E SYSTEMATIC REVIEW ARTICLE

Hypertonic Saline in Human Sepsis: A Systematic

Review of Randomized Controlled Trials
Diego Orbegozo, MD, Jean-Louis Vincent, MD, PhD, Jacques Creteur, MD, PhD,
and Fuhong Su, MD, PhD

The role of hypertonic saline in sepsis remains unclear because clinical data are limited and
the balance between beneficial and adverse effects is not well defined. In this systematic
literature review, we searched PubMed and Embase to identify all randomized controlled trials
up until January 31, 2018 in which hypertonic saline solutions of any concentration were used
in patients of all ages with sepsis and compared to a cohort of patients receiving an isotonic
fluid. We identified 8 randomized controlled trials with 381 patients who had received hypertonic
saline. Lower volumes of hypertonic saline than of isotonic solutions were needed to achieve
the desired hemodynamic goals (standardized mean difference, −0.702; 95% CI, −1.066 to
−0.337; P < .001; moderate-quality evidence). Hypertonic saline administration was associ-
ated with a transient increase in sodium and chloride concentrations without adverse effects
on renal function (moderate-quality evidence). Some data suggested a beneficial effect of hyper-
tonic saline solutions on some hemodynamic parameters and the immunomodulatory profile
(very low–quality evidence). Mortality rates were not significantly different with hypertonic saline
than with other fluids (odds ratio, 0.946; 95% CI, 0.688–1.301; P = .733; low-quality evidence).
In conclusion, in our meta-analysis of studies in patients with sepsis, hypertonic saline reduced
the volume of fluid needed to achieve the same hemodynamic targets but did not affect survival. 
(Anesth Analg 2019;128:1175–84)

S
epsis and septic shock are common causes of hospi-
tal admission worldwide and are associated with high
costs for society1,2 and considerable related morbidity
and mortality.1–6 Sepsis is characterized by increased capil-
lary permeability, which generates loss of protein and liquid
into the interstitial space and a decrease in vascular tone.
Both lead to a decrease in effective intravascular volume.7–9
Fluid requirements are, therefore, increased during sepsis,
but excessive fluid administration can be harmful.10–13
The use of colloids to limit fluid overload while main-
taining an adequate intravascular volume is controver-
sial.14,15 Hypertonic saline solutions may represent a
cheaper and potentially safer alternative. These fluids
can promote a shift of water from the intracellular to the
extracellular (interstitial and intravascular) compartment.16
Moreover, they may increase cardiac output,17 modulate the
immune response,18 reduce the permeability of the vascu-
lar wall,19 reduce the volume of fluids needed to obtain the
same hemodynamic targets as with crystalloid solutions,20
and promote the endogenous secretion of vasopressin.21
However, the high chloride content of hypertonic saline
may have adverse effects, including acidosis, coagulopathy,
and impaired renal function.22–25
From the Department of Intensive Care, Erasme University Hospital,
Université Libre de Bruxelles, Brussels, Belgium.
Accepted for publication October 23, 2018.
Funding: None.
The authors declare no conflicts of interest.
Supplemental digital content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions of
this article on the journal’s website (www.anesthesia-analgesia.org).
Reprints will not be available from the authors.
Address correspondence to Jean-Louis Vincent, MD, PhD, Department of
Intensive Care, Erasme University Hospital, Route de Lennik 808, 1070 Brus-
sels, Belgium. Address e-mail to jlvincent@intensive.org.
Copyright © 2019 International Anesthesia Research Society
DOI: 10.1213/ANE.0000000000003955
Although preparation of hypertonic saline is simple and
cheap and the fluid is available worldwide, it is not widely
used, and clinical studies are limited. In view of its potential
beneficial effects on a background of ongoing debate about
optimal fluid choices, we reviewed the clinical experience
with hypertonic saline in patients with sepsis by perform-
ing a systematic review of randomized controlled trials in
which hypertonic saline infusion was compared to any iso-
tonic fluid in patients of all ages with sepsis. We focused
our review on the effects on hemodynamic status, electro-
lyte balance, acid-base status, renal function, coagulation,
immune function, microcirculation, intensive care unit
length of stay, and mortality.
METHODS
We adhered to Preferred Reporting Items for Systematic
Reviews and Meta-analyses guidelines.26 We searched PubMed
and Embase from creation to January 31, 2018, looking for all
original articles exploring the role of hypertonic saline solu-
tions in patients with sepsis. The search was performed by a
statistician to obtain sensitive queries for each database. MeSH
and Emtree (for PubMed and Embase, respectively) were
browsed to identify the most relevant and synonymous terms
in each database. Selected queries were searched as MeSH and
Emtree terms as well as free text within the titles, abstracts, and
keywords of each article. The retained query for Embase was:
([“sepsis”/exp OR “sepsis”] OR [“septic shock”/exp OR “sep-
tic shock”]) AND ([“hypertonic solution”/exp OR “hypertonic
solution”] OR [hypertonic AND {“saline”/exp OR saline}]
OR [hypertonic AND {“sodium”/exp OR sodium}]), and for
PubMed was (“hypertonic solutions” OR “hypertonic saline”
OR “hypertonic sodium”) AND (“sepsis” OR “septic shock”).
The search was not limited by date of publication, language,
population age, or type of study.
Articles were retained for review if they were random-
ized controlled trials performed in patients with sepsis who

