controlled trials. BMJ 1998; 317:
235–40.
14 Jones D, McEvoy S, Merz TM et al. International albumin use:
1995 to 2006. Anaesth Intensive Care 2010; 38: 266–73.
15 Finfer S, Bellomo R, McEvoy S et al. Effect of baseline serum
albumin concentration on outcome of resuscitation with
albumin or saline in patients in intensive care units: analysis
 J. A. Myburgh
of data from the saline versus albumin fluid evaluation
(SAFE) study. BMJ 2006; 333: 1044.
16 Myburgh J, Cooper DJ, Finfer S et al. Saline or albumin for
fluid resuscitation in patients with traumatic brain injury. N
Engl J Med 2007; 357: 874–84.
17 Cooper DJ, Myburgh J, Heritier S et al. Albumin resuscitation
for traumatic brain injury: is intracranial hypertension the
cause of increased mortality? J Neurotrauma 2013; 30: 512–
8.
18 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. Inten-
sive Care Med 2011; 37: 86–96.
19 Caironi P, Tognoni G, Masson S et al. Albumin replacement in
patients with severe sepsis or septic shock. N Engl J Med
2014; 370: 1412–21.
20 Dickenmann M, Oettl T, Mihatsch MJ. Osmotic nephrosis:
acute kidney injury with accumulation of proximal tubular
lysosomes due to administration of exogenous solutes. Am J
Kidney Dis 2008; 51: 491–503.
21 Bellmann R, Feistritzer C, Wiedermann CJ. Effect of molec-
ular weight and substitution on tissue uptake of hydroxyethyl
starch: a meta-analysis of clinical studies. Clin Pharmacoki-
net 2012; 51: 225–36.
22 Cittanova ML, Leblanc I, Legendre C, Mouquet C, Riou B,
Coriat P. Effect of hydroxyethylstarch in brain-dead kidney
donors on renal function in kidney-transplant recipients.
Lancet 1996; 348: 1620–2.
23 Schortgen F, Lacherade JC, Bruneel F et al. Effects of
hydroxyethylstarch and gelatin on renal function in severe
sepsis: a multicentre randomised study. Lancet 2001; 357:
911–6.
24 Brunkhorst FM, Engel C, Bloos F et al. Intensive insulin
therapy and pentastarch resuscitation in severe sepsis. N
Engl J Med 2008; 358: 125–39.
25 James MF, Michell WL, Joubert IA, Nicol AJ, Navsaria PH,
Gillespie RS. Resuscitation with hydroxyethyl starch
improves renal function and lactate clearance in penetrating
trauma in a randomized controlled study: the FIRST trial
(Fluids in Resuscitation of Severe Trauma). Br J Anaesth
2011; 107: 693–702.
26 Guidet B, Martinet O, Boulain T et al. Assessment of hemo-
dynamic 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.
27 Perel P, Roberts I. Colloids versus crystalloids for fluid
resuscitation in critically ill patients. Cochrane Database
Syst Rev 2007; (4): CD000567.
28 Shafer SL. Notice of retraction. Anesth Analg 2010; 111:
1567.
29 Shafer SL. Shadow of doubt. Anesth Analg 2011; 112: 498–
500.
30 Myburgh J. CHEST and the impact of fraud in fluid resus-
citation research. Crit Care Resusc 2011; 13: 69–70.
31 Haase N, Perner A, Hennings LI et al. Hydroxyethyl starch
130/0.38-0.45 versus crystalloid or albumin in patients with
sepsis: systematic review with meta-analysis and trial
sequential analysis. BMJ 2013; 346: f839.
32 Gattas DJ, Dan A, Myburgh J, Billot L, Lo S, Finfer S. Fluid
resuscitation with 6% hydroxyethyl starch (130/0.4 and
130/0.42) in acutely ill patients: systematic review of effects
Review: Fluid resuscitation in acute medicine
on mortality and treatment with renal replacement therapy.
Intensive Care Med 2013; 39: 558–68.
33 Perel P, Roberts I. Colloids versus crystalloids for fluid
resuscitation in critically ill patients. Cochrane Database
Syst Rev 2012; 6: CD000567.
34 Mutter TC, Ruth CA, Dart AB. Hydroxyethyl starch (HES)
versus other fluid therapies: effects on kidney function.
Cochrane Database Syst Rev 2013; 7: CD007594.
35 Bellomo R, Bion J, Finfer S, Myburgh J, Perner A, Reinhart K.
Open letter to the Executive Director of the European Med-
icines Agency concerning the licensing of hydroxyethyl starch
solutions for fluid resuscitation. Br J Anaesth 2014; 112:
595–600.
36 Annane D, Siami S, Jaber S et al. Effects of fluid resuscitation
with colloids vs crystalloids on mortality in critically ill
patients presenting with hypovolemic shock: the CRISTAL
randomized trial. JAMA 2013; 310: 1809–17.
