Review Article

Osmotherapy for
Address correspondence to
Dr Matthew Fink, Weill Cornell
Medical College, Department of
Neurology and Neuroscience,
525 East 68th Street, F-610,
New York, NY 10065,
mfink@med.cornell.edu.
Intracranial
Relationship Disclosure:
Dr Fink serves as a consultant
for MAQUET/Datascope and
Hypertension: Mannitol
Proctor & Gamble and serves as
an advisor for Navigant
Consulting, Inc. Dr Fink serves
as the editor of Neurology Alert.
Versus Hypertonic Saline
Unlabeled Use of Matthew E. Fink, MD, FAAN
Products/Investigational
Use Disclosure:
Dr Fink reports no disclosure.
* 2012, American Academy of
ABSTRACT
Neurology. Purpose of Review: Hyperosmolar therapy is one of the core medical treatments for
brain edema and intracranial hypertension, but controversy exists regarding the use
of the most common agents, mannitol, and hypertonic saline. This article describes
the relative merits and adverse effects of these agents using the best available clinical
evidence.
Recent Findings: Mannitol is effective and has been used for decades in the treat-
ment of traumatic brain injury, but it may precipitate acute renal failure if serum
osmolarity exceeds 320 mOsm/L. Hypertonic saline appears to be safe, and serum
sodium has been elevated to as high as 180 mEq/L in clinical settings without sig-
nificant neurologic, cardiac, or renal injury. In small comparative trials both agents are
effective and no clinically significant difference has been noted, but a properly pow-
ered trial has not yet been performed.
Summary: Both mannitol and hypertonic saline are effective and have an acceptable
risk profile for use in the treatment of elevated intracranial pressure secondary to
brain edema.

Continuum Lifelong Learning Neurol 2012;18(3):640–654.

THE PROBLEM: CEREBRAL to progressive intracranial hypertension,

EDEMA AND RAISED
INTRACRANIAL PRESSURE
Cerebral edema and raised intracranial
pressure (ICP) are some of the most
common problems encountered in the
practice of neurologic critical care. These
problems are the consequences of acute
brain injuries as well as chronic brain
disorders, including ischemic stroke,
intracranial hemorrhage, traumatic brain
injury, subarachnoid hemorrhage, brain
infections, and intracranial tumors.
Raised ICP is recognized as one of the
major causes of mortality, as well as poor
neurologic outcome, in many of these
conditions. Because severe brain edema
that is not successfully treated can lead
640 www.aan.com/continuum
cerebral ischemia, brain herniation,
and eventual progression to death, the
identification and treatment of these
conditions is a major effort among neu-
rocritical care specialists (Table 8-1).1
The purpose of this article is to
discuss the use of the hyperosmolar
agents mannitol and hypertonic saline,
which are core therapies for the treat-
ment of raised ICP, but not to give an
exhaustive discussion of the manage-
ment of raised ICP. Figure 8-12,3 is a
brief algorithm that is derived from the
Brain Trauma Foundation’s Guidelines
for the Management of Severe Trau-
matic Brain Injury and may be appli-
cable to other conditions. The reader is
June 2012

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 KEY POINTS
a h Increased intracranial
TABLE 8-1 Causes of Intracranial Hypertension
pressure is one of the
Primary (Intracranial) Secondary (Extracranial) most common problems
encountered in the
Nontraumatic Airway obstruction
neurocritical care unit
Intracranial hemorrhages Hypoventilation and must be identified
(eg, parenchymal, subarachnoid, early.
subdural, epidural)
h Osmotic agents such as
Ischemic infarction Hypoxia mannitol and hypertonic
Hydrocephalus (communicating Hypercarbia saline are most effective
and noncommunicating) in conditions causing
vasogenic edema.
Brain edema Head position or posture
Brain tumor Venous outflow obstruction
Status epilepticus Hyperpyrexia
Cerebral venous thrombosis Hyponatremia
Cerebral vasospasm Agitation, pain
Infection (eg, encephalitis, Diabetic ketoacidosis
meningitis, abscess, etc)
Traumatic Eclampsia or hypertensive
encephalopathy
Mass lesion (eg, epidural or Convulsive or nonconvulsive seizure
subdural hematomas,
hemorrhagic contusions)
Hydrocephalus Increased intrathoracic or intra-abdominal
pressure (eg, Valsalva maneuvers,
mechanical ventilation setup)
Diffuse brain edema Fulminant hepatic encephalopathy
Depressed skull fracture High-altitude cerebral edema
Drugs (eg, lead, tetracycline,
doxycycline, retinoic acid)
a
Adapted from Mariano GSL, Fink ME, Hoffman C, Rosengart AJ. Intracranial pressure: monitoring and
management. In: Hall J, Schmidt G, Wood L, editors. Principles of critical care. 4th ed. New York, NY: McGraw-
Hill, 2012.1 B 2012, with permission from The McGraw-Hill Companies, Inc.

referred to other references for further brain contusions. Disruption of the

information.4Y6
Two forms of cerebral edema are
generally recognized: vasogenic and
cytotoxic. Vasogenic edema is defined
as excess fluid within the interstitial
space due to blood-brain barrier (BBB)
disruption from a variety of pathologic
causes, including malignant brain tu-
mors or brain abscesses, intracranial
hemorrhages, direct surgical manipula-
tion, meningitis or encephalitis, and
BBB allows diffusion of water into the
interstitial space. The presence of an
intact or partially intact BBB is what
allows an osmotic agent to help remove
brain water, reduce ICP, and lessen the
harmful effects of cerebral edema.7,8
Cytotoxic edema, defined as excess
fluid within the intracellular space, com-
monly occurs with both focal and global
acute brain ischemia as well as severe
metabolic disorders such as fulminant

Continuum Lifelong Learning Neurol 2012;18(3):640–654 www.aan.com/continuum 641

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Osmotherapy

FIGURE 8-1 Intracranial pressure management protocol for

traumatic brain injury.
2
Data from Brain Trauma Foundation.
Modified from Clausen T, Bullock R. Medical treatment and neuroprotection in traumatic
3
brain injury. Curr Pharm Des 2001;7(15):1519. B 2001, with permission from Bentham
Science Publishers. www.benthamdirect.org/pages/content.php?CPD/2001/00000007/
00000015/0004B.SGM.