June 2019 • Volume 128 • Number 6 www.anesthesia-analgesia.org 1175

Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
 EE Systematic Review Article
received an infusion of hypertonic saline of any concentration.
The included studies must have compared data from a cohort
of patients receiving hypertonic saline alone and a cohort
receiving an isotonic solution or a cohort receiving hypertonic
saline in combination with a colloid if there was also a compari-
son cohort that received just a colloid. Given the limited data in
this field, we included pediatric and adult studies. Two authors
(D.O. and F.S.) independently reviewed the search results
(initially titles and abstracts and then full texts of remaining
articles) to retrieve the studies fulfilling the previous criteria,
and any discrepancy was resolved by consensus. Data for any
reported outcome were extracted to prespecified tables, but we
focused on the following prespecified outcomes: administered
volumes of liquids, effects on hemodynamic status, electrolyte
concentrations, acid-base status, renal function, coagulation,
immune function, microcirculation, intensive care unit length
of stay, and mortality. The results were summarized and are
presented as a systematic review. Risk of bias for each retrieved
study was evaluated using the Cochrane Risk of Bias tool (RoB
2.0 tool).27 For assessment of the overall quality of evidence
for each reported outcome, the Grading of Recommendations
Assessment, Development, and Evaluation approach was used
to create a summary of findings table with the GRADEpro
Guideline Development Tool software (McMaster University,
Evidence Prime, Inc, Hamilton, ON, Canada).
Exploratory, post hoc meta-analyses (in an attempt to
generate hypotheses for future studies) were performed for
outcomes reported in ≥6 of the studies, that is, for the amount
of fluid administered and for mortality. The estimated effect
of each study for these outcomes was calculated with 95%
CIs, and the global effect was then assessed using a random
effects model to address inherent differences among experi-
mental protocols. Results are presented as forest plots with
estimates of heterogeneity. For the amount of fluid admin-
istered, the global effect was reported as the standardized
mean difference because infused volumes were not reported
using the same units across the studies. If reported fluid vol-
umes were assessed at different timepoints in a study, we
selected the longest time period in an attempt to evaluate
whether the effect was sustained over time. For mortality,
the global effect was reported as the odds ratio for survival.
Reported time periods for mortality assessment differed
considerably among studies, but if >1 time period was
reported, we selected the longest to determine whether the
effect was sustained over time. Subgroup analyses consid-
ering only those studies comparing hypertonic saline with
an isotonic crystalloid and only those studies comparing
hypertonic saline + hydroxyethyl starch with hydroxy-
ethyl starch alone were performed. Funnel plots, including
imputed studies using Duval and Tweedie’s trim-and-fill
methodology, were created to assess the risk of publication
bias. All analyses were performed using Comprehensive
Meta-Analysis V3 software (Englewood, CO).
RESULTS
A total of 483 articles were identified, but only 13 had been
performed in patients with sepsis.28–40 Eight of the articles
(including 381 patients who had received hypertonic saline)
were randomized controlled trials29,30,32–35,39,40 and were
retained for further analysis (Supplemental Digital Content,
Figure S1, http://links.lww.com/AA/C670). Five articles
1176   
www.anesthesia-analgesia.org
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
were published in English and 3 in Chinese. The 5 single-
cohort studies28,31,36–38 are shown in Supplemental Digital
Content, Table S1, http://links.lww.com/AA/C670.
The study by Li et al30 had 4 cohorts. For our analyses, we
compared the hypertonic saline versus normal saline groups
separately from the hypertonic saline + hydroxyethyl starch
versus hydroxyethyl starch alone groups.30 A study by Van
Haren et al34 presented the same cohort of patients as in an
article published in 2012 by the same author.35 However,
because the presented outcomes were completely different,
we kept it for descriptive purposes.
Five randomized controlled trials, 2 of which were in
pediatric populations,33,39 compared hypertonic saline with
an isotonic crystalloid (Table 1).29,30,33,39,40 Four randomized
controlled trials compared hypertonic saline + hydroxy-
ethyl starch with hydroxyethyl starch alone (Table 2).30,32,34,35
In general, the risk of bias was high; the results of each indi-
vidual domain for each study are shown in Figure 1.
The tonicity of the infused hypertonic saline solutions
varied between 3%33 and 7.5%.32 The administered dose
was either given according to body weight, between 432
and 15 mL/kg,33 or as a fixed dose between 250 mL34,35 and
500 mL.30
Effects on Amount of Fluid Administered
Six randomized controlled trials reported data on the
amount of fluid administered. Five randomized controlled
trials showed a significant decrease in the total volume
of fluid administered when a hypertonic saline solution
was infused regardless of whether the comparison was
between hypertonic saline and isotonic crystalloid30,33,39
or between hypertonic saline + hydroxyethyl starch and
hydroxyethyl starch alone.30,32,35 One randomized con-
trolled trial reported no differences in the total volume of
fluid administered during the first 72 hours when compar-
ing hypertonic saline with an isotonic crystalloid.40 Meta-
analysis of the 6 studies that reported this outcome found a
significant reduction in the volume of fluids administered
(standardized mean difference, −0.702; 95% CI, −1.066
to −0.337; P < .001; moderate-quality evidence [Table 3])
when hypertonic saline and hypertonic saline + hydroxy-
ethyl starch were compared to an isotonic crystalloid and
hydroxyethyl starch (Figure 2). However, there was a mod-
erate degree of heterogeneity in this analysis (I2 = 72%).
Subgroup analyses of studies that compared only hyper-
tonic saline with an isotonic crystalloid or only hypertonic
saline + hydroxyethyl starch with hydroxyethyl starch
alone gave similar results (Supplemental Digital Content,
Figure S2, http://links.lww.com/AA/C670). The Funnel
plot showed minor publication bias not affecting the esti-
mated global effect on the volume of fluids administered
(Supplemental Digital Content, Figure S3, http://links.
lww.com/AA/C670).
Hemodynamic Effects
Two randomized controlled trials reported no differences in
cardiac filling pressures between hypertonic saline and an
isotonic crystalloid or between hypertonic saline + hydroxy-
ethyl starch and hydroxyethyl starch alone.30,35
An increase in arterial pressure during the first hours of
resuscitation was reported in 2 randomized controlled trials
ANESTHESIA & ANALGESIA
 Table 1.  Summary of Results From Randomized Controlled Trials Comparing Hypertonic Saline With an Isotonic Crystalloid
Country, Acid-Base Status
Author, No. of Hypertonic Fluid and Electrolytes Other
Year Centers Population Exclusion Criteria Groups Saline Dose Volume Hemodynamics (Maximal Difference) Mortality Outcomes
Fang et al, China, Adults with MI, trauma, pregnancy, Hypertonic saline 5 mL/kg Not reported No differences pH 7.34 vs 7.36 17% vs 16%
200829 5 severe expected early 3.5% (number of in 15 min in HR, MAP, (P = not at 28 d
sepsis or death, multiple organ patients = 30) and CO significant). (P = not
septic shock dysfunction syndrome Normal saline sodium 136 vs 135 significant)
(number of mEq/L (P = not
patients = 32) significant).
Hypertonic Saline in Human Sepsis

