Topical Review

Section Editor: Marc Fisher, MD

Antiedema Therapy in Ischemic Stroke

Juergen Bardutzky, MD; Stefan Schwab, MD

Abstract—Life-threatening, space-occupying brain edema occurs in up to 10% of patients with supratentorial infarcts and
is traditionally associated with a high mortality rate of up to 80%. Management of these patients is currently being
changed to an earlier and more aggressive treatment regimen. Early surgical decompression has recently been proven
effective to reduce mortality and increase the number of patients with a favorable outcome in randomized controlled
trials and is now the “antiedema” therapy of first choice for patients with large middle cerebral artery infarction aged
60 years or younger. Several medical treatment strategies have been proposed to control brain edema and reduce
intracranial pressure, including different osmotherapeutics, hyperventilation, tromethamine, hypothermia, and barbitu-
rate coma. None of these treatments is supported by level 1 evidence of efficacy in clinical trials, and some of them may
even be detrimental. Preliminary results on hypothermia for space-occupying hemispheric infarction are encouraging,
but far from definitive. (Stroke. 2007;38:3084-3094.)
Key Words: conventional antiedema therapy 䡲 edema, brain 䡲 hemicraniectomy 䡲 hypothermia 䡲 ischemia

S evere middle cerebral artery (MCA) or hemispheric

infarctions account for ⬇10% of all ischemic strokes.1,2
This type of stroke is commonly associated with variable
degrees of brain edema and a high mortality rate.2 The
clinical course in most of these patients typically follows a
Downloaded from http://ahajournals.org by on December 15, 2018
the efficacy of these conventional therapies to improve short-
or long-term clinical outcome.5 Thus, management of space-
occupying brain edema remained controversial and poorly
understood, and the value of these measures has come into
question.

predictable pattern. The patients initially show a severe
hemispheric syndrome with head turning, eye deviation,
hemiplegia, aphasia, or severe speech disturbance. Progres-
sive brain edema leads to neurological deterioration caused
by tissue shifts compressing the midline structures, eventually
leading to transtentorial and uncal herniation. Space-
occupying edema usually manifests between the second and
fourth day after stroke onset, although neurological deterio-
ration within 24 hours of symptom onset has been reported in
up to one-third of the patients.3 Outcome is fatal in the
majority of patients with space-occupying hemispheric in-
farction, with a mortality of ⬇80% in intensive care-based
series, despite maximal conservative treatment.1,4
Several treatment strategies have been proposed to reduce
brain edema and control elevated intracranial pressure (ICP)
for malignant MCA infarction including artificial ventilation
and hyperventilation, osmotherapy, tromethamine (THAM),
and barbiturate administration. There are a number of widely
accepted tenets about the effects and limitations on the use of
these therapies; however, only some of them appear to be
substantiated. No controlled randomized trial has addressed
Today, new therapeutic options can be offered. Early
surgical decompression has now been shown effective in
lowering mortality and improving neurological outcome in
randomized trials. Therapeutic hypothermia may represent
another option, which has been demonstrated to be neuropro-
tective in animal models, as well as in clinical studies after
cardiac arrest.6
This review examines the available data on the use of
conservative medical treatment modalities for antiedema
therapy in ischemic stroke and discusses its value with regard
to new therapeutic options that are decompression and
hypothermia.
Ischemic Brain Edema
It has become customary to distinguish 2 major types of
edema generation in ischemic stroke, the intracellular (cyto-
toxic) and the extracellular (vasogenic) edema, although this
classic vasogenic and cytotoxic paradigm is probably an
oversimplification. The cytotoxic edema develops within
minutes in the ischemic core and is mainly caused by energy
failure and anoxic membrane depolarization with subsequent

Continuing medical education (CME) credit is available for this article. Go to http://cme.ahajournals.org to take the quiz.
Received April 4, 2007; final revision received June 8, 2007; accepted June 18, 2007.
From Department of Neurology, University of Erlangen, Germany.
Correspondence to Juergen Bardutzky, Department of Neurology, University of Erlangen, Schwabachanlage 6, Germany 91054. E-mail
juergen.bardutzky@uk-erlangen.de
© 2007 American Heart Association, Inc.
Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.107.490193