hepatic failure.9 It is unclear whether sult in cytotoxic edema because the

osmotic agents have a therapeutic role osmotic gradient is disrupted across all
to play in pathologic conditions that re- cell membranes and the BBB fails to

642 www.aan.com/continuum June 2012

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.


maintain an osmotic gradient.10 It is rare
to have a pure form of cerebral edema,
however; most clinical situations involve
a combination of both vasogenic and
cytotoxic edema, and it is often difficult
to separate the two. A special case is the
extreme water intoxication that results in
severe hyponatremia and diffuse brain
edema, but that is the result of passive
movement of water down a diffusion
gradient into the intracellular space.
OSMOLALITY AND OSMOLARITY
In order to better understand the role
of osmotic agents in the treatment of
edema and raised ICP, a few definitions
are necessary. Osmolarity refers to the
osmotic concentration of a solution ex-
pressed as osmoles of solute per liter of
solution. Osmolality refers to the osmo-
tic concentration of a solution expressed
as osmoles of solute per kilogram of so-
lution. The formula to calculate osmo-
lality is mOsm/kg = (Na  2 + glucose/
18) + (blood urea nitrogen [BUN]/2.3).
In this calculated formula, the sodium
concentration is measured in mEq/L,
and glucose and BUN are measured in
mg/dL. In clinical situations, measures of
serum osmolality and serum osmolarity
are almost identical. A more important
determination is the osmolar gap (OG),
which is the difference between the
calculated osmolality and the measured
osmolarity; it is particularly important
when using mannitol.11 The term os-
motic pressure refers to the pressure
necessary to prevent osmosis into a
solution when it is separated from the
pure water by a semipermeable mem-
brane. Osmotic pressure (mm Hg) is
calculated by the formula 19.3  osmo-
lality (mOsm/kg). Hypertonicity refers
to the ability of a hyperosmolar solution
to redistribute fluid from the intracellu-
lar to the extracellular compartment. In
clinical situations, this requires an intact
or partially intact BBB, which functions
like a semipermeable membrane.
Continuum Lifelong Learning Neurol 2012;18(3):640–654
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
KEY POINTS
PHYSIOLOGY OF INTRACRANIAL
h Osmolality can be
HYPERTENSION calculated, but
Cerebral edema may exist with or with- osmolarity is directly
out intracranial hypertension, depend- measured. The osmolar
ing on the extent and magnitude of the gap is an important
edema. Intracranial hypertension is measurement to
defined as a sustained ICP of greater monitor.
than 20 mm Hg. Normal ICP ranges h Cerebral perfusion
between 5 cm H2O and 20 cm H2O or pressure equals mean
3 mm Hg and 15 mm Hg. Bedside arterial pressure minus
monitoring equipment measures ICP in intracranial pressure and
mm Hg, although measurement of should be maintained
pressure with a manometer during between 50 mm Hg and
70 mm Hg.
lumbar puncture reports ICP in cm
H2O. The conversion factor is 1.36 cm
H2O equals 1 mm Hg. Cerebral perfu-
sion pressure (CPP) equals the mean
arterial pressure minus ICP in mm Hg.
Normal CPP is between 50 mm Hg and
70 mm Hg. In a healthy adult, CPP is
maintained constant at mean arterial
pressures between 50 mm Hg and 150
mm Hg because of cerebral autoregu-
lation. If autoregulation is impaired, an
elevation in mean arterial pressure will
cause a rise in CPP and often a rise in
ICP as well. Figure 8-21 depicts the clas-
sic volume-pressure compliance curve
that demonstrates the relationship be-
tween intracranial contents and in-
creased ICP. Any additional mass added
to the cranial cavity, such as blood or
neoplasm, brain edema, or excess CSF,
can result in a rise in ICP according to
the relationships between pressure and
volume. Figure 8-31 in Case 8-1 is an
example of progressive brain swelling
and herniation from a large infarct in
spite of maximum medical therapy.12
The role of osmotic therapy with
either mannitol or hypertonic saline is
based on the principle that these agents
will help to remove water from brain
tissue across an intact BBB by creating an
osmotic gradient that draws water from
the interstitium into the vascular space.
In general, it is thought that these agents
are more effective in clinical situations
that result in vasogenic edema rather
www.aan.com/continuum 643
 Osmotherapy

KEY POINT
h Osmotic agents lower
intracranial pressure by
removing brain water
when the blood-brain
barrier is intact.

FIGURE 8-2 Intracranial compliance curve. The curve

represents brain compliance as it relates to
intracranial pressure (ICP) and volume.
As volume increases (edema, blood, CSF, mass lesion), a
compensatory response maintains a normal ICP (A). Additional
volume is tolerated with a rise in ICP (B), but at a critical point
compensation is exhausted and the ICP rises exponentially to
dangerous levels (C ).
Modified from Mariano GSL, Fink ME, Hoffman C, Rosengart AJ. Intracranial
pressure: monitoring and management. In: Hall J, Schmidt G, Wood L, editors.
1
Principles of critical care. 4th ed. New York, NY: McGraw-Hill, 2012. B 2012,
with permission from The McGraw-Hill Companies, Inc.

than cytotoxic edema. It is also quite pos-
sible that prolonged infusions of osmotic
agents may worsen cerebral edema by
crossing the damaged BBB into the in-
terstitial and intracellular spaces and
causing an increase in brain edema.
Mannitol has been used as a treat-
ment for traumatic brain injury since the
1960s. Despite its long history of use, no
Class I evidence confirms its benefit in
improving neurologic outcome, and no
definitive randomized clinical trials have
been performed.13 The use of hyper-
tonic saline has increased in recent
years, primarily because of fear of acute
renal failure from repeated doses of
mannitol and the problems associated
with rising mannitol concentrations in
the blood. Again, however, no Class I
evidence supports the use of hyper-
tonic saline as an agent to reduce ICP
or brain edema, and no large compa-
rative trials comparing mannitol and
hypertonic saline in these clinical sit-
uations have been performed.14 Never-
theless, it is worth discussing the
relative merits of these two agents to
644 www.aan.com/continuum
determine which agent should be se-
lected in different clinical situations.
Major factors contributing to the diffi-
culty in gathering Class I evidence for
hyperosmolar therapy include the wide
variety of etiologies that cause elevated
ICP and the variability of treatment
algorithms used by various institutions
and clinicians. In addition, it is ex-
tremely difficult to organize and per-
form randomized, controlled clinical
trials in any group of critically ill pa-
tients to obtain such information.15,16
OSMOTIC THERAPY WITH
MANNITOL
Mannitol is a sugar alcohol that is not
significantly metabolized and is ex-
creted unchanged in the urine after IV
infusion. It is filtered by the kidney at
the glomerulus and is reabsorbed in the
distal nephron, thereby acting as a
potent osmotic diuretic. The half-life of
mannitol in the blood is affected by the
glomerular filtration rate but averages
between 39 and 103 minutes.
June 2012