chloride 98 vs 97
mEq/L (P = not

June 2019 • Volume 128 • Number 6

significant)
Li et al, China, Adults with Hematological disorders, Hypertonic saline 300–500 mL 1883 vs No differences in Not reported 67% vs 67% No differences in
200830 1 severe liver or renal failure, 4% (number of in 30 min 2916 mL HR, MAP, CVP, at 28 d lactate at 24 h
sepsis or pregnancy, expected patients = 15) (P < .01) or vasopressors (P = not
septic shock early death, cardiac Normal saline >3 d significant)
arrest, heart failure (number of
patients = 15)
Chopra India, Children with Multiple organ Hypertonic saline 15 mL/kg 38 vs No differences Sodium change 27% vs 33% Intensive care
et al, 1 septic shock dysfunction syndrome, 3% (number of in 30 min 69 mL/kg in HR, systolic +10 vs +5 mEq/L at intensive unit stay 5 vs
201133 chronic liver disease, patients = 30) (P < .001) arterial pressure, (P < .001). care unit 6 d (P = not
chronic heart failure, Normal saline MAP, diastolic Sodium at 24 h discharge significant)
severe malnutrition, (number of arterial pressure, similar (P = not
dysnatremia, patients = 30) time under significant)
leukemia, lymphoma vasopressor, or
shock reversal
time
Liu et al, China, Children with Age ≤ 28 d, tumors, Hypertonic saline 6 mL/kg in 103 vs No differences in Na+ was higher in 5.0% vs 8.3% Mechanical
201539 1 septic shock liver and kidney 3% (number of 10–15 min, 131 mL/kg HR and MAP the hypertonic at 28 d ventilation 80
failure, congenital patients = 20) maximum (P = .005) saline group than (P = .66) vs 103 h
heart disease, Normal saline 2 boluses at 24 h in the normal (P = .786)
hypernatremia, (number of saline group at 1
expected death within patients = 24) h, but not at 3, 6,
72 h and 24 h after
fluid resuscitation
Asfar et al, France, Adults with Pao2/Fio2 < 100, Hypertonic saline Repeated boluses 2600 vs No differences in Sodium >155 or 46% vs 44% No differences in
201740 22 septic intracranial 3% (number of of 280 mL 2300 mL vasopressor sodium change in at 90 d renal sequential
shock under hypertension, patients = 214) during the first (P = .22)a dose or in 24 h >12 mEq/L (P = .48). organ failure
mechanical dysnatremia, cardiac Normal saline 72 h cardiovascular (=stop of blinded 42% vs 37% assessment
ventilation arrest, overt cardiac (number of sequential protocol) 39.3% vs at 28 d score or need
failure, pregnancy, patients = 220) organ failure 4.1% (P < .0001) (P = .25) for renal
DNR order, inclusion assessment replacement
in another study score therapy or
intensive care
unit length of
stay
Abbreviations: CO, cardiac output; CVP, central venous pressure; DNR, do-not-resuscitate; Fio2, inspired oxygen fraction; HR, heart rate; MAP, mean arterial pressure; MI, myocardial infarction.
a

www.anesthesia-analgesia.org 1177
At day 3 including blinded and open boluses. All comparisons are presented in the following order: hypertonic saline versus normal saline groups.

Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
 Table 2.  Summary of Results From Randomized Controlled Trials Comparing Hypertonic Saline With a Colloid
Acid-Base
Hypertonic Status and

1178   
Country, Saline + Electrolytes
Author, No. of Hydroxyethyl Fluid (Maximal
Year Centers Population Exclusion Criteria Groups Starch Dosea Volume Hemodynamics Difference) Mortality Other Outcomes
Li et al, China, Adults with Hematological Hypertonic saline 4% + 300–500 1446 vs Higher MAP at 1 h in Not reported 33% vs 67% Higher lactate clearance
200830 1 septic shock disorders, liver hydroxyethyl starch mL/30 min 2800 mL the hypertonic saline at 28 d in the hypertonic saline
or renal failure, (number of (P < .01) + hydroxyethyl starch (P = not + hydroxyethyl starch
pregnancy, patients = 15) group. significant)
expected early Hydroxyethyl starch No differences in
death, previous (number of vasopressors >33 d
cardiac arrest, patients = 15)
EE Systematic Review Article

heart failure
Zhu et al, China, Adults with Liver or kidney Hypertonic saline 4 mL/kg 5475 vs Higher MAP at 6 h in Not reported 2% vs 4% Higher urine output and
201132 1 severe failure, 7.5% + hydroxyethyl in 30 min 6383 mL the hypertonic saline not clear lactate clearance in

www.anesthesia-analgesia.org
sepsis disseminated starch at 24 h + hydroxyethyl starch at which the hypertonic saline
intravascular (number of (P < .01) group time point + hydroxyethyl starch
coagulation, patients = 45) (P = not group
expected early Hydroxyethyl starch significant)
death (number of
patients = 45)
Van Haren New Adults with Sodium <130 or Hypertonic saline 250 mL Not reported Not reported Not reported Not reported Lower gene expression
et al, Zealand, septic shock >150 mEq/L, 7.2% + hydroxyethyl in 15 min for matrix
201134 1 <24 h after arrhythmias, MI in starch metalloproteinase-9
intensive the last month, (number of and l-selectin in the
care unit pregnancy patients = 12) hypertonic saline +
admission Hydroxyethyl starch hydroxyethyl starch
(number of group.
patients = 12) No differences in the
cluster of differentiation
11b, interleukin-8,
interleukin-10, soluble
intercellular adhesion
molecule-1 or monocyte
chemoattractant
protein-1
Van Haren New Adults with Sodium <130 or Hypertonic saline 250 mL Decreased Higher MAP/ Higher sodium 25% vs 33% No differences in
et al, Zealand, septic shock >150 mEq/L, 7.2% + hydroxyethyl in 15 min need for norepinephrine, SV and chloride at hospital the sublingual
201235 1 <24 h after arrhythmias, MI starch ongoing fluids index, and systolic in the discharge microcirculation or
intensive in the last month, (number of (P = .046) velocity of the mitral hypertonic (P = not gastric tonometry
care unit pregnancy patients = 12) annulus in the saline + significant)
admission Hydroxyethyl starch hypertonic saline + hydroxyethyl
(number of hydroxyethyl starch starch group
patients = 12) No differences in HR, for 4 h.
MAP, CVP, or CO No differences
in pH
All comparisons are presented in the following order: hypertonic saline + hydroxyethyl starch versus hydroxyethyl starch groups.
Abbreviations: CO, cardiac output; CVP, central venous pressure; HR, heart rate; MAP, mean arterial pressure; MI, myocardial ischemia; SV, stroke volume.
a
Presented values represent the total volume given as a bolus of hypertonic saline + hydroxyethyl starch in the experimental cohort.