3084
 Bardutzky and Schwab
accumulation of intracellular Na⫹, leading to an influx of
water and cellular swelling. The main factor for vasogenic
edema is a disruption of the blood– brain barrier (BBB),
leading to increased permeability and movement of proteins
and fluid from the intravascular space to the interstitial and
intracellular compartments. However, experimental data sug-
gest that the early increase in brain water content is mainly
caused by an abnormal Na⫹ transport into the extracellular
space rather than structural changes of the BBB with a gross
breakdown to larger molecules. The shift of ions and water
into the cells during the formation of cytotoxic edema creates
a new gradient for Na⫹ between the intravascular and the
extracellular space, which probably acts as driving force for
transcapillary electrolyte and water transport into the brain
parenchyma.7 Breakdown of the BBB with an almost indis-
criminate shift of intravascular proteins and ions into the
extravascular compartments occurs later in the course of
vasogenic edema. Although the exact mechanisms by which
ischemia disrupts the BBB are not fully understood, several
modulators and mediators have been identified in experimen-
tal investigations including aquaporins, free radicals, pro-
teases such as matrix metalloproteases, inflammatory cells
and their mediators, bradykinin, vascular endothelial growth
factor, and nitric oxide synthase.7,8
The effect of reperfusion on edema evolution is still a
matter of debate. In experimental focal ischemia, reperfusion
has been reported to both reduce infarct growth and brain
edema and aggravate vasogenic edema formation and hem-
orrhage. Whether reperfusion has beneficial or harmful ef-
Downloaded from http://ahajournals.org by on December 15, 2018
fects largely depends on severity and duration of previous
ischemia, and the efficacy of reperfusion.9 Interestingly, total
cessation of blood flow does not cause a change in brain
water and Na⫹ content, suggesting that brain swelling re-
quires active cerebral blood flow.7,8 Certainly, if reperfusion
is induced in already irreversibly damaged areas, the in-
creased vascular permeability during reperfusion leads to
vasogenic edema formation. In addition, late restoration of
blood flow may even increase ischemic damage by inducing
different pathophysiological mechanisms (eg, increase in
reactive oxygen species and excitatory amino acids, Ca2⫹
influx), with a further worsening of edema.8,9 However, the
clinical significance of these experimental data should be
interpreted with caution, because reperfusion was not consis-
tently associated with aggravation of edema, and in case of
human stroke, the benefits of early reperfusion by
thrombolytic therapy likely outweigh the potential worsening
of ischemic edema.
There is a considerable degree of variation in the timing
and extend of edema formation among patients. Different
clinical and radiological predictors for fatal brain edema
formation have been identified, such as early nausea/vomit-
ing, rapidly declining level of consciousness, dilated pupils,
NIHSS ⬎15, and early CT hypodensity involving ⬎50% of
the MCA territory and other vascular territories1,10,11. Unfor-
tunately, none of these predictors alone or in combination
proved sufficient prognostic value. It seems plausible that the
extent of swelling strongly depends on the extent and location
of the infarcted area, but substantial individual variability
exists.12
Antiedema Therapy 3085
Medical Antiedema Therapy
An overview of the clinical studies on medical treatment
strategies with proposed antiedema effects for acute ischemic
stroke is shown in Table 1.
Osmotherapy
Mannitol
Mannitol mainly acts as an osmotic agent, drawing water
from the interstitial and intracellular spaces of the brain
across the BBB. In addition, mannitol may improve micro-
vascular cerebral blood flow by hemodilution and increased
deformability of erythrocytes with a subsequent reduction of
cerebral blood volume and ICP via vasoconstriction. Manni-
tol may also cause an increase in cerebral perfusion pressure
(CPP) by augmentation of mean arterial blood pressure,
which in turn may cause cerebral vasoconstriction with
consequent reduction in cerebral blood volume and ICP.13
Numerous experimental studies have investigated the ef-
fects of mannitol on focal cerebral ischemia and edema
formation in various animal models, but the results have not
been consistent. Most investigations found beneficial effects
of intravenous mannitol with respect to infarct volume,14
edema formation,14,15 and elevated ICP.15 The effective dose
used varied markedly among the studies, ranging between a
single bolus of 1g/kg and repeated boluses of 2.5 g/kg every
4 hours.15 Other studies on focal ischemia failed to show a
significant positive effect of mannitol therapy on infarct size
or cerebral edema,16 and some studies even found detrimental
effects on brain swelling and midline shift after multiple
doses of mannitol.17
Given the fact that mannitol is one of the most frequently
used osmotic agent for the treatment of brain edema of
various types worldwide, surprisingly few clinical trials have
been performed to study the effects of mannitol in acute
ischemic stroke. A recent Cochrane report18 identified a total
of 5 randomized controlled trials, 4 of which were character-
ized as confounded, whereas the fifth19 had several method-
ical issues. The authors stated that no conclusion could be
drawn about the effect of mannitol in acute ischemic stroke
patients due to the lack of adequate controlled randomized
trials.
In a more recent observational study, Bereczki et al20
analyzed the case fatality with respect to mannitol treatment
in 805 stroke patients. The authors found that depending on
the prognostic factors used in the multivariate analysis,
mannitol had either a nonsignificant or an adverse effect on
30-day and 1-year case fatality. However, the results of this
nonrandomized study should be interpreted with caution. The
characteristics differed substantially between groups, the
decision to give mannitol was based on to the discretion of
the treating physician and CT was performed in only 73% of
the patients, with no lesion detected in 10%.
However, all of these clinical trials were not primarily
aimed at reducing edema formation in large ischemic stroke.
Randomized controlled studies on this subject are lacking,
although ICP-lowering properties of mannitol have been
reported in some case series on space-occupying hemispheric
infarction. Single doses of 40-g mannitol in 8 patients with
 3086 Stroke November 2007

Table 1. Summary of Clinical Studies Evaluating Medical Treatment Strategies With Proposed Antiedema Effect in Acute Ischemic Stroke
Study Type of Study No. Treated Interventions Treatment Results
Mannitol
Santambrogio 1978 Ran-con 36 0.8–0.9g/kg daily No effect on outcome and 10-day
survival
Schwarz 1998 Prospec 8 ⬇0.5g/kg Reduced ICP in 10/14 crises
Manno 1999 Prospec 7 1.5 mg/kg No effect on tissue shift
Videen 2001 Prospec 6 1.5 mg/kg Preferential shrinkage of normal
hemisphere
Bereczki 2003 Prospec 546 47⫾22 g/d for 6⫾3 d No or adverse effect on 1-yr
mortality

Hypertonic saline
Schwarz 1998 Prospec 8 100 mL 7.5%⫹60 g/L HES Reduced ICP in 16/16 crises
Schwarz 2002 Prospec 6 75 mL 10% Reduced ICP in 22/22 crises after
mannitol failed

Glycerol (only ran-con studies34)

Mathew 1972 Ran-con 34 500 mL 10% for 4–6 d No effect on mortality
Friethz 1975 Ran-con 50 500 mL 10% for 6 d No effect on mortality, neurological
improvement in intermediately
disabled patients
Larsson 1976 Ran-con 12 500 mL 10% for 6 d No effect on mortality and
outcome
Fawer 1978 Ran-con 26 2⫻250 mL 10% for 6 d No effect on mortality, transient
improvement of neurological
recovery improvement in
intermediately disabled patients
Friedli 1979 Ran-con 32 500 mL 10% for 6 d No effect on mortality, neurological
Downloaded from http://ahajournals.org by on December 15, 2018

Frei 1987 Ran-con 18 500 mL 10% for 7 d No effect on mortality and

outcome
Bayer 1987 Ran-con 85 500 mL 10% for 6 d Reduced mortality, no effect on
outcome
Yu 1993 Ran-con 56 500 mL 10% for 6 d No effect on mortality, trend
toward better Barthel Index
Sakamaki 2003 Prospec 6 300 mL 10% No effect on tissue shift

Hyperventilation
Christensen 1972 Ran-con 24 CO2 ⬇25 mm Hg for 72 hr No effect on outcome and
mortality
Simard 1973 Ran-con 50 CO2 ⬇22 mm Hg for 72 hr No effect on outcome

THAM
No study published

Barbiturate
Rockhoff 1979 Retrospec 4 Pentobarbital (dose?) for 5 d All patients died
Woodcock 1982 Prospec 5 Pentobarbital 50–300 mg/hr for 2–5 d 4 of 5 patients died
Schwab 1997 Prospec 60 Thiopental titrated for burst Reduced ICP in 50 of 60 patients
suppression after failure of conventional
therapy, significant CPP decline,
only 5 patients survived
Steiner 2001 Prospec 21 300–500 mg thiopental Reduced ICP in every case, but
reduced cerebral oxygenation and CPP
Furosemid
No study published
(Continued)
 Bardutzky and Schwab Antiedema Therapy 3087