37 Bion J, Bellomo R, Myburgh J, Perner A, Reinhart K, Finfer S.
Hydroxyethyl starch: putting patient safety first. Intensive
Care Med 2013; 40: 256–59.
38 Dellinger RP, Levy MM, Rhodes A et al. Surviving sepsis
campaign: international guidelines for management of severe
sepsis and septic shock, 2012. Intensive Care Med 2013; 39:
165–228.
39 American College of Surgeons Committee on Trauma.
Advanced Trauma Life Support for Doctors. American College
of Surgeons Committee on Trauma, 2012. (http://www.facs.
org/trauma/atls/index.html).
40 Arlati S, Storti E, Pradella V, Bucci L, Vitolo A, Pulici M.
Decreased fluid volume to reduce organ damage: a new
approach to burn shock resuscitation? A preliminary study
Resuscitation 2007; 72: 371–8.
41 Chua HR, Venkatesh B, Stachowski E et al. Plasma-Lyte 148
vs 0.9% saline for fluid resuscitation in diabetic ketoacidosis.
J Crit Care 2012; 27: 138–45.
42 Powell-Tuck J, Gosling P, Lobo DN. et al. British Consensus
Guidelines on Intravenous Fluid Therapy for Adult Surgical
Patients (GIFTASUP). March, 2011. (http://www.baden.org.
uk/pdfs.bapen_pubs/giftasup.pdf).
43 Hadimioglu N, Saadawy I, Saglam T, Ertug Z, Dinckan A. The
effect of different crystalloid solutions on acid-base balance
and early kidney function after kidney transplantation.
Anesth Analg 2008; 107: 264–9.
44 Kellum JA, Song M, Li J. Science review: extracellular
acidosis and the immune response: clinical and physiologic
implications. Crit Care 2004; 8: 331–6.
45 Yunos NM, Bellomo R, Hegarty C, Story D, Ho L,
Bailey M. Association between a chloride-liberal vs chloride-
restrictive intravenous fluid administration strategy and
kidney injury in critically ill adults. JAMA 2012; 308: 1566–
72.
46 Raghunathan K, Shaw A, Nathanson B et al. Association
between the choice of IV crystalloid and in-hospital mortality
among critically ill adults with sepsis*. Crit Care Med 2014;
42: 1585–91.
47 Shaw AD, Bagshaw SM, Goldstein SL et al. Major complica-
tions, mortality, and resource utilization after open abdom-
inal surgery: 0.9% saline compared to Plasma-Lyte. Ann Surg
2012; 255: 821–9.
48 Guidet B, Soni N, Della RG et al. A balanced view of balanced
solutions. Crit Care 2010; 14: 325.
ª 2014 The Association for the Publication of the Journal of Internal Medicine 67
Journal of Internal Medicine, 2015, 277; 58–68
 J. A. Myburgh
49 Vincent JL, De BD. Circulatory shock. N Engl J Med 2013;
369: 1726–34.
50 Yealy DM, Kellum JA, Huang DT et al. A randomized trial of
protocol-based care for early septic shock. N Engl J Med 2014;
370: 1683–93.
51 Maitland K, George E, Evans J et al. Exploring mechanisms of
excess mortality with early fluid resuscitation: insights from
the FEAST trial. BMC Med 2013; 11: 68.
52 Myburgh J, Finfer S. Causes of death after fluid bolus
resuscitation: new insights from FEAST. BMC Med 2013;
11: 67.
53 Prowle JR, Bellomo R. Fluid administration and the kidney.
Curr Opin Crit Care 2010; 16: 332–6.
54 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 associ-
ated with increased mortality. Crit Care Med 2011; 39:
259–65.
68 ª 2014 The Association for the Publication of the Journal of Internal Medicine
Journal of Internal Medicine, 2015, 277; 58–68
Review: Fluid resuscitation in acute medicine
55 Wiedemann HP, Wheeler AP, Bernard GR et al. Comparison of
two fluid-management strategies in acute lung injury. N Engl
J Med 2006; 354: 2564–75.
56 Brandstrup B, Svendsen PE, Rasmussen M et al. Which goal
for fluid therapy during colorectal surgery is followed by the
best outcome: near-maximal stroke volume or zero fluid
balance? Br J Anaesth 2012; 109: 191–9.
57 Mikkelsen ME, Christie JD, Lanken PN et al. The adult
respiratory distress syndrome cognitive outcomes study:
long-term neuropsychological function in survivors of acute
lung injury. Am J Respir Crit Care Med 2012; 185: 1307–15.
Correspondence: Professor John A. Myburgh, Professor of Inten-
sive Care Medicine, Department of Intensive Care Medicine, St
George Hospital, Gray St, Kogarah 2217, Sydney, Australia.
(fax: +61291135971; e-mail: jmyburgh@georgeinstitute.org.au).