ANESTHESIA & ANALGESIA

Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
 Hypertonic Saline in Human Sepsis

Three randomized controlled trials reported equivalent

vasopressor doses when hypertonic saline was compared
with an isotonic crystalloid,30,33,40 and 1 randomized con-
trolled trial reported a higher mean arterial pressure-to-nor-
epinephrine dose ratio with the infusion of hypertonic saline
+ hydroxyethyl starch compared to hydroxyethyl starch
alone.35 One randomized controlled trial reported equivalent
cardiovascular sequential organ failure assessment scores
during the first 7 days of intensive care unit stay when hyper-
tonic saline was compared with an isotonic crystalloid.40

Electrolytes and Acid-Base Status

Five studies reported data on sodium and chloride levels. Four
showed a transient increase in their concentrations,33,35,39,40
and 1 reported no difference 2 hours after the hypertonic
saline infusion.29 One randomized controlled trial reported
that safety limits (sodium >155 mEq/L or sodium change in
24 hours >12 mEq/L) were reached more frequently (39.3%
vs 4.1%; P < .0001) when hypertonic saline was administered
compared to an isotonic crystalloid.40 However, in this ran-
domized controlled trial, repeated boluses were allowed dur-
ing the first 3 days in the intensive care unit, leading to the
infusion of a median (p25–p75) volume of 1.4 (0.6–2.0) L of
3% hypertonic saline in the intervention group.
Two studies reported no changes in blood pH after
hypertonic saline infusion.29,35

Figure 1. Estimated risk of bias for each included study.
when hypertonic saline + hydroxyethyl starch was compared
with hydroxyethyl starch alone,30,32 but no such differences
were reported in 4 other randomized controlled trials.29,33,35,39
Two randomized controlled trials evaluated cardiac func-
tion using echocardiography: 1 reported similar increases in
cardiac output with hypertonic saline and an isotonic crys-
talloid,29 and the other reported higher stroke volume and
tissue Doppler velocities with hypertonic saline + hydroxy-
ethyl starch than with hydroxyethyl starch alone.35
Renal Function and Coagulation
One randomized controlled trial reported a higher urine
output in patients receiving hypertonic saline + hydroxy-
ethyl starch than in those receiving just hydroxyethyl
starch.32 One randomized controlled trial reported no differ-
ences in the renal sequential organ failure assessment score
and in the need for renal replacement therapy in patients
after receiving hypertonic saline compared to an isotonic
crystalloid.40
No study reported any data on coagulation parameters,
bleeding, or transfusion.

Table 3.  Summary of Quality of the Evidence Using the GRADE Approach

Certainty of the
Outcome Impact No. of Participants (Studies) Evidence (GRADE)
Infused fluid volume Decreased infused volume in the cohort receiving 712 (6 randomized controlled trials) ⊕⊕⊕O
hypertonic saline moderatea,b
Hemodynamics A potential small benefit on hemodynamics with 774 (7 randomized controlled trials) ⊕OOO
hypertonic saline infusion. Harm was not detected very lowa,c
Electrolytes Transient increase in sodium and chloride levels after 572 (5 randomized controlled trials) ⊕⊕⊕O
a bolus of hypertonic saline moderated
Renal function No evidence for a deterioration in renal function after 524 (2 randomized controlled trials) ⊕⊕⊕O
hypertonic saline infusion moderatea,e
Immune system Hypertonic saline may have some impact on 24 (1 randomized controlled trial) ⊕OOO
immune function very lowa,f,g
Mortality No difference in mortality rates between hypertonic 774 (7 randomized controlled trials) ⊕⊕OO
saline and the control groups lowa,c,h
Abbreviation: GRADE, Grading of Recommendations Assessment, Development, and Evaluation.
a
Risk of inadequate blinding and its effects on the obtained outcome.
b
Largest trial did not show any differences at day 3.
c
Different reported outcomes in different studies.
d
One study did not show this response.
e
Only reported in 2 studies.
f
Only reported in 1 study.
g
Some biomarkers in a large panel of molecules and measurements highly dependent on manipulations.
h
Studies underpowered for this outcome.

June 2019 • Volume 128 • Number 6 www.anesthesia-analgesia.org 1179

Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
 EE Systematic Review Article

Immune System groups; however, the majority of studies were underpowered

In 1 randomized controlled trial, the hypertonic saline +
hydroxyethyl starch cohort had a lower gene expression
for matrix metalloproteinase-9 and for l-selectin than the
hydroxyethyl starch cohort but no changes in other inflam-
matory biomarkers (Table 2).34
Microcirculation and Gastric Tonometry
One randomized controlled trial reported no significant
differences in the sublingual microcirculation (using side-
stream dark field videomicroscopy) or gastric tonometry
variables between hypertonic saline + hydroxyethyl starch
and hydroxyethyl starch alone cohorts.35
Intensive Care Unit Length of Stay and Mortality
Two randomized controlled trials reported no significant
differences in the intensive care unit length of stay between
a hypertonic saline and an isotonic crystalloid cohort.33,40
Seven randomized controlled trials reported data on mor-
tality (Tables 1 and 2).29,30,32,33,35,39,40 None of the studies reported
a statistically significant difference in mortality rates between
for this outcome. A meta-analysis of the 7 studies showed no
differences in mortality rates between hypertonic saline and
other fluids (odds ratio, 0.946; 95% CI, 0.688–1.301; P = .733;
low-quality evidence; Figure  3). There was no significant
heterogeneity in this analysis (I2  =  0%). Subgroup analyses
of studies comparing only hypertonic saline with an isotonic
crystalloid or only hypertonic saline + hydroxyethyl starch
with hydroxyethyl starch alone also showed no significant
differences (Supplemental Digital Content, Figure S4, http://
links.lww.com/AA/C670). The Funnel plot showed some
degree of publication bias but not affecting the estimated
global effect on mortality (Supplemental Digital Content,
Figure S5, http://links.lww.com/AA/C670).
DISCUSSION
Use of hypertonic saline in the treatment of sepsis and
septic shock has been widely investigated in animal mod-
els,18–21,41–44 but data in humans remain limited. The major-
ity of studies we identified had small patient cohorts, and
relevant outcomes (renal function, coagulation, and other

Figure 2. Forest plots of studies reporting administered fluid volumes. HES indicates hydroxyethyl starch; HS, hypertonic saline;
NS, normal saline.