Table 1. Continued
Study Type of Study No. Treated Interventions Treatment Results
Steroids (only ran-con studies53)
Patten 1972 Ran-con 17 Dexamethasone Improved 2-wk neurological
outcome
Bauer 1973 Ran-con 28 Dexamethasone No effect on outcome and 2-wk
mortality
Norris 1976 Ran-con 26 Dexamethasone Impaired 4-wk neurological
outcome
Mulley 1978 Ran-con 61 Dexamethasone No effect on outcome and 1-yr
mortality
Santambrogio 1978 Ran-con 48 Dexamethasone No effect on outome and 10-d
mortality
Gupta 1978 Ran-con 13 Betamethasone No effect on outcome and 3-wk
mortality
McQueen 1978 Ran-con 24 Betamethasone No effect on outcome and 12-wk
mortality
Norris 1986 Ran-con 54 Dexamethasone No effect on outcome and 3-wk
mortality

Indomethacine
Schwarz 1999 Case 1 50-g bolus Reduced ICP after failure of
conventional therapy
Prospec indicates prospective; ran-con, randomized controlled; retrospect, retrospective.

hemispheric stroke and massive edema were effective in
temporarily (up to 4 hours) reducing elevated ICP in 10 of 14
episodes.21 Effects on long-term outcome were not investi-
gated. Similarly, in another observational study with 21
Downloaded from http://ahajournals.org by on December 15, 2018
Summing up these results, surprisingly few clinical studies
exist testing mannitol in the setting of acute ischemic stroke.
In a systemic Cochrane review, outcome analysis could not
be performed because of lack of appropriate trials and a few