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Case 8-1
A 74-year-old right-handed woman with
chronic atrial fibrillation who was poorly
adherent to warfarin was admitted to the
hospital with an acute right middle cerebral
artery territory infarction. She was found
lying on the floor at home alone. The time of
onset was unknown, preventing the use
of any thrombolytic therapies. Head CT scans
(Figure 8-3A and Figure 8-3B) were obtained
approximately 48 hours after the ischemic
infarction. There were local mass effects but
still some remaining compressible CSF spaces
(ventricular system, basal cisterns), indicating
reduced, but not exhausted, intracranial FIGURE 8-3 Head CT scans (A and B) obtained 48 hours
after an acute right middle cerebral artery
compliance. Intermittent mannitol boluses at ischemic infarction. There are local mass
a dose of 0.5 g/kg were administered every effects but still some remaining compressible CSF spaces
(ventricular system, basal cisterns), indicating reduced but not
4 hours. Repeat CT scans taken at 96 hours exhausted intracranial compliance. Repeat CT scans taken at
(Figure 8-3C and Figure 8-3D) showed 96 hours (C and D) show compression of all available spaces
compression of all available spaces with with subfalcial herniation and secondary infarction in the
territory of the right anterior cerebral artery (arrow on panel D).
subfalcial herniation and secondary infarction
in the territory of the right anterior cerebral ICP = intracranial pressure.
artery (arrow on panel D). This combination of Modified from Mariano GSL, Fink ME, Hoffman C, Rosengart AJ. Intracranial
pressure: monitoring and management. In: Hall J, Schmidt G, Wood L, editors.
features is associated with a high mortality 1
Principles of critical care. 4th ed. New York, NY: McGraw-Hill, 2012. B 2012,
from transtentorial herniation (80%), and with permission from The McGraw-Hill Companies, Inc.
hypertonic saline (250 mL of 3% saline) was
administered repeatedly until the serum sodium was 160 mEq/L. An estimated compliance curve is
illustrated in the graph below the images.
Comment. Clinical indications of intracranial hypertension may be inferred from imaging features,
even if intracranial pressure monitoring is not available. Hyperosmolar therapy was not successful
in this case because most of the edema was cytotoxic in the setting of brain infarction, so it would
not be expected to have a significant effect.

In addition to its use in the reduction
of elevated ICP, mannitol is used for
osmotic diuresis in some forms of renal
failure and for reduction of refractory
intraocular pressure in glaucoma. The
effects of mannitol on ICP begin within
minutes, peak between 15 and 120
minutes, and last from 1 to 5 hours,
depending on the dose of mannitol
administered. Mannitol concentrations
vary widely around the world, but the
most commonly used concentration is a
20% solution. Mannitol’s primary effect is
to increase the osmotic gradient across
the BBB, if it remains intact, and it pro-
motes osmosis of water from the brain
parenchyma, thereby decreasing brain
Continuum Lifelong Learning Neurol 2012;18(3):640–654
water content and increasing intravascu-
lar volume acutely. In brain tumors with
extensive vasogenic edema, mannitol will
reduce perilesional edema, a finding that
has been demonstrated in multiple ani-
mal studies and clinical trials. In addition,
mannitol has secondary mechanisms
that improve cardiovascular function
and blood viscosity. IV mannitol infusion
will transiently increase plasma volume,
decrease hematocrit and blood viscosity,
and decrease red blood cell deforma-
bility, thereby improving blood flow
through the microvasculature while in-
creasing cardiac output and increasing
mean arterial pressure.17Y21 This results
in enhanced tissue oxygen delivery, and
www.aan.com/continuum 645

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 Osmotherapy
KEY POINTS
h IV infusion of 20% autoregulation of cerebral blood flow
mannitol will promptly compensates with cerebral vasoconstric-
lower intracranial tion to reduce cerebral blood volume
pressure, and this and decrease ICP. These effects of man-
response is dose nitol have been studied primarily in
dependent. experimental animal models of trau-
h Continuous infusion of matic brain injury but also have been
mannitol does not documented in humans by a variety of
confer any benefits over methods, including PET scanning.22Y24
bolus administration Mannitol has been shown to be
and may be harmful. effective for reducing raised ICP (Class
II evidence) and is indicated for the
treatment of intracranial hypertension
when signs and symptoms suggest her-
niation or other complications of raised
ICP.4,25,26 Although not a requirement,
it is strongly recommended that some
form of ICP monitoring be used if
osmolar agents are used for treatment
of elevated ICP, and most centers will
institute such therapy when ICP rises
above 25 mm Hg. Some clinicians use
prophylactic mannitol in patients with
traumatic brain injury, but there is no
evidence to support this use and no
clinical trials indicating that it is benefi-
cial. Mannitol is more effective the
higher the initial ICP; when patients
are stratified by ICP at the onset of
therapy, patients who have higher ICP
exhibit a greater response to mannitol
treatment than those with marginally
elevated ICP. In addition, the response
to mannitol seems to be influenced by
the preceding doses; frequent repeated
doses and a high cumulative dose of
mannitol seem to result in less effec-
tiveness with the passage of time.27Y30
Therefore, the use of prophylactic man-
nitol may be counterproductive and may
make it less effective when patients be-
come symptomatic and need the ther-
apy more urgently.
The continuous infusion of mannitol
does not offer an advantage over bolus
use. It is prudent not to exceed a serum
osmolarity of 320 mOsm/L when using
mannitol because acute renal failure
may develop.31Y33 An osmotic gradient
646 www.aan.com/continuum
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
of 10 mOsm or more between blood
and brain is the most effective gradient
for lowering ICP, so rapid bolus IV
infusion of 0.5 g/kg to 1 g/kg of man-
nitol results in the most effective re-
sponse. There appears to be a significant
dose-response relationship when using
mannitol, and doses less than 0.5 g/kg
appear to be less efficacious and to
have a shorter duration of action. How-
ever, because the cumulative dose of
mannitol given seems to determine its
ongoing effectiveness, there are some
benefits to using the lowest effective
dose in a given patient. The lowest ef-
fective dose can be determined by re-
cording each bolus dose of mannitol,
monitoring its effects on ICP (using an
ICP monitor), and plotting its duration
of action. In addition to monitoring
serum osmolarity and avoiding osmo-
larity of greater than 320 mOsm/L, it is
useful to monitor the OG, which is
representative of the actual serum man-
nitol level and its clearance. Calculated
osmolality does not determine how
much mannitol is actually in the circu-
lation, but measured osmolarity includes
this molecule. Retrospective analyses of
acute renal failure cases that have
occurred during the use of mannitol
suggest that this complication is exceed-
ingly rare if the OG is less than 55.
Therefore, the OG appears to be a good
measurement to help monitor hyper-
osmolar therapy with mannitol when
treating intracranial hypertension.11
Other significant adverse effects with
the use of mannitol include electrolyte
abnormalities (eg, hypernatremia, hypo-
kalemia, and metabolic acidosis), hypo-
tension, and congestive heart failure
with pulmonary edema. Because manni-
tol is a potent osmotic diuretic, after the
acute expansion of intravascular volume
that occurs with the initial IV adminis-
tration there can then be a profound
diuresis and contraction of intravascular
volume that can lead to hypotension.
June 2012