Figure 3. Forest plots of studies reporting mortality. HES indicates hydroxyethyl starch; HS, hypertonic saline; NS, normal saline.

1180   
www.anesthesia-analgesia.org ANESTHESIA & ANALGESIA
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
 Hypertonic Saline in Human Sepsis
adverse events) were not always reported, limiting their
potential conclusions. The main observation from our
review is that lower volumes of hypertonic saline than other
fluid types are required to obtain the desired hemodynamic
goals. As expected, the studies we reviewed observed a
transient increase in natremia and chloremia with hyper-
tonic saline but no associated safety issues (notably on
hemodynamics or renal function). In the few studies we
identified, there were no significant mortality differences
between hypertonic saline and other fluids.
About two-thirds of human body water is found in
the intracellular compartment.45 When hypertonic saline
is infused, it diffuses rapidly into the intravascular and
interstitial spaces because electrolytes can freely cross the
endothelial barrier. However, at the cellular membrane
level, electrolytes are unable to proceed; thus, the tonic-
ity of the extracellular compartment increases and water
escapes from the intracellular compartment.45,46 Various
animal studies in septic models confirm that less volume is
needed to obtain similar hemodynamic goals when using
hypertonic saline compared to an isotonic crystalloid,20,42
and human studies in healthy volunteers, which directly
measured the blood volume, have shown that hypertonic
saline is 4–5 times more efficient than a crystalloid solution
in terms of volume expansion.16,47 In our review, we found
a clear association between hypertonic saline infusion and
reduction in the total volume of fluids needed to achieve the
same hemodynamic effect.
Hypertonic saline can have a direct influence on endothe-
lial permeability. Some animal data suggest that hypertonic
saline can decrease the transcapillary flow of proteins in
hemorrhagic or septic states.19 Evidence suggests that hyper-
tonic saline can decrease sepsis-induced endothelial activa-
tion by reducing the expression of different chemotactic and
proinflammatory molecules or by modulating the function
of different immune cells.18,41,48,49 Ding et al36 suggested some
modulation of the inflammatory response with infusion of
hypertonic saline, but the lack of a control group limits the
interpretation of their findings. Perhaps more relevant is the
study by Van Haren et al,34 which showed that hypertonic
saline could inhibit the gene expression of matrix metallo-
proteinase-9 and l-selectin. Importantly, patients with sepsis
may have either a predominantly proinflammatory or an
anti-inflammatory state, and data on the role of hypertonic
saline for different phenotypes of sepsis do not exist.
Different animal studies have also suggested that hyper-
tonic saline can preserve the microcirculation in shock
states,50 decreasing the apoptosis and edema of endothelial
cells,51 inducing selective arteriolar vasodilatation while
improving tissue perfusion46 or interfering with the rheo-
logical behavior of red blood cells.52 Only 1 human study
has evaluated the effects of hypertonic saline on the sublin-
gual microcirculation, and it reported no significant differ-
ences compared to isotonic saline.35 However, the sample
size of this study was small, and the patients studied had
no significant alterations in capillary density at base-
line, making it difficult to show improvement with fluid
administration.
The infusion of hypertonic saline generates an impor-
tant stimulus (via different osmoreceptors) to the poste-
rior pituitary gland, leading to vasopressin secretion.21,53,54
June 2019 • Volume 128 • Number 6 www.anesthesia-analgesia.org 1181
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Vasopressin decreases water loss by the kidneys through
the V2 receptors and simultaneously increases vascular tone
through the V1 receptors.55 The vascular effects of vasopres-
sin may contribute substantially to the hemodynamic effects
of hypertonic saline (mainly when shock is present), as sug-
gested by the lack of increase in arterial pressure following a
bolus of hypertonic saline after blockade of the V1 receptor
in endotoxemic rats.21 In our review, we found no marked
change in arterial pressure after a bolus of hypertonic saline,
but none of the included randomized controlled trials evalu-
ated a possible role of vasopressin in the obtained responses.
It is interesting to note that in recent single-cohort studies in
which plasma vasopressin concentrations were measured
before and after a bolus of hypertonic saline, an abnormal
response was detected in around half of the patients.31,37,38
In these studies, this alteration was evident from 24 hours
after the diagnosis of sepsis38 and persisted for ≤5 days after
discontinuation of vasopressors.37 In 1 study, patients with
impaired osmoregulation had an increased intensive care
unit length of stay and mortality.38
The protocols used to infuse hypertonic saline in our
included randomized controlled trials, and the doses admin-
istered were heterogeneous. For a hypothetical adult patient
of 70 kg, the proposed quantity of chloride or sodium to
be infused during each bolus of hypertonic saline (that will
preferentially remain in the extracellular compartment)
would vary between 20929 and 539 mEq.33 This electrolyte
load is quite considerable when considering that, for such
a subject, the estimated chloride content in the extracellu-
lar space is already around 1750 mEq, with only 230 mEq
in the intracellular space. While most of the studies used
only 1 fluid bolus of hypertonic saline, the large random-
ized controlled trial by Asfar et al40 exposed the interven-
tional group to repeated boluses of hypertonic saline during
the first 3 days in the intensive care unit. Unfortunately, this
strategy led to the experimental protocol being stopped in
39% of the patients because the sodium concentration was
>155 mEq/L or increased by >12 mEq/L in 24 hours. Any
solution is potentially harmful if given in excess or if admin-