patients with severe MCA infarction, mannitol was effective
in reducing ICP in most patients. This was associated with
substantial increase in CPP and brain tissue oxygen pressure
in both the ischemic and the nonischemic hemisphere.22
Several concerns have to be considered when using man-
nitol in stroke patients. Theoretically, the effectiveness of
osmotherapy depends on an intact BBB and hypertonic
solutions may cause a preferential shrinkage of normal
hemisphere where BBB is still intact, with a subsequent
worsening of tissue shifts. The clinical significance of this
phenomenon remains unclear and the few available data in
humans are inconsistent.23,24 Moreover, accumulation of
mannitol in damaged brain tissue has been reported after
repeated doses in one animal study, causing a reversal of the
osmotic gradient and aggravation of brain edema.17 Again,
quantifying and demonstrating these differences between
compartments after multiple mannitol doses have been
proven difficult.15,25 It should be noted that the same pre-
sumed theoretical concerns apply to other osmotic agents
such as glycerol and hypertonic saline.
Given the lack of systematic clinical trial in stroke patients,
there is no clear information on optimal timing, dosing, and
application schedule of mannitol. A range between 310 and
320 mOsm/L has been commonly recommended as target
serum osmolality,22,26 and the maximum tolerable threshold is
generally believed to be 320 mOsm/L. However, this practice
is not supported by systemic experimental or clinical stud-
ies.13 Crossing this threshold is not necessarily dangerous as
long as the patient is not volume-depleted.
retrospective and observational studies found no or even a
tendency toward harmful effects in patients with acute stroke.
Similarly, no randomized clinical trial has addressed the
effect of mannitol treatment on clinical outcome in patients
with space-occupying edema after large MCA infarction.
Thus, the use of mannitol in these patients is solely founded
by the results of experimental studies, or observations in
small nonrandomized case series in humans.
Hypertonic Saline
Over the past few years, hypertonic saline solutions have
increasingly been used as an alternative to mannitol to control
brain edema of various types. As is the case with mannitol,
several mechanisms may be responsible for the reduction of
brain edema achieved with hypertonic saline. Because so-
dium chloride is completely excluded from an intact BBB, it
has been proposed that hypertonic saline may be a more
favorable osmotic agent compared with mannitol. Further-
more, hypertonic saline has the effect of expanding the
intravascular volume with increasing mean arterial blood
pressure leading to improved CPP, whereas mannitol is an
osmotic diuretic that secondary leads to volume depletion.
Other proposed mechanisms of action include modulation of
inflammatory response and neuron excitation, and improved
oxygenation.27
Experimental studies comparing hypertonic saline with
mannitol in different brain lesion models others than ischemia
have provided somewhat conflicting results with respect to
antiedema and ICP-lowering properties.28 In a few small
 3088 Stroke November 2007
randomized studies on patients with head injury, there is
some support that hypertonic saline may be more effective in
lowering ICP compared with mannitol.28,29 However, no
definitive conclusion can be drawn at present because the
experimental and clinical studies used a wide range of saline
concentrations and equiosmolar solutions were not consis-
tently used.
In stroke animal models, results have also been ambiguous;
7.5% saline worsened infarct volume when given as a bolus
immediately after reperfusion followed by continuous infu-
sion in rats.30 In contrast, using the same stroke model, the
authors found a significant reduction of water content in both
hemispheres when infusion was started 6 hours31 or 24 hours
after ischemia induction.32 Of note, these positive effects
were seen after prolonged continuous infusion (1 to 4 days),
with a subsequent serum osmolality clearly exceeding 350
mOsm/L. Lower saline concentration (3%), however, had no
effect on infarct volume30 and water content of ischemic brain
regions.31 Somewhat surprisingly, the effects of repeated
boluses instead of continuous infusion have not yet been
systematically investigated.
Data on the clinical use of hypertonic saline in ischemic
stroke are derived from only 2 small case series. In 9 patients
with elevated ICP caused by massive edema after hemi-
spheric stroke, a bolus of 100 mL of 7.5% saline plus
hydroxyethyl starch solution significantly lowered ICP in 16
of 16 episodes. Hypertonic saline solution appeared to de-
crease ICP more rapidly and effectively than equimolar doses
of mannitol, and was still successful after mannitol had
Downloaded from http://ahajournals.org by on December 15, 2018
failed.21 Similarly, in 8 severe stroke patients with 22 epi-
sodes of ICP crisis, a 75-mL bolus of 10% saline led to a
significant reduction of ICP in all episodes after failure of
other medical therapies including mannitol. This was associ-
ated with a significant increase in CPP, whereas mean arterial
blood pressure remained essentially unchanged.33
At present, there is no clear information on the optimum
form of application of hypertonic saline. Multiple concentra-
tions ranging from 3% to 23.4% have been tested in contin-
uous and bolus infusion with different application schedules
for the treatment of traumatic brain injury, and the results are
controversial.34 For the use of hypertonic saline in ischemic
stroke, almost no evidence exists for any one formula because
of the lack of systemic trials.
A frequently emphasized argument against the nonrestric-
tive use of hypertonic saline is the fear of complications
attributable to severe hypernatremia and to inducing pontine
myelinolysis, although these concerns are more hypothetical
and such complications have been rarely observed in clinical
studies.27 Other potential side effects of hypertonic saline
include congestive heart failure and pulmonary edema, hy-
perchloremic acidosis, hypokalemia, and hypomagnesemia.
In summary, although data on the use of hypertonic saline
for the treatment of brain edema and elevated ICP are
promising in various conditions, it is far away from defini-
tive. Only 2 small case series investigated hypertonic saline
solutions in the setting of brain edema after large MCA
infarction. No clinical trial addressed its effect in the acute
stage of ischemia or its impact on functional outcome. Thus,
whereas hypertonic saline may ameliorate elevated ICP over
a short period of observation, its long-term effects in patients
with ischemic stroke are largely unknown.
Glycerol
The sugar glycerol is another osmotic agent that may also has
neuroprotective properties. In human stroke, increase of
blood flow to ischemic territories and improvement in ische-
mic brain energy metabolism after glycerol administration
have also been postulated.35 The occurrence of a rebound
phenomenon has been controversial,12 but glycerol has the
theoretical advantage over other osmotics in being metabo-
lized by the brain on crossing BBB, thus reducing the risk of
rebound edema. Using MRI before and after glycerol admin-
istration in 6 patients with large hemispheric infarction, no
effect on tissue shift and noninfarcted hemisphere volume has
been observed.36 In addition, glycerol has almost no major
side effects.37
Glycerol has been evaluated in numerous randomized and
nonrandomized clinical trials of acute stroke, either ischemic
or hemorrhagic. A systemic (Cochrane) review identified 10
randomized studies in which a total of 482 glycerol-treated
patients were compared with 463 control patients.37 Glycerol
was associated with a nonsignificant reduction in odds of
death within the scheduled treatment period (OR, 0.78; CI,
0.58 to 1.06), which was significant for definite or probable
ischemic stroke (OR, 0.65; CI, 0.44 to 0.97). However, there
was no evidence of any effect of glycerol on mortality at the
end of scheduled follow-up. Functional outcome was compa-
rable only in 2 trials, and there was a nonsignificant reduction
in odds of being death or dependent (OR, 0.73; CI, 0.37 to
1.42). However, these results should be interpreted with
caution because most trials were performed in the pre-CT era,
and only a few patients had their stroke confirmed by brain
imaging. Furthermore, in only 1 study the clinical suspicion
of brain edema was considered in the inclusion criteria.
Therefore, the review does not provide reliable evidence on
the efficacy of glycerol in patients with established brain
edema.
Summarizing these results, glycerol is one of the most
frequently tested treatments for acute human stroke. Glycerol
therapy seems to have a beneficial effect on short-term
survival in patients with acute ischemic stoke, but no long-
term efficacy. The lack of proven benefit on long-term
clinical outcome does not support its routine use in patients
with acute ischemic stroke. No randomized trial specifically
addressed its effect on outcome in patients with massive
edema secondary to large hemispheric infarction.
Barbiturates
The main effect of barbiturates consists of a decrease in
cerebral metabolism. Reduced metabolic rate and subsequent
reduction of cerebral blood volume and cerebral blood flow
may theoretically reduce edema formation and lower ICP.
Case series suggest that barbiturate therapy may be effective
to reduce elevated ICP in various conditions, although ICP-