Therefore, cardiovascular function and
intravascular volume as well as total IV
fluid input and urinary output must be
closely monitored. Hyponatremia may
occur during or immediately after infu-
sion and then may change to hyper-
natremia as osmotic diuresis occurs. The
initial hyponatremia is often called pseu-
dohyponatremia because it is a function
of mannitol altering the absolute sodium
concentration in the serum. There is not
a true deficit of sodium and it should not
be replaced.
There has been much discussion
about ICP rebound after prolonged use
of IV mannitol. The hypothesis states
that mannitol will leak into the brain tis-
sue and cause a rise in brain osmolality
due to a leaky BBB and a diminished
gradient between the intravascular and
extravascular spaces. As the mannitol is
cleared from the circulation, a reverse
gradient may be established, and water
may enter the brain and flow down the
osmotic gradient, which has changed to
favor water movement from the intra-
vascular space to the interstitial com-
partment. This phenomenon has been
demonstrated in animal models of brain
edema and has certainly been observed
clinically by most practitioners in the
neurologic intensive care unit. It is
clearly a function of the disruption of
the BBB that occurs in many of the
conditions being treated.27,28
To avoid a rebound rise of ICP with
the use of mannitol, it is recommended
that the smallest effective dose of bolus
therapy be used to allow a diuresis of the
mannitol bolus before repeating the
next dose. Sometimes this is possible,
but in cases of intractable intracranial
hypertension more frequent boluses of
mannitol are given, leading to a rise in
serum osmolarity and an increase in
mannitol concentration that cannot be
avoided. In such situations it may help to
slowly withdraw the mannitol and grad-
ually lower serum osmolarity during the
Continuum Lifelong Learning Neurol 2012;18(3):640–654
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
KEY POINTS
recovery phase of the acute condition h Adverse effects of
that is being treated. Of note, no Class I mannitol include acute
evidence supports concerns about a clin- renal failure,
ically significant rebound ICP elevation hypernatremia,
in patients receiving mannitol therapy. congestive heart failure,
However, this phenomenon has been and hypovolemia.
described in a variety of scholarly arti- h Intracranial pressure
cles, so it is best to be aware of the pos- rebound may occur
sibility of this complication.5,34 after repeated boluses
and prolonged
OSMOTIC THERAPY WITH administration of
HYPERTONIC SALINE mannitol.
In spite of many years of effective use of h Hypertonic saline and
mannitol for the treatment of raised ICP, mannitol have a very
the use of hypertonic saline has become similar effect on
more and more popular in the past elevated intracranial
decade. The increased popularity appears pressure when they are
to be due to concerns about complica- given in equimolar doses.
tions from mannitol use, particularly
acute renal failure and rebound of intra-
cranial hypertension when mannitol is
withdrawn. Although the rebound phe-
nomenon has been observed, it is not
clear whether it results in worse neu-
rologic outcome for patients. Neverthe-
less, it remains a significant concern for
clinicians.
Hypertonic saline comes in several
different concentrations. Normal saline
(isotonic NaCl) is 0.9% sodium chloride.
Hypertonic saline is available in 2%, 3%,
7.5%, and 23.4% concentrations. It is
easy to calculate equimolar quantities
of hypertonic saline; these quantities
can be directly compared to equimolar
amounts of mannitol and are used in
clinical trials that have attempted to
compare the effectiveness of mannitol
and hypertonic saline. Hypertonic saline
has similar osmotic effects as mannitol
but is a less potent diuretic, so the initial
advantage in certain clinical situations
is that hypertonic saline expands the
intravascular volume, increases blood
pressure, increases cardiac output, and,
theoretically, increases cerebral blood
flow.35Y37 The osmotic load from hyper-
tonic saline increases blood volume and
mean arterial pressure while at the
www.aan.com/continuum 647
 Osmotherapy
KEY POINT
h Hypertonic saline same time decreasing ICP. It is gener-
concentrations higher ally recommended that concentrations
than 2% should be of sodium chloride greater than 2% be
administered via a administered via a central venous cath-
central venous catheter eter because of the risk of phlebitis and
to prevent vein peripheral vein thrombosis, which can
thrombosis. be induced by high concentrations of
sodium chloride. However, no trial has
demonstrated that administering higher
concentrations of sodium chloride via a
central venous catheter diminishes the
risk of thromboembolic complications.
As with mannitol, hypertonic saline im-
proves the microcirculation, probably
by reducing the size of red blood cells
and enhancing their passage through
capillaries. Similar to mannitol, a con-
tinuous infusion of hypertonic saline
may cause a rebound exacerbation of
cerebral edema after the infusion is
stopped. The mechanism of action of
hypertonic saline is quite similar to
mannitol. When the BBB is intact, al-
most all of the hypertonic saline is
excluded from the interstitium because
the BBB prevents intravascular sodium
chloride from crossing membranes into
the interstitium as long as there are
tight junctions between endothelial
cells and intact astrocytic foot processes
on those capillaries.
One of the most common problems
associated with hypertonic saline is that
repeated doses or continuous infusion
of concentrated solutions leads to a
hyperchloremic acidosis, sometimes re-
quiring buffering with sodium bicarbon-
ate. It has been very difficult to evaluate
the efficacy of hypertonic saline or to
compare it to protocols using mannitol
because of the wide variety of con-
centrations available and the varied
protocols used. Some clinicians use a
continuous infusion, some use bolus
therapy, and both of these approaches
are done with varying concentrations of
saline. However, there is some agree-
ment that when transtentorial hernia-
tion of the brain is imminent, the most
648 www.aan.com/continuum
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
rapid response to high ICP is the use of
a bolus infusion of 30 mL to 60 mL of
23.4% saline. This protocol has been
shown to reverse the clinical features of
transtentorial herniation and reduce ICP
in most of the patients treated.26,38
Alternatively, a bolus of 250 mL to
500 mL of 3% sodium chloride can be
given, followed by continuous infusion
of 2% or 3% sodium chloride at 50 mL
to 100 mL per hour, with a goal of in-
creasing the serum sodium by approx-
imately 5 mEq/L in the first hour and
maintaining the serum sodium between
145 mEq/L and 155 mEq/L thereafter
(as demonstrated in Case 8-1). Central
pontine myelinolysis (CPM) is a possi-
ble complication if a rapid increase in
serum sodium occurs. CPM has never
been reported after the use of hyper-
tonic saline, but it may be best to avoid
using it in a patient with chronic hy-
ponatremia or other risk factors associ-
ated with CPM, including alcoholism,
malnutrition, and chronic hyponatremia
from the syndrome of inappropriate se-
cretion of antidiuretic hormone. Other
potential side effects of hypertonic sa-
line include renal failure (although this
appears to be much less common than
with the use of mannitol), dehydration,
cardiac arrhythmias, hemolysis, and
congestive heart failure with pulmonary
edema. In general, serum sodium should
be maintained below 160 mEq/L. How-
ever, in clinical situations of intractable
intracranial hypertension, sodium con-
centrations as high as 180 mEq/L have
been tolerated without complications,
and patients have made full recoveries
from extreme levels of hypernatremia
in spite of the fear of more complica-
tions. Patients who are treated with a
continuous infusion of hypertonic saline
usually develop hypokalemia because of
the exchange of sodium and potassium,
and they generally require an addition
of 20 mg to 40 mg of potassium
chloride per liter of infusion.39,40
June 2012