istered to high-risk populations. Animal and human studies
have shown that a bolus of hypertonic saline causes minor
alterations in the acid-base status when renal function is
normal,56,57 but the presence of hyperchloremia added to
renal failure, and the related metabolic acidosis may be
problematic.58–60 Nevertheless, administration of hypertonic
saline was generally well tolerated in our studies, with a
transient increase in the sodium and chloride concentration
levels and no major changes in the acid-base balance.
Hypertonic saline was associated with renal vasocon-
striction in a canine denervated and auto-transplanted
kidney model.22 We recently infused equivalent isosmotic
doses of normal or hypertonic saline in an animal model
of peritonitis. Renal function did not deteriorate despite
the development of hyperchloremia, and survival time was
prolonged in animals that received hypertonic saline.61 In
their large randomized controlled trial, Asfar et al40 did
not report any increase in the rate of renal adverse events
or a decrease in kidney function despite administration of
large doses of hypertonic saline. Future studies in this field
should routinely report effects on renal function.
 EE Systematic Review Article
Hyperchloremic acidosis can also alter the coagula-
tion system. Animal studies suggest that, compared to
Ringer’s lactate, normal saline can induce a coagulopathic
state.24 A recent meta-analysis in humans (mainly includ-
ing patients in the operating room) found an increase in
the estimated blood loss or the need for red blood cell
transfusion in patients who received normal saline, espe-
cially in high-risk populations.25 No studies in our analy-
sis reported on coagulation system outcomes, highlighting
the need for future trials on the effects of hypertonic saline
on coagulation.
The available data in our review show no statistically
significant differences in clinically important outcomes,
such as mortality, but the studies were generally small and
underpowered for this outcome, so an effect on mortality
cannot be ruled out. Unfortunately, identifying clinical sig-
nals in heterogeneous groups of critically ill patients can be
extremely challenging because some patients may benefit
from the intervention while others are harmed.62 Studies that
have evaluated the role of hypertonic saline in resuscitation
from hemorrhagic shock have reported no beneficial effect
on survival in the whole population,63–65 although some
post hoc analyses have suggested that patients with hypo-
tension may have benefited, and those not needing blood
transfusions in the first 24 hours may have been harmed by
the hypertonic saline infusion.65–67 Future research is needed
to identify and target those patients most likely to benefit
from hypertonic saline.
Our study has several limitations. First, the included stud-
ies were considerably heterogeneous in design. Although
we used a conservative approach in our forest plots by using
random effects models and calculated the degree of hetero-
geneity for each analysis, all the studies had important dif-
ferences in their protocols (infused solutions, observation
periods, studied outcomes, etc). Second, we combined data
from adult and pediatric randomized controlled trials in our
forest plots, although they represent different populations.
Nevertheless, the individual studies all showed similar
trends in fluid balance and mortality independent of popu-
lation age. Finally, in general, the risk of bias in the included
studies was high. However, this finding is the result of our
extensive literature search (including studies in foreign
languages) retrieving several studies that did not clearly
describe the methodology. It is also important to recognize
that completely blinding the medical staff to the type of fluid
in these studies is difficult because close observation of chlo-
ride levels will unmask the intervention cohort.
CONCLUSIONS
Hypertonic saline infusion in patients with sepsis is asso-
ciated with the need for reduced fluid volumes compared
to other solutions to achieve the same hemodynamic goals.
Hypertonic saline administration seems to be safe. More
research is needed to determine whether specific subpopu-
lations of patients with sepsis are more likely to benefit from
hypertonic saline administration than others. E
DISCLOSURES
Name: Diego Orbegozo, MD.
Contribution: This author helped design the study, helped perform
the literature search and extract the data, wrote the first draft of the
manuscript, and read and approved the final version.
1182   
www.anesthesia-analgesia.org
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
Name: Jean-Louis Vincent, MD, PhD.
Contribution: This author helped design the study, critically
reviewed the manuscript for intellectual content, and read and
approved the final version.
Name: Jacques Creteur, MD, PhD.
Contribution: This author helped design the study, critically
reviewed the manuscript for intellectual content, and read and
approved the final version.
Name: Fuhong Su, MD, PhD.
Contribution: This author helped design the study, helped per-
form the literature search and extract the data, critically reviewed
the manuscript for intellectual content, and read and approved the
final version.
This manuscript was handled by: Avery Tung, MD, FCCM.
REFERENCES
1. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo
J, Pinsky MR. Epidemiology of severe sepsis in the United
States: analysis of incidence, outcome, and associated costs of
care. Crit Care Med. 2001;29:1303–1310.
2. Tiru B, DiNino EK, Orenstein A, et al. The economic and
humanistic burden of severe sepsis. Pharmacoeconomics.
2015;33:925–937.
3. Friedman G, Silva E, Vincent JL. Has the mortality of septic
shock changed with time. Crit Care Med. 1998;26:2078–2086.
4. Levy MM, Dellinger RP, Townsend SR, et al; Surviving Sepsis
Campaign. The Surviving Sepsis Campaign: results of an inter-
national guideline-based performance improvement program
targeting severe sepsis. Crit Care Med. 2010;38:367–374.
5. Asfar P, Meziani F, Hamel JF, et al; SEPSISPAM Investigators.
High versus low blood-pressure target in patients with septic
shock. N Engl J Med. 2014;370:1583–1593.