lowering effect appears to be relatively inconsistent.38 – 40 A
prospective comparative study with 95 severely head-injured
patients reported that barbiturate coma was less effective than
mannitol for control of raised ICP, and may be even harmful
 Bardutzky and Schwab
in the subgroup of patients without hematoma.41 The use of
high-dose barbiturates is frequently associated with severe
complications, including hypotension, hepatic dysfunction,
and increased risk of infections.
There are only very limited data on the use of barbiturates
for brain edema after severe infarction. Some case studies in
the 1970s and 1980s on ICP-lowering effects of barbiturates
included a few patients with brain swelling secondary to
ischemic infarction, and the observations were disappoint-
ing.38,39 In a more recent case series with 21 patients with
elevated ICP after large MCA infarction, barbiturate treat-
ment was persistently associated with a reduction in cerebral
oxygen pressure and reduced CPP, although ICP was tempo-
rarily decreased in every case.22 In the only prospective, but
uncontrolled study on this subject, 60 consecutive patients
with elevated ICP caused by severe MCA infarction were
treated with barbiturates, after failure of osmotherapy and
hyperventilation. Only 5 patients survived. Although barbi-
turates were initially effective in lowering ICP in 50 patients,
sustained ICP control was only achieved in the 5 survivors.
Moreover, a significant decline in CPP was observed during
barbiturate infusion.42 Randomized controlled trials are lack-
ing.
Thus, in patients with large MCA infarctions, barbiturates
only seem to offer limited and short-lasting benefits that may
be counterbalanced by severe adverse effects, especially if
hypotension occurs with a subsequent critical decline in CPP.
Hyperventilation
Downloaded from http://ahajournals.org by on December 15, 2018
Hyperventilation decreases ICP by the induction of cerebral
vasoconstriction, with a subsequent decrease of cerebral
blood volume and cerebral blood flow. The ICP-lowering
capacity of hyperventilation has been shown in numerous
case series in traumatic brain-injured patients, with duration
of hyperventilation mostly ranging between 10 and 30 min-
utes.43 However, some studies suggest that the effect of
hyperventilation may diminish within a few hours.43,44 De-
spite the wide use of hyperventilation in the treatment of
elevated ICP after head trauma, its effect on clinical outcome
was evaluated in only 1 randomized clinical trial. Depending
on subgroup analysis, prolonged hyperventilation had no or
even harmful effect on functional outcome at 3 and 6
months.44
The major drawback of hyperventilation is that cerebral
vasoconstriction may decrease cerebral blood flow to ische-
mic levels.45 In addition, various clinical studies on head
injury suggested deleterious effects of hyperventilation on
cerebral oxygenation46 and cerebral metabolism.45 Moreover,
rebound vasodilation may occur along with the risk of
increasing ICP after discontinuation of hyperventilation.44
Data on the use of hyperventilation in stroke patients are
rare and were derived from clinical studies in the early 1970s.
In these trials, no effect of hyperventilation (CO2 22 to
25 mm Hg) on clinical outcome and mortality was ob-
served.47,48 More recent clinical studies on this subject are
lacking.
With regard to the large body of evidence indicating the
possible deleterious effects of hyperventilation on cerebral
oxygenation, metabolism and blood flow, and the lack of
Antiedema Therapy 3089
evidence of any beneficial effect on outcome, hyperventila-
tion cannot be recommended in stroke patients.
THAM
THAM is supposed to act by entering the cerebrospinal
fluid compartment and neutralizing the acidosis-induced
vasodilation, thereby reducing ICP. ICP-lowering proper-
ties as well as beneficial effects on edema formation and
cerebral energy disturbance of THAM have been demon-
strated in animal models of head injury.49 In animal models
of focal cerebral ischemia, THAM infusion was associated
with a significant reduction of infarct size, brain edema
and lactate concentration.50
There are only a few data published on the clinical use of
THAM treatment for brain edema and elevated ICP. In the
only prospective randomized trial in 149 patients with severe
head injury assigned to receive THAM or placebo for 5 days,
THAM treatment was associated with significantly fewer
episode of ICP elevation and a significantly lower incidence
of requiring barbiturate coma. However, no difference in
outcome was observed.51
There are no controlled studies on the use of THAM in
acute stroke patients. Nevertheless, some authors consider
THAM administration as an option for treating brain edema
and elevated ICP secondary to large MCA infarction.2,12,26
Elevated Head Position
Elevation of the head is thought to decrease ICP via increas-
ing the venous outflow and reducing venous hydrostatic
pressure and volume at the cranial level. Thus, patients with
massive hemispheric stroke are traditionally nursed with the
head moderately (15° to 45°) elevated. These recommenda-
tions for stroke patients are largely derived from pathophys-
iological considerations and the results from head-injured
patients. In addition, elevated backrest position may reduce
the risk of ventilator-associated pneumonia.52 Head elevation,
however, may also cause marked decrease in CPP because of
decrease in mean arterial blood pressure,53 and sustained CPP
declines can result in further ischemic damage. Despite this
concern, only 1 study systematically investigated the effects
of body position in 12 patients with large supratentorial
stroke.54 Mean ICP decreased only slightly from 13 mm Hg
in horizontal position to 11.4 mm Hg at 30° backrest eleva-
tion, whereas CPP markedly declined from 77 mm Hg to
64 mm Hg. These findings and our own experience give little
support for the routine use of 30° head elevation, because the
increase in ICP in the horizontal position is likely not
clinically significant, but the decrease in CPP may be harmful
for patients with massive stroke. Instead, optimal body
position should be established individually: the flat position
may probably be preferable to achieve sufficient CPP; mod-
erate head elevation appears reasonable as long as CPP is
adequately maintained (⬎70 mm Hg) and in those patients in
whom ICP increases substantially in a horizontal position.
Indomethacin
Indomethacin is a potent cerebral vasoconstrictor that also
may exhibit antiedema and anti-inflammatory effects. ICP-
lowering capacity of indomethacin has been described in few
 3090 Stroke November 2007
Table 2. Summary of Clinical Studies on Hypothermia (33°C) for Antiedema Therapy After Severe (More Than Two-Thirds) MCA
Infarction
Study No. Patients Induction Latency (h)
Schwab 1998 25 14⫾7
Schwab 2001 50 22⫾9
Georgiadis 2001 6 28⫾17
Georgiadis 2002 19 24 (18–24)
Milhaud 2005 12 10⫾7
BI indicates Barthel Index.
case reports on head trauma.55 In a patient with space-
occupying hemispheric infarction, indomethacin was effec-
tive in transiently reducing ICP and improving CPP after all
other conventional therapies had failed, and repeated boluses
continued to be effective.56 However, a critical decline in
cerebral blood flow with secondary ischemia may occur as a
result of cerebral vasoconstriction.55 Given the lack of sys-
temic clinical trials, indomethacin for the treatment of post-
ischemic swelling remains experimental.
Steroids
A systematic (Cochrane) review57 identified only 7 of 22
published trials of acute steroid treatment in acute presumed
ischemic stroke acceptable for further analysis, and these
comprised only 453 patients in total with no uniformity of
evaluation and assessment. Only 1 study performed CT to
exclude hemorrhagic stroke.58 The authors concluded that
these trials of steroids do not provide any evidence of a
Downloaded from http://ahajournals.org by on December 15, 2018
beneficial effect on mortality or functional outcome after
presumed ischemic stroke. Clinical studies on intracerebral
hemorrhage also failed to show beneficial effects, whereas
the rate of complications significantly increased.59 Obviously,
the possibility that inclusion of hemorrhagic infarcts may
have influenced the results of early studies cannot be ruled
out. Furthermore, 2 trials with high case fatality rates in the
control group reported nonsignificant trends in favor of
steroid treatment.58,60 Thus, given the likely mode of action of
steroids, it appears possible, that patients with vasogenic
edema secondary to large infarction may potentially benefit
from steroid treatment. However, there is no trial that
specifically addressed this issue. In addition, steroids are
known to increase the risk of infections, hyperglycemia, and
muscle catabolism. These adverse effects and the lack of
evidence of any benefit in randomized stroke trials certainly
discourage the conduction of further investigations.
Furosemide
Loop diuretics like furosemide may act by decreasing total
body water and increasing blood osmolality, thereby remov-
ing water from the brain. Only a few experimental studies
addressed the use of furosemide for treating brain edema after
focal cerebral ischemia, and results are inconsistent.25,32 No
clinical trials have been published evaluating the effects of
furosemide on outcome after ischemic stroke. As a note of
caution, the risk of substantial volume depletion with subse-
quent decrease in mean arterial blood pressure and CPP to