COMPARISONS OF MANNITOL
AND HYPERTONIC SALINE
As stated earlier, no Class I evidence
clearly demonstrates greater benefits
with mannitol versus hypertonic saline
in the treatment of cerebral edema
and elevated ICP, but a number of
small clinical trials have compared the
two regimens in a variety of clinical
situations.41Y43 A recent experimental
study by da Silva and colleagues44 in
Brazil compared mannitol and 10% hy-
pertonic saline using a rabbit model
and an intracranial balloon to elevate
ICP. There were three groups, includ-
ing a control that was compared to the
hypertonic saline and mannitol groups.
The control group had higher mortal-
ity, and hypertonic saline was signifi-
cantly better than mannitol at lowering
ICP. In 2011, Scalfani and colleagues22
at Washington University in St. Louis
studied the effects of mannitol and hy-
pertonic saline on cerebral blood flow.
They studied eight patients with severe
traumatic brain injury, using PET scan-
ning to measure regional cerebral blood
flow before and 1 hour after adminis-
tration of equimolar quantities of either
20% mannitol at 1 g/kg or 23.4% hypo-
tonic saline at 0.686 mL/kg. They found
that both agents were effective in low-
ering ICP and in increasing cerebral
perfusion pressure. They did not find a
significant difference between the two
agents, but it is not possible to make a
definitive statement on the basis of
such small numbers.
Kamel and colleagues14 at the Uni-
versity of California, San Francisco, did
a meta-analysis of all of the randomized
trials comparing mannitol and hyper-
tonic saline for the treatment of ele-
vated ICP as of 2011. They found five
well-designed trials that included 112
patients with 184 episodes of elevated
ICP. They noted that the odds ratio of
controlling intracranial hypertension
was 1.16 in favor of hypertonic saline
Continuum Lifelong Learning Neurol 2012;18(3):640–654
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
KEY POINT
and that hypertonic saline reduced ICP h There is no Class I
a mean of 2.0 mm Hg more than man- evidence to indicate that
nitol; both of these results were statis- hypertonic saline is
tically significant. Their conclusion was more effective than
that hypertonic saline may be more ef- mannitol in lowering
fective than mannitol for the treatment elevated intracranial
of elevated ICP, but this meta-analysis pressure.
was limited by the small number and
size of eligible trials. The authors clearly
state that to answer this question a large,
multicenter, randomized trial to defini-
tively establish first-line medical therapy
for intracranial hypertension should be
designed and implemented. Table 8-2
and Table 8-3 compare and contrast
the preparations, relative benefits, and
adverse effects of mannitol and hyper-
tonic saline.12
In 2009, the Neurocritical Care Soci-
ety surveyed its members to determine
practice patterns for the treatment of in-
tracranial hypertension, asking the clin-
icians about their preferred agent, dose,
and method for monitoring therapy.15 A
total of 295 responses were received,
80% of which were from physicians.
Most respondents (89.9%) reported
using osmotherapy as needed for intra-
cranial hypertension, with a small
minority reporting that they used it as
a prophylactic treatment. The practi-
tioners were fairly evenly divided
between those who preferred hy-
pertonic saline (54.9%) and those who
preferred mannitol (45.1%), with some
respondents reserving hypertonic saline
for patients with refractory intracranial
hypertension not responding to other
modalities, including mannitol. Those
who preferred hypertonic saline over
mannitol stated that they felt that hyper-
tonic saline had fewer systemic side
effects, had a more prolonged benefit,
caused less rebound cerebral edema,
was easier to titrate, and caused less
renal failure. Those practitioners who
preferred mannitol over hypertonic sal-
ine stated that they had more experi-
ence with the agent and were therefore
www.aan.com/continuum 649
 Osmotherapy