6. De Backer D, Biston P, Devriendt J, et al; SOAP II Investigators.
Comparison of dopamine and norepinephrine in the treatment
of shock. N Engl J Med. 2010;362:779–789.
7. Fleck A, Raines G, Hawker F, et al. Increased vascular perme-
ability: a major cause of hypoalbuminaemia in disease and
injury. Lancet. 1985;1:781–784.
8. Holcroft JW, Trunkey DD, Carpenter MA. Extravasation of
albumin in tissues of normal and septic baboons and sheep. J
Surg Res. 1979;26:341–347.
9. Levy B, Collin S, Sennoun N, et al. Vascular hyporesponsive-
ness to vasopressors in septic shock: from bench to bedside.
Intensive Care Med. 2010;36:2019–2029.
10. Vincent JL, Sakr Y, Sprung CL, et al; Sepsis Occurrence in
Acutely Ill Patients Investigators. Sepsis in European inten-
sive care units: results of the SOAP study. Crit Care Med.
2006;34:344–353.
11. Boyd JH, Forbes J, Nakada TA, Walley KR, Russell JA. Fluid
resuscitation in septic shock: a positive fluid balance and ele-
vated central venous pressure are associated with increased
mortality. Crit Care Med. 2011;39:259–265.
12. Payen D, de Pont AC, Sakr Y, Spies C, Reinhart K, Vincent JL;
Sepsis Occurrence in Acutely Ill Patients (SOAP) Investigators.
A positive fluid balance is associated with a worse outcome in
patients with acute renal failure. Crit Care. 2008;12:R74.
13. Sirvent JM, Ferri C, Baró A, Murcia C, Lorencio C. Fluid balance
in sepsis and septic shock as a determining factor of mortality.
Am J Emerg Med. 2015;33:186–189.
14. 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 analysis of all-cause mortal-
ity. BMJ. 2014;349:g4561.
15. Perel P, Roberts I, Ker K. Colloids versus crystalloids for fluid
resuscitation in critically ill patients. Cochrane Database Syst Rev.
2013;2:CD000567.
16. Drobin D, Hahn RG. Kinetics of isotonic and hypertonic plasma
volume expanders. Anesthesiology. 2002;96:1371–1380.
17. Oliveira RP, Weingartner R, Ribas EO, Moraes RS, Friedman G.
Acute haemodynamic effects of a hypertonic saline/dextran
solution in stable patients with severe sepsis. Intensive Care
Med. 2002;28:1574–1581.
18. Pascual JL, Khwaja KA, Ferri LE, et al. Hypertonic saline
resuscitation attenuates neutrophil lung sequestration and
ANESTHESIA & ANALGESIA
 Hypertonic Saline in Human Sepsis
transmigration by diminishing leukocyte-endothelial interac-
tions in a two-hit model of hemorrhagic shock and infection. J
Trauma. 2003;54:121–130.
19. de Carvalho H, Matos JA, Bouskela E, Svensjö E. Vascular per-
meability increase and plasma volume loss induced by endo-
toxin was attenuated by hypertonic saline with or without
dextran. Shock. 1999;12:75–80.
20. Rahal L, Garrido AG, Cruz RJ Jr, Silva E, Poli-de-Figueiredo LF.
Fluid replacement with hypertonic or isotonic solutions guided
by mixed venous oxygen saturation in experimental hypody-
namic sepsis. J Trauma. 2009;67:1205–1212.
21. Giusti-Paiva A, Martinez MR, Bispo-da-Silva LB, Salgado MC,
Elias LL, Antunes-Rodrigues J. Vasopressin mediates the pres-
sor effect of hypertonic saline solution in endotoxic shock.
Shock. 2007;27:416–421.
22. Wilcox CS. Regulation of renal blood flow by plasma chloride. J
Clin Invest. 1983;71:726–735.
23. Reid F, Lobo DN, Williams RN, Rowlands BJ, Allison SP.
(Ab)normal saline and physiological Hartmann’s solution:
a randomized double-blind crossover study. Clin Sci (Lond).
2003;104:17–24.
24. Kiraly LN, Differding JA, Enomoto TM, et al. Resuscitation with
normal saline (NS) vs lactated ringers (LR) modulates hyperco-
agulability and leads to increased blood loss in an uncontrolled
hemorrhagic shock swine model. J Trauma. 2006;61:57–64.
25. Orbegozo Cortés D, Rayo Bonor A, Vincent JL. Isotonic crystal-
loid solutions: a structured review of the literature. Br J Anaesth.
2014;112:968–981.
26. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group.
Preferred reporting items for systematic reviews and meta-
analyses: the PRISMA statement. PLoS Med. 2009;6:e1000097.
27. Higgins JPT, Sterne JAC, Savovic J, et al. A revised tool for
assessing risk of bias in randomized trials. Cochrane Database of
Systematic Reviews. 2016;10(Suppl 1):29–31.
28. Muller L, Lefrant JY, Jaber S, et al. Short term effects of hyper-
tonic saline during severe sepsis and septic shock [in French].
Ann Fr Anesth Reanim. 2004;23:575–580.
29. Fang ZX, Li YF, Zhou XQ, et al. Effects of resuscitation with
crystalloid fluids on cardiac function in patients with severe
sepsis. BMC Infect Dis. 2008;8:50.
30. Li F, Sun H, Han XD. The effect of different fluids on early fluid
resuscitation in septic shock [in Chinese]. Zhongguo Wei Zhong
Bing Ji Jiu Yi Xue. 2008;20:472–475.
31. Siami S, Bailly-Salin J, Polito A, et al. Osmoregulation of vaso-
pressin secretion is altered in the postacute phase of septic
shock. Crit Care Med. 2010;38:1962–1969.
32. Zhu GC, Quan ZY, Shao YS, Zhao JG, Zhang YT. The study
of hypertonic saline and hydroxyethyl starch treating severe
sepsis [in Chinese]. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue.
2011;23:150–153.
33. Chopra A, Kumar V, Dutta A. Hypertonic versus normal saline
as initial fluid bolus in pediatric septic shock. Indian J Pediatr.
2011;78:833–837.
34. van Haren FM, Sleigh J, Cursons R, La Pine M, Pickkers P, van
der Hoeven JG. The effects of hypertonic fluid administration
on the gene expression of inflammatory mediators in circu-
lating leucocytes in patients with septic shock: a preliminary
study. Ann Intensive Care. 2011;1:44.
35. van Haren FM, Sleigh J, Boerma EC, et al. Hypertonic fluid
administration in patients with septic shock: a prospective ran-
domized controlled pilot study. Shock. 2012;37:268–275.
36. Ding WW, Li WQ, Tong ZH, Li N, Li JS. The immunomodula-
tory effects of hypertonic saline on sepsis patients [in Chinese].
Zhongguo Wei Zhong Bing Ji Jiu Yi Xue. 2012;24:465–469.
37. Siami S, Polito A, Porcher R, et al. Thirst perception and osmo-
regulation of vasopressin secretion are altered during recovery
from septic shock. PLoS One. 2013;8:e80190.
38. Zhou Q, Yang X, Sun J, Wang C, Li D. Prognostic value of
decreased vasopressin modulation in the late-phase of septic
shock patients [in Chinese]. Zhonghua Wei Zhong Bing Ji Jiu Yi