ischemic levels may outweigh any potential benefit on brain
edema and ICP.
Hypothermia Duration Mortality, % Outcome of Survivors
65 (48–72) hr 44 BI 70 (60–85), mRS 2.6 (2–4)
55 (24–72) hr 38 BI 65 (10–85), mRS 2.9 (2–5)
48–78 hr 17 Not given
71⫾21 hr 42 Not given
19⫾4 d 42 BI 75 (50–95), mRS 3 (1–5)
New Therapeutic Approaches
Hypothermia
Most likely, hypothermia exerts multiple and synergistic
neuroprotective effects. Hypothermia reduces the cerebral
metabolic rate, stabilizes BBB, reduces brain edema, free
radical formation, and the release of excitatory neurotrans-
mitters, and attenuates postischemic inflammatory response
and apoptosis.61
In animal models of transient focal ischemia, hypothermia
consistently decreased infarct volume when initiated before
or at the onset of ischemia,62 and to a less extend with
increasing delay of cooling.63 A longer cooling period results
in better neuroprotective effects when initiation of cooling is
delayed.62,63 Another crucial factor for its neuroprotective
potential is the depth of hypothermia. A recent study system-
atically evaluating temperatures between 32°C and 37°C
found the smallest infarct volumes with a body temperature
of 33°C and 34°C.64
Although early intervention to protect the ischemic pen-
umbra until restoration of blood flow may be seen as the
ultimate goal of therapeutic hypothermia for ischemic stroke,
most of the initial human studies were designed to address the
effects of hypothermia on space-occupying edema after
infarction was completed.
In the 5 observational case series on this subject published
to date, a total of 112 patients has been investigated, all of
them with space-occupying infarction involving at least
two-thirds of the MCA territory65– 69 (Table 2). All trials used
33°C as target temperature and treatment entailed deep
sedation, relaxation, and mechanical ventilation. The delay
from stroke onset to induction of hypothermia was relatively
long in all trials, with a mean ranging between 10 and 28
hours. Duration of cooling also varied, ranging between 55
and 71 hours in the first 4 studies. In the study by Milhaud et
al67 testing more prolonged hypothermia, mean cooling pe-
riod was 19 days. A significant and rapid decrease in ICP was
described in the first 2 studies by Schwab et al.68,69 Mortality
was relatively consistent among trials, being 44% and 38% in
the studies reported by Schwab et al,68,69 47% in the study
described by Georgiadis et al,65 and 42% in the study by
Milhaud et al.67 The rewarming period seems to be crucial,
because rebound intracranial hypertension during rewarming
was a common occurrence and a major contributor to mor-
tality,65,68,69 and a shorter rewarming period (⬍16hours) was
associated with a more pronounced increase in ICP.68 Though
all patients fulfilled the criteria for malignant MCA infarc-
tion, the survivors reached a fairly favorable outcome at 3
 Bardutzky and Schwab
months (Barthel Index 65 and 70, respectively)68,69 or at 6
months (Barthel Index 75).67 The most common complica-
tions were arterial hypotension (40% to 100%), thrombopenia
(37% to 76%), pneumonia (11% to 48%), and bradycardia
(30% to 60%).61 Randomized controlled clinical trials on
hypothermia for the treatment of space-occupying MCA
infarction are still lacking.
Other clinical trials focused on the feasibility of the early
(⬍6 to 9 hours) use of therapeutic hypothermia. Two recent
studies reported early cooling to be safe and feasible in awake
stroke patients, even when combined with thrombolysis.70,71
Overall, no significant differences were observed between
hypothermic and normothermic patients, but studies were not
powered to evaluate the efficacy of early cooling.
In summary, although numerous experimental studies con-
sistently demonstrated robust neuroprotection when cooling
was initiated during ischemia evolution, its effect on space-
occupying edema after established large infarction has barely
been tested in animal studies. Five observational case series
specifically addressed the effect of hypothermia on massive
brain edema after severe MCA infarction. Hypothermia
seems to reduce mortality, with a good outcome of survivors
when compared with historical controls with malignant MCA
infarction.1,4 Although these results are encouraging, they are
preliminary and far from firm. The interpretation is strongly
limited by confounding bias of such comparisons, the highly
selective nature of nonrandomized trials, and the low number
of patients included. Evidence from randomized trials on
cooling for the treatment of space-occupying infarction is
Downloaded from http://ahajournals.org by on December 15, 2018
lacking. Thus, the use of hypothermia in these patients
remains experimental and no evidence-based recommenda-
tion can currently be given.
Future Directions
Further investigations should define the optimal methodical
approach for inducing hypothermia in stroke including pa-
tient selection, onset timing, depth and duration of hypother-
mia, and rewarming. One oft the most pressing issues is the
optimal timing of hypothermia. All studies on hypothermia
for antiedema therapy awaited clinical worsening before
deciding for hypothermia treatment. The long delay in cool-
ing induction surely precluded any significant neuroprotec-
tive effect.
Optimal duration of cooling represents another important
issue that also should be addressed. Most of studies for
massive edema applied cooling for 48 to 72 hours. However,
given the natural time course of edema evolution in which
swelling is usually maximal on days 2 to 5,12 cerebral edema
is still present even after 72 hours, and the patient’s condition
frequently worsens during rewarming. Thus, from a patho-
physiological point of view, it seems reasonable that hypo-
thermia might be more effective in controlling ICP when
cooling period exceeds the natural peak of brain swelling, ie,
⬎120 hours. Prolonged (8 days) hypothermia has been shown
to be safe and feasible in a trial on severe head injury72 and,
more recently, in 1 study on space-occupying MCA infarc-
tion,67 with a complication rate comparable to previous
studies using shorter cooling times and a fairly good func-
tional outcome of survivors.
Antiedema Therapy 3091
Thus, with regard to neuroprotection and the natural course
of edema evolution, early induced (⬍6 hours) and more
prolonged (⬎120 hours) hypothermia may achieve increased
benefit in patients with evolving space-occupying hemi-
spheric infarction.
Decompressive Surgery
Over the past decades, decompressive hemicraniectomy for
malignant MCA infarction has been reported to lower mor-
tality without increasing the number of severely disabled
survivors in several uncontrolled case series.12,26,73 However,
given the highly selected nature of such uncontrolled studies,
and the lack of prospective randomized trials, there was as yet
substantial concern about allowing survival at the cost of a
high degree of disability with this aggressive therapy.
Fortunately, the results of a pooled analysis of 3 European
randomized controlled trials (DECIMAL, DESTINY, HAM-
LET) assessing the effect of decompressive surgery in pa-
tients with space-occupying MCA infarction are now avail-
able.74 Inclusion criteria for the pooled analysis were age 18
to 60 years, NIHSS ⬎15, decreased level of consciousness,
infarct volume ⬎50% of the MCA territory on CT or ⬎145
cm3 on diffusion-weighted MRI, and surgery initiated within
48 hours after stroke onset. A total of 93 patients were
included. Survival rate was significantly higher in patients
treated with decompressive surgery compared with patients
without surgery (78% vs 29%; Figure 1). Significantly more
patients in the surgery group had a favorable outcome with a
modified Ranking scale (mRS) score ⱕ4 (75% vs 24%) and
mRS ⱕ3 (42% vs 21%; Figure 2). Number needed to treat
was 2 for survival with mRS ⱕ4, 4 for survival with mRS
ⱕ3, and 2 for survival irrespective of functional outcome.
Further analysis suggested that age older than 50 years, time
window to treatment ⬎24 hours, and the presence of aphasia
did not alter the beneficial effects of surgery, although
subgroups were small and not powered to detect small
differences in treatment effects.
It must be noted that this study design is an unconventional
approach in which a pooled analysis from 3 independent trials
was planned in advance while these trials were still ongoing.
The obvious advantage of such an approach is to keep the
number of patients included to a minimum and to report the
results much earlier than would have been possible based on
individual trials alone. However, this study design certainly
has limitations that are strongly related to the nature of pooled
analysis of independent trials. Eligibility criteria such as age,
time to randomization, and neuroimaging criteria were not
identical, resulting in differences in baseline characteristics,
infarct morphology, and in timing of surgery between indi-
vidual trials. Nevertheless, there was no significant heteroge-
neity between the three trials with regard to all outcome
measures and adjusted analysis did not make any differences
to the results. More importantly, 2 of the 3 trials (DESTINY
and DECIMAL) stopped recruitment in 2006 and the results
of both trials are consistent to those of the pooled analy-
sis,75,76 confirming the significance of the pooled analysis.
The results will be published soon in Stroke.
Thus, based on these findings, early (⬍48 hours) decom-
pressive surgery can now be recommended as the treatment
 3092 Stroke November 2007