a
TABLE 8-2 Use of Mannitol and Hypertonic Saline

2% Hypertonic 3% Hypertonic 23.4% Hypertonic

Mannitol Saline Saline Saline
Dose 0.25 g/kg/dose to Initial infusion at Initial infusion at Refractory elevated
recommendations 1 g/kg/dose IV 1 mL/kg/h to 2 mL/kg/h, 1 mL/kg/h to 2 mL/kg/h, intracranial pressure
bolus over 1 min 250 mL bolus over 250 mL bolus over (ICP): IV (30 mL to
to 30 min 30 min may be 30 min may be 60 mL) given over
administered administered 2 min to 20 min
May be given Bolus can be repeated Bolus can be repeated Bolus can be repeated
repeatedly every after 30 min after 30 min after 15 min
4 h to 8 h
Recommended 2 g/kg/dose 1 mEq/kg/h = 1 mEq/kg/h = May be repeated in
maximum dose 2.9 mL/kg/h 1.9 mL/kg/h 6 h if target sodium
level not met
Route Peripheral IV or Peripheral IV or Central IV Central IV
central IV central IV
Osmolarity 1098 mOsm/L 684 mOsm/L 1027 mOsm/L 8008 mOsm/L
Onset and duration Onset: diuretic Onset: rapid Onset: rapid Onset: rapid
of action effect 1 h to 3 h
Reduction of ICP: Duration: 2 h to 6 h Duration: 2 h to 6 h Duration: 2 h to 6 h
15 min
Duration: diuretic
effect 4 h to 6 h
Reduction of ICP:
3 h to 8 h
a
Adapted from Mariano GSL, Fink ME, Hoffman C, Rosengart AJ. Intracranial pressure: monitoring and management. In: Hall J, Schmidt G, Wood L, editors.
Principles of critical care. 4th ed. New York, NY: McGraw-Hill, 2012.1 B 2012, with permission from The McGraw-Hill Companies, Inc.

more comfortable with it. They also felt
that it did not require central venous
access, which made it easier to use, and
a few clinicians felt that it was more ef-
fective. Of note, 95% of survey partici-
pants stated that they use mannitol in
their clinical practices, and it appears
that many of them use both mannitol
and hypertonic saline in different clinical
situations and in patients who require
both. Eighty-nine percent of the clini-
cians reported that they use hypertonic
saline as well, so it seems clear that both
agents enjoy the confidence of most
neurocritical care practitioners, although
the dosages and administrations are
extremely variable across the sites sur-
veyed. Table 8-3 lists differences that
the clinicians noted in their compar-
isons of mannitol and hypertonic saline.
The indications appear to be the same,
although the monitoring is different,
and it appears that the greatest con-
cerns related to the use of mannitol are
that it will reduce intravascular volume
and compromise cerebral perfusion
pressure, lead to nephrotoxicity, and
accumulate in the injured brain, thereby
causing rebound edema. Most of these
concerns are theoretical, however, and
have not been demonstrated to occur
on a frequent basis. Hays and colleagues
conclude that the use of mannitol and
hypertonic saline in neurocritical care