Xue. 2014;26:706–709.
39. Liu S, Ren X, Gun L, Zhang Q, Zhang J, Zhu Y. Effect of 3%
hypertonic saline as early fluid resuscitation in pediatric septic
shock [in Chinese]. Zhonghua Er Ke Za Zhi. 2015;53:599–604.
June 2019 • Volume 128 • Number 6 www.anesthesia-analgesia.org 1183
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
40. Asfar P, Schortgen F, Boisramé-Helms J, et al; HYPER2S
Investigators; REVA Research Network. Hyperoxia and hyper-
tonic saline in patients with septic shock (HYPERS2S): a two-
by-two factorial, multicentre, randomised, clinical trial. Lancet
Respir Med. 2017;5:180–190.
41. Shih CC, Chen SJ, Chen A, Wu JY, Liaw WJ, Wu CC. Therapeutic
effects of hypertonic saline on peritonitis-induced septic shock
with multiple organ dysfunction syndrome in rats. Crit Care
Med. 2008;36:1864–1872.
42. Garrido Adel P, Cruz RJJr, Poli de Figueiredo LF, Rocha e Silva M.
Small volume of hypertonic saline as the initial fluid replacement
in experimental hypodynamic sepsis. Crit Care. 2006;10:R62.
43. Weeren FR, Tobias TA, Schertel ER, Allen DA, Brourman
JD. Comparative effects of 7% NaCl in 6% dextran 70 and
0.9% NaCl on oxygen transport in endotoxemic dogs. Shock.
1994;1:159–165.
44. Wang YL, Chen JH, Zhu QF, et al. In vivo evaluation of the ame-
liorating effects of small-volume resuscitation with four differ-
ent fluids on endotoxemia-induced kidney injury. Mediators
Inflamm. 2015;2015:726243.
45. De Backer D, Cortés DO. Characteristics of fluids used for intra-
vascular volume replacement. Best Pract Res Clin Anaesthesiol.
2012;26:441–451.
46. Mazzoni MC, Borgström P, Arfors KE, Intaglietta M. Dynamic
fluid redistribution in hyperosmotic resuscitation of hypovole-
mic hemorrhage. Am J Physiol. 1988;255:H629–H637.
47. Svensén C, Hahn RG. Volume kinetics of Ringer solution, dex-
tran 70, and hypertonic saline in male volunteers. Anesthesiology.
1997;87:204–212.
48. Huang GS, Shih CM, Wu CC, et al. Hypertonic saline, mannitol
and hydroxyethyl starch preconditioning of platelets obtained
from septic patients attenuates CD40 ligand expression in vitro.
J Trauma. 2010;68:331–336.
49. Shields CJ, O’Sullivan AW, Wang JH, Winter DC, Kirwan
WO, Redmond HP. Hypertonic saline enhances host response
to bacterial challenge by augmenting receptor-independent
neutrophil intracellular superoxide formation. Ann Surg.
2003;238:249–257.
50. Pascual JL, Ferri LE, Seely AJ, et al. Hypertonic saline resuscita-
tion of hemorrhagic shock diminishes neutrophil rolling and
adherence to endothelium and reduces in vivo vascular leak-
age. Ann Surg. 2002;236:634–642.
51. Qian Y, Du YH, Tang YB, et al. ClC-3 chloride channel prevents
apoptosis induced by hydrogen peroxide in basilar artery
smooth muscle cells through mitochondria dependent path-
way. Apoptosis. 2011;16:468–477.
52. Zhao L, Wang B, You G, Wang Z, Zhou H. Effects of different
resuscitation fluids on the rheologic behavior of red blood cells,
blood viscosity and plasma viscosity in experimental hemor-
rhagic shock. Resuscitation. 2009;80:253–258.
53. Liard JF, Dolci W, Vallotton MB. Plasma vasopressin levels after
infusions of hypertonic saline solutions into the renal, portal,
carotid, or systemic circulation in conscious dogs. Endocrinology.
1984;114:986–991.
54. Yesberg NE, Henderson M, Budtz-Olsen OE. The effect of
intravenous hypertonic saline infusion on renal function and
vasopressin excretion in sheep. Q J Exp Physiol Cogn Med Sci.
1978;63:331–339.
55. Sharshar T, Annane D. Endocrine effects of vasopressin in criti-
cally ill patients. Best Pract Res Clin Anaesthesiol. 2008;22:265–273.
56. Wan L, Bellomo R, May CN. The effects of normal and hyper-
tonic saline on regional blood flow and oxygen delivery. Anesth
Analg. 2007;105:141–147.
57. Kolsen-Petersen JA, Nielsen JO, Tonnesen E. Acid base and
electrolyte changes after hypertonic saline (7.5%) infusion:
a randomized controlled clinical trial. Scand J Clin Lab Invest.
2005;65:13–22.
58. Zhou F, Peng ZY, Bishop JV, Cove ME, Singbartl K, Kellum JA.
Effects of fluid resuscitation with 0.9% saline versus a balanced
electrolyte solution on acute kidney injury in a rat model of sep-
sis. Crit Care Med. 2014;42:e270–e278.
59. Shaw AD, Raghunathan K, Peyerl FW, Munson SH,
Paluszkiewicz SM, Schermer CR. Association between intra-
venous chloride load during resuscitation and in-hospital
 EE Systematic Review Article
mortality among patients with SIRS. Intensive Care Med. 2014;40:
1897–1905.
60. Orbegozo D, Su F, Santacruz C, et al. Effects of different crys-
talloid solutions on hemodynamics, peripheral perfusion,
and the microcirculation in experimental abdominal sepsis.
Anesthesiology. 2016;125:744–754.
61. Su F, Xie K, He X, et al. The harmful effects of hypertonic
sodium lactate administration in hyperdynamic septic shock.
Shock. 2016;46:663–671.
62. Vincent JL. We should abandon randomized controlled trials in
the intensive care unit. Crit Care Med. 2010;38:S534–S538.
63. Younes RN, Aun F, Accioly CQ, Casale LP, Szajnbok I, Birolini
D. Hypertonic solutions in the treatment of hypovolemic shock:
a prospective, randomized study in patients admitted to the
emergency room. Surgery. 1992;111:380–385.
1184   
www.anesthesia-analgesia.org
Copyright © 2019 International Anesthesia Research Society. Unauthorized reproduction of this article is prohibited.
64. Vassar MJ, Fischer RP, O’Brien PE, et al; The Multicenter Group
for the Study of Hypertonic Saline in Trauma Patients. A mul-
ticenter trial for resuscitation of injured patients with 7.5%
sodium chloride: the effect of added dextran 70. Arch Surg.
1993;128:1003–1011.
65. Bulger EM, May S, Brasel KJ, et al; ROC Investigators. Out-
of-hospital hypertonic resuscitation following severe trau-
matic brain injury: a randomized controlled trial. JAMA.
2010;304:1455–1464.
66. Younes RN, Aun F, Ching CT, et al. Prognostic factors to predict
outcome following the administration of hypertonic/hyperon-
cotic solution in hypovolemic patients. Shock. 1997;7:79–83.
67. Dubick MA, Shek P, Wade CE. ROC trials update on prehos-
pital hypertonic saline resuscitation in the aftermath of the
US-Canadian trials. Clinics (Sao Paulo). 2013;68:883–886.
ANESTHESIA & ANALGESIA