Survival Rate
100

90
14 / 17
11 / 14 40 / 51
80 15 / 20

70

60
Conservative
% 50 7 / 15
Surgery
40
12 / 42
30
4 / 18
20
1/9
10

0
DECIMAL DESTINY HAMLET Pooled Data
Figure 1. Survival rate at 12 months for the 3 individual randomized controlled trials (DECIMAL, DESTINY, HAMLET) and the pooled
data on decompressive surgery for malignant MCA infarction.

of choice for patients aged 60 years or younger, with severe
infarction of at least 50% of the MCA territory, to reduce
mortality and increase the number of patients with favorable
functional outcome, independent of laterality. However, some
aspects must be considered when offering this treatment.
Downloaded from http://ahajournals.org by on December 15, 2018
Second, several nonrandomized trials suggest that decom-
pressive surgery is less effective in elderly patients.73 The 3
randomized trials excluded patients older than 55 or 60 years,
and the results of the pooled data probably cannot be
extrapolated to patients who are older. Thus, to date, it

First, although surgery doubles the probability to survive in a
favorable condition (mRS ⱕ3), and the risk of severe disability
(mRS 5) is not increased, the chance of surviving in a condition
requiring assistance from others (mRS 4) increases ⬎10 times.
Thus, the decision to perform decompressive surgery in patients
with space-occupying infarction should always be based on the
wishes of the patient and their families in light of the potential to
survive with long-term moderate disability.
remains unclear whether decompressive surgery is also ben-
eficial in patients older than 60 years. A randomized con-
trolled trial addressing this subject is underway.
Third, whether decompressive hemicraniectomy is still
beneficial if performed after the first 48 hours remains also
unclear and is currently being tested in HAMLET.77
Fourth, several complications of surgical decompression
have been reported, including wound infection, subdural