650 www.aan.com/continuum June 2012

Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

 a,b
TABLE 8-3 Comparison of Mannitol and Hypertonic Saline By Consensus Opinion
Mannitol
Indications Intracranial hypertension secondary to
cerebral edema
Dosing 0.25 g/kg to 1.5 g/kg boluses as needed or
at fixed intervals (eg, every 6 h)
Targets Intracranial pressure, radiographic midline
shift, clinical examination
Monitoring (1) Follow osmolar gap (goal G20 before
the next dose)
(2) Achieve hypernatremia (without systemic
hypovolemia)
(3) Blood urea nitrogen/creatinine
Advantages (1) Rapid effect in reducing intracranial
pressure (ICP) even before the onset of
osmotic diuresis
(2) No need for central venous access
Disadvantages (1) Could lead to reduction of intravascular
volume and compromise of mean arterial
pressure and cerebral perfusion pressure
(2) Impaired clearance can lead to
nephrotoxicity
(3) Accumulation into injured brain via
disrupted BBB and rebound edema
Maximum serum 330 mOsm/L
osmolality
Effectiveness May cause tolerance with repeated
administration
Continuum Lifelong Learning Neurol 2012;18(3):640–654
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
Hypertonic Saline
Intracranial hypertension secondary to
cerebral edema
Different concentrations available as
boluses or infusion (2%, 3%, 7.5%, and
23.4%)
Intracranial pressure, radiographic midline
shift, clinical examination
Prespecified sodium ranges (eg, 140 mEq/L
to 150 mEq/L or 150 mEq/L to 160 mEq/L).
Avoid exceeding 160 mEq/L.
(1) 30 mL bolus of 23.4% has an immediate
effect on ICP reduction
(2) High-throughput screening permeability
coefficient 1 vs. 0.9 for mannitol (intact
blood-brain barrier [BBB])
(3) Volume expander
(4) Positive inotropy
(5) Mixed immunomodulatory results
(1) Hypervolemia and pulmonary edema
(2) Hyperchloremic metabolic acidosis (consider
50:50 solutions with sodium acetate)
(3) Accumulation into injured brain via
disrupted BBB and rebound edema
(4) Hyperoncotic hemolysis
(5) Requires central venous access (peripheral
administration may lead to vein sclerosis)
(6) Bolus administration (especially 23.4%)
can lead to transient hypotension
(7) Mixed immunomodulatory results
60 mOsm/L
Effective after repeated administration and
when tolerance of mannitol has occurred
Beneficial as a rescue therapy to rapidly
induce hyperosmolality
Continued on next page
www.aan.com/continuum 651
 Osmotherapy
TABLE 8-3 Continued
Mannitol
Change in mean Moderate increase initially
arterial pressure
Diuretic effect Osmotic diuretic, may necessitate volume
replacement to avoid hypovolemia and
hypotension
Other suggested Antioxidant effects
interactions
Cautions Transient volume overload, pulmonary
edema, osmotic nephrosis, congestive
heart failure, hypertension, sodium
depletion, electrolyte abnormalities,
acidosis, increased cerebral blood flow,
and risk of postoperative bleeding
Additional One preparation
comments
Most reasonable price requires in-line filter
for administration
a
Adapted from Mariano GSL, Fink ME, Hoffman C, Rosengart AJ. Intracranial pressure: monitoring and management. In: Hall J, Schmidt G, Wood L, editors.
Principles of critical care. 4th ed. New York, NY: McGraw-Hill, 2012.1 B 2012, with permission from The McGraw-Hill Companies, Inc.
b
Adapted with permission from Hays AN, Lazaridis C, Neyens R, et al. Osmotherapy: use among neurointensivists. Neurocrit Care 2011;14(2):222Y228.15
www.springerlink.com/content/k365766155673r45/.
patients varies considerably and that
there is a lack of consensus as to which
agent is better.15 A large randomized
trial will be necessary to answer this
question, but currently it appears that
both agents are effective and have a
reasonable risk profile for the treatment
of cerebral edema and raised ICP.
REFERENCES
1. Mariano GSL, Fink ME, Hoffman C,
Rosengart AJ. Intracranial pressure:
monitoring and management. In: Hall J,
Schmidt G, Wood L, editors. Principles of
critical care. 4th ed. New York, NY:
McGraw-Hill, 2012.
652 www.aan.com/continuum
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
Hypertonic Saline
Greater, more prolonged
Diuresis through the stimulation of atrial
natriuretic peptide release
Restoration of resting membrane potential
and cell volume, inhibition of inflammation
(1) Hypotension (infusion related) phlebitis (less
than with higher sodium concentration)
(2) Thrombophlebitis tissue necrosis if
excavated hypotension (infusion related)
(3) Transient hypotension thrombophlebitis
tissue necrosis if extravasated
(4) Rebound ICP elevation, central pontine
myelinolysis, coagulopathy (bleeding),
electrolyte abnormalities (hypokalemia,
hyperchloremia, hypernatremia),
metabolic acidosis, congestive heart
failure, and pulmonary edema
Other preparations available:
5% saline (mOsm/kg = 1710)
7.5% saline (mOsm/kg = 2566)
14.6% saline (mOsm/kg = 5370)
2. Brain Trauma Foundation, American
Association of Neurological Surgeons,
Congress of Neurological Surgeons,
AANS/CNS Joint Section on Neurotrauma
and Critical Care. Guidelines for the
management of severe traumatic brain injury.
3rd ed. www.braintrauma.org/pdf/protected/
Guidelines_Management_2007w_
bookmarks.pdf. Accessed April 12, 2012.
3. Clausen T, Bullock R. Medical treatment and
neuroprotection in traumatic brain injury.
Curr Pharm Des 2001;7(15):1519.
4. Bratton SL, Chestnut RM, Ghajar J, et al.
Guidelines for the management of severe
traumatic brain injury. II. Hyperosmolar therapy.
J Neurotrauma 2007;24(suppl 1):S14YS20.
5. Bhardwaj A. Osmotherapy in neurocritical
care. Curr Neurol Neurosci Rep
2007;7(6):513Y521.
June 2012
 6. Broderick J, Connolly S, Feldmann E, et al.
Guidelines for the management of
spontaneous intracerebral hemorrhage in
adults: 2007 update: a guideline from the
American Heart Association/American
Stroke Association Stroke Council, High
Blood Pressure Research Council, and the
Quality of care and Outcomes in Research
Interdisciplinary Working Group. Stroke
2007;38(6):2001Y2023.
7. Marmarou A. A review of progress in
understanding the pathophysiology and
treatment of brain edema. Neurosurg Focus
2007;22(5):E1.
8. Chen CH, Toung TJ, Sapirstein A, Bhardwaj
A. Effect of duration of osmotherapy on
blood-brain barrier disruption and regional
cerebral edema after experimental stroke.
J Cereb Blood Flow Metab 2006;26(7):
951Y958.
9. Murphy N, Auzinger G, Bernel W, Wendon J.
The effect of hypertonic sodium chloride on
intracranial pressure in patients with acute
liver failure. Hepatology 2004;39(2):464Y470.
10. Schwarz S, Georgiadis D, Aschoff A,
Schwab S. Effects of hypertonic (10%) saline
in patients with raised intracranial pressure