Figure 2. Distribution of the scores of modified Ranking scale (MRS) after 12 months for patients treated with or without decompres-
sive surgery. Reprinted from Vahedi K, et al. Early decompressive surgery in malignant infarction of the middle cerebral artery: a pooled
analysis of three randomised controlled trials. Lancet Neurol. 2007;6:215–222, with kind permission from Elsevier.
 Bardutzky and Schwab
hygroma and hematoma, tissue shearing from inadequately
large craniotomy, brain sagging, and hydrocephalus that
should not be shunted when skull remains open.
Conclusions
Several medical treatment strategies have been proposed to
be effective in controlling cerebral edema and reducing
elevated ICP after severe hemispheric stroke. However,
conservative antiedema therapy, as exists today, is disap-
pointing. None of these treatments is supported by adequate
evidence of efficacy from experimental studies or clinical
trials, and some interventions may even be detrimental.42,44
Thus, at present, no evidence-based recommendation on the
use of any of these conventional strategies for space-
occupying postischemic edema can be made. Earlier and
more aggressive therapeutic approaches are strongly needed
for these patients. Immediate access to reperfusion therapy
remains the cornerstone of stroke therapy and also the first
step of antiedema treatment, because limiting final infarct
size by restoration of blood flow may likely reduce the chance
of developing massive postischemic edema. In cases of
evolving large MCA infarction, early surgical decompression
is now the “antiedema” therapy of first choice for patients
younger than 60 years independent of affected hemisphere.
Whether this aggressive treatment is beneficial for elderly
patients or when performed after 48 hours is currently being
tested.
Disclosures
Downloaded from http://ahajournals.org by on December 15, 2018
None.
References
1. Hacke W, Schwab S, Horn M, Spranger M, De Georgia M, von Kummer
R. “Malignant” middle cerebral artery territory infarction: Clinical course
and prognostic signs. Arch Neurol. 1996;53:309 –315.
2. Aiyagari V, Diringer MN. Management of large hemispheric strokes in
the neurological intensive care unit. Neurologist. 2002;8:152–162.
3. Qureshi AI, Suarez JI, Yahia AM, Mohammad Y, Uzun G, Suri MF,
Zaidat OO, Ayata C, Ali Z, Wityk RJ. Timing of neurologic deterioration
in massive middle cerebral artery infarction: A multicenter review. Crit
Care Med. 2003;31:272–277.
4. Berrouschot J, Sterker M, Bettin S, Koster J, Schneider D. Mortality of
space-occupying (‘malignant’) middle cerebral artery infarction under
conservative intensive care. Intensive Care Med. 1998;24:620 – 623.
5. Hofmeijer J, van der Worp HB, Kappelle LJ. Treatment of space-
occupying cerebral infarction. Crit Care Med. 2003;31:617– 625.
6. Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypo-
thermia to improve the neurologic outcome after cardiac arrest. N Engl
J Med. 2002;346:549 –556.
7. Simard JM, Kent TA, Chen M, Tarasov KV, Gerzanich V. Brain oedema
in focal ischaemia: Molecular pathophysiology and theoretical impli-
cations. Lancet Neurol. 2007;6:258 –268.
8. Ayata C, Ropper AH. Ischaemic brain oedema. J Clin Neurosci. 2002;9:
113–124.
9. Schaller B, Graf R. Cerebral ischemia and reperfusion: The pathophysi-
ologic concept as a basis for clinical therapy. J Cereb Blood Flow Metab.
2004;24:351–371.
10. Krieger DW, Demchuk AM, Kasner SE, Jauss M, Hantson L. Early
clinical and radiological predictors of fatal brain swelling in ischemic
stroke. Stroke. 1999;30:287–292.
11. Kasner SE, Demchuk AM, Berrouschot J, Schmutzhard E, Harms L,
Verro P, Chalela JA, Abbur R, McGrade H, Christou I, Krieger DW.
Predictors of fatal brain edema in massive hemispheric ischemic stroke.
Stroke. 2001;32:2117–2123.
12. Steiner T, Ringleb P, Hacke W. Treatment options for large hemispheric
stroke. Neurology. 2001;57:S61–S68.
Antiedema Therapy 3093
13. Diringer MN, Zazulia AR. Osmotic therapy: Fact and fiction. Neurocrit
Care. 2004;1:219 –233.
14. Karibe H, Zarow GJ, Weinstein PR. Use of mild intraischemic hypo-
thermia versus mannitol to reduce infarct size after temporary middle
cerebral artery occlusion in rats. J Neurosurg. 1995;83:93–98.
15. Paczynski RP, He YY, Diringer MN, Hsu CY. Multiple-dose mannitol
reduces brain water content in a rat model of cortical infarction. Stroke.
1997;28:1437–1443.
16. Oktem IS, Menku A, Akdemir H, Kontas O, Kurtsoy A, Koc RK.
Therapeutic effect of tirilazad mesylate (u-74006f), mannitol, and their
combination on experimental ischemia. Res Exp Med (Berl). 2000;199:
231–242.
17. Kaufmann AM, Cardoso ER. Aggravation of vasogenic cerebral edema
by multiple-dose mannitol. J Neurosurg. 1992;77:584 –589.
18. Bereczki D, Liu M, do Prado GF, Fekete I. Mannitol for acute stroke.
Cochrane Database Syst Rev. 2001:CD001153.
19. Santambrogio S, Martinotti R, Sardella F, Porro F, Randazzo A. Is there
a real treatment for stroke? Clinical and statistical comparison of different
treatments in 300 patients. Stroke. 1978;9:130 –132.
20. Bereczki D, Mihalka L, Szatmari S, Fekete K, Di Cesar D, Fulesdi B,
Csiba L, Fekete I. Mannitol use in acute stroke: Case fatality at 30 days
and 1 year. Stroke. 2003;34:1730 –1735.
21. Schwarz S, Schwab S, Bertram M, Aschoff A, Hacke W. Effects of
hypertonic saline hydroxyethyl starch solution and mannitol in patients
with increased intracranial pressure after stroke. Stroke. 1998;29:
1550 –1555.
22. Steiner T, Pilz J, Schellinger P, Wirtz R, Friederichs V, Aschoff A, Hacke
W. Multimodal online monitoring in middle cerebral artery territory
stroke. Stroke. 2001;32:2500 –2506.
23. Manno EM, Adams RE, Derdeyn CP, Powers WJ, Diringer MN. The
effects of mannitol on cerebral edema after large hemispheric cerebral
infarct. Neurology. 1999;52:583–587.
24. Videen TO, Zazulia AR, Manno EM, Derdeyn CP, Adams RE, Diringer
MN, Powers WJ. Mannitol bolus preferentially shrinks non-infarcted
brain in patients with ischemic stroke. Neurology. 2001;57:2120 –2122.
25. Paczynski RP, Venkatesan R, Diringer MN, He YY, Hsu CY, Lin W.
Effects of fluid management on edema volume and midline shift in a rat
model of ischemic stroke. Stroke. 2000;31:1702–1708.
26. Schwab S. Therapy of severe ischemic stroke: Breaking the conventional
thinking. Cerebrovasc Dis. 2005;20(Suppl 2):169 –178.
27. Kempski O. Hypertonic saline and stroke. Crit Care Med. 2005;33:
259 –260.
28. Rabinstein AA. Treatment of cerebral edema. Neurologist. 2006;12:
59 –73.
29. Vialet R, Albanese J, Thomachot L, Antonini F, Bourgouin A, Alliez B,
Martin C. 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;31:1683–1687.
30. Bhardwaj A, Harukuni I, Murphy SJ, Alkayed NJ, Crain BJ, Koehler RC,
Hurn PD, Traystman RJ. Hypertonic saline worsens infarct volume after
transient focal ischemia in rats. Stroke. 2000;31:1694 –1701.
31. Toung TJ, Chen CH, Lin C, Bhardwaj A. Osmotherapy with hypertonic
saline attenuates water content in brain and extracerebral organs. Crit
Care Med. 2007;35:526 –531.
32. Toung TJ, Hurn PD, Traystman RJ, Bhardwaj A. Global brain water
increases after experimental focal cerebral ischemia: Effect of hypertonic
saline. Crit Care Med. 2002;30:644 – 649.
33. Schwarz S, Georgiadis D, Aschoff A, Schwab S. Effects of hypertonic
(10%) saline in patients with raised intracranial pressure after stroke.
Stroke. 2002;33:136 –140.
34. Ogden AT, Mayer SA, Connolly ES Jr. Hyperosmolar agents in neuro-
surgical practice: The evolving role of hypertonic saline. Neurosurgery.
2005;57:207–215.
35. Meyer JS, Itoh Y, Okamoto S, Welch KM, Mathew NT, Ott EO, Sakaki
S, Miyakawa Y, Chabi E, Ericsson AD. Circulatory and metabolic effects
of glycerol infusion in patients with recent cerebral infarction. Circulation.
1975;51:701–712.
36. Sakamaki M, Igarashi H, Nishiyama Y, Hagiwara H, Ando J, Chishiki T,
Curran BC, Katayama Y. Effect of glycerol on ischemic cerebral edema
assessed by magnetic resonance imaging. J Neurol Sci. 2003;209:69 –74.
37. Righetti E, Celani MG, Cantisani T, Sterzi R, Boysen G, Ricci S.
Glycerol for acute stroke. Cochrane Database Syst Rev. 2004:CD000096.
 3094 Stroke November 2007
38. Woodcock J, Ropper AH, Kennedy SK. High dose barbiturates in non-
traumatic brain swelling: ICP reduction and effect on outcome. Stroke.
1982;13:785–787.
39. Rockoff MA, Marshall LF, Shapiro HM. High-dose barbiturate therapy in
humans: A clinical review of 60 patients. Ann Neurol. 1979;6:194 –199.
40. Cormio M, Gopinath SP, Valadka A, Robertson CS. Cerebral hemody-
namic effects of pentobarbital coma in head-injured patients. J Neuro-
trauma. 1999;16:927–936.
41. Schwartz ML, Tator CH, Rowed DW, Reid SR, Meguro K, Andrews DF.
The University of Toronto head injury treatment study: A prospective,
randomized comparison of pentobarbital and mannitol. Can J Neurol Sci.
1984;11:434 – 440.
42. Schwab S, Spranger M, Schwarz S, Hacke W. Barbiturate coma in severe
hemispheric stroke: Useful or obsolete? Neurology. 1997;48:1608 –1613.
43. Stocchetti N, Maas AI, Chieregato A, van der Plas AA. Hyperventilation
in head injury: A review. Chest. 2005;127:1812–1827.
44. Muizelaar JP, Marmarou A, Ward JD, Kontos HA, Choi SC, Becker DP,
Gruemer H, Young HF. Adverse effects of prolonged hyperventilation in
patients with severe head injury: A randomized clinical trial. J Neurosurg.