after stroke. Stroke 2002;33(1):136Y140.
11. Garcia-Morales EF, Cariappa R, Parvin CA,
et al. Osmole gap in neurologic-neurosurgical
intensive care unit: its normal value,
calculation, and relationship with mannitol
serum concentrations. Crit Care Med 2004;
32(4):986Y991.
12. Mariano GSL, Fink ME, Hoffman C, et al.
Intracranial pressure: monitoring and
management. In: Hall J, Schmidt G, Wood L,
editors. Principles of critical care. 4th ed.
New York, NY: McGraw Hill, 2012.
13. Wakai A, Roberts I, Schierhout G. Mannitol
for acute traumatic brain injury. Cochrane
Database Syst Rev 2007;1:CD001049.
14. Kamel H, Navi BB, Nakagawa K, et al.
Hypertonic saline versus mannitol for the
treatment of elevated intracranial pressure:
a meta-analysis of randomized clinical
trials. Crit Care Med 2011;39(3):554Y559.
15. Hays AN, Lazaridis C, Neyens R, et al.
Osmotherapy: use among neurointensivists.
Neurocrit Care 2011;14(2):222Y228.
16. Mortazavi MM, Romeo AK, Deep A, et al.
Hypertonic saline for treating raised
intracranial pressure: literature review with
meta-analysis. J Neurosurg 2012;116(1):
210Y221.
17. Muizelaar JP, Wei EP, Kontos HA, Becker DP.
Mannitol causes compensatory cerebral
vasoconstriction and vasodilation in
Continuum Lifelong Learning Neurol 2012;18(3):640–654
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
response to blood viscosity changes.
J Neurosurg 1983;59(5):822Y828.
18. Burke AM, Quest DO, Chien S, Cerri C. The
effects of mannitol on blood viscosity.
J Neurosurg 1981;55(4):550Y553.
19. Mendelow AD, Teasdale GM, Russell T, et al.
Effect of mannitol on cerebral blood flow
and cerebral perfusion pressure in human
head injury. J Neurosurg 1985;63(1):43Y48.
20. Sorani MD, Morabito D, Rosenthal G, et al.
Characterizing the dose-response relationship
between mannitol and intracranial pressure
in traumatic brain injury patients using a
high-frequency physiological data collection
system. J Neurotrauma 2008;25(4):291Y298.
21. Sorani MD, Manley GT. Dose-response
relationship of mannitol and intracranial
pressure: a metaanalysis. J Neurosurg 2008;
108(1):80Y87.
22. Scalfani MT, Dhar R, Zazulia AR, et al. Effect
of osmotic agents on regional cerebral
blood flow in traumatic brain injury
[published ahead of print]. J Crit Care 2011;
doi:10.1016/j.jcrc.2011.10.008.
23. Mirski AM, Denchev ID, Schnitzer SM,
Hanley FD. Comparison between hypertonic
saline and mannitol in the reduction of
elevated intracranial pressure in a rodent
model of acute cerebral injury. J Neurosurg
Anesthesiol 2000;12(4):334Y344.
24. Qureshi AJ, Wilson DA, Traystman RJ.
Treatment of elevated intracranial pressure
in experimental intracerebral hemorrhage:
comparison between mannitol and
hypertonic saline. Neurosurgery 1999;44(5):
1055Y1063; discussion 1063Y1064.
25. Cruz J, Minoja G, Okuchi K. Major clinical
and physiological benefits of early high
doses of mannitol for intraparenchymal
temporal lobe hemorrhages with abnormal
papillary widening: a randomized trial.
Neurosurgery 2002;51(3):628Y637; discussion
637Y638.
26. Koenig MA, Bryan M, Lewin JL III, et al.
Reversal of transtentorial herniation with
hypertonic saline. Neurology 2008;70(13):
1023Y1029.
27. Kaufmann AM, Cardoso ER. Aggravation of
vasogenic cerebral edema by multiple-dose
mannitol. J Neurosurg 1992;77(4):584Y589.
28. McManus ML, Soriano SG. Rebound swelling
of astroglial cells exposed to hypertonic
mannitol. Anesthesiology 1998;88(6):
1586Y1591.
29. Sankar T, Assina R, Karis JP, et al.
Neurosurgical implications of mannitol
accumulation within a meningioma and its
www.aan.com/continuum 653
 Osmotherapy
peritumoral region demonstrated by
magnetic resonance spectroscopy: case
report. J Neurosurg 2008;108(5):1010Y1013.
30. Palma L, Bruni G, Fiaschi AI, Mariottini A.
Passage of mannitol into the brain around
gliomas: a potential cause of rebound
phenomenon. A study on 21 patients.
J Neurosurg Sci 2006;50(3):63Y66.
31. Dorman HR, Sondheimer JH,
Cadnapaphornchai P. Mannitol-induced
acute renal failure. Medicine (Baltimore)
1990;69(3):153Y159.
32. Visweswaran P, Massin EK, Dubose TD Jr.
Mannitol-induced renal failure. J Am Soc
Nephrol 1997;8(6):1028Y1033.
33. Weaver A, Sica DA. Mannitol-induced acute
renal failure. Nephron 1987;45(3):233Y235.
34. Prough DS, Whitley JM, Taylor CL, et al.
Rebound intracranial hypertension in dogs
after resuscitation with hypertonic solutions
from hemorrhagic shock accompanied by
an intracranial mass lesion. J Neurosurg
Anesthesiol 1999;11(2):102Y111.
35. Oddo M, Levine JM, Frangos S, et al. Effect
of mannitol and hypertonic saline on
cerebral oxygenation in patients with severe
traumatic brain injury and refractory
intracranial hypertension. J Neurol
Neurosurg Psychiatry 2009;80(8):916Y920.
36. Vialet R, Albanese J, Thomachot L, et al.
Isovolume hypertonic solutes (sodium
chloride or mannitol) in the treatment of
refractory posttraumatic intracranial
hypertension: 2 mL/kg 7.5% saline is more
effective than 2 mL/kg 20% mannitol. Crit
Care Med 2003;31(6):1683Y1687.
37. Suarez JI, Qureshi AI, Bhardwaj A, et al.
Treatment of refractory intracranial
654 www.aan.com/continuum
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.
hypertension with 23.4% saline. Crit Care
Med 1998;26(6):1118Y1122.
38. Qureshi AI, Suarez JI, Bhardwaj A, et al.
Use of hypertonic (3%) saline/acetate
infusion in the treatment of cerebral edema:
effect on intracranial pressure and lateral
displacement of the brain. Crit Care Med
1998;26(3):440Y446.
39. Aiyagari V, Deibert E, Diringer MN.
Hypernatremia in the neurologic intensive
care unit: how high is too high? J Crit Care
2006;21(2):163Y172.
40. Froelich M, Ni Q, Wess C, et al. Continuous
hypertonic saline therapy and the occurrence
of complications in neurocritically ill patients.
Crit Care Med 2009;37(4):1433Y1441.
41. Battison C, Andrews PJ, Graham C, Petty T.
Randomized, controlled trial on the effect
of a 20% mannitol solution and a 7.5%
saline/6% dextran solution on increased
intracranial pressure after brain injury.
Crit Care Med 2005;33(1):196Y202; discussion
257Y258.
42. Francony G, Fauvage B, Falcon D, et al.
Equimolar doses of mannitol and hypertonic
saline in the treatment of increased
intracranial pressure. Crit Care Med 2008;
36(3):795Y800.
43. Zeng HK, Wang QS, Deng YY, et al. A
comparative study on the efficacy of 10%
hypertonic saline and equal volume of 20%
mannitol in the treatment of experimentally
induced cerebral edema in adult rats.
BMC Neurosci 2010;11:153.
44. da Silva JC, de Lima FMT, Valenca MM, et al.
Hypertonic saline more efficacious than
mannitol in lethal intracranial hypertension
model. Neurol Res 2010;32(2):139Y143.
June 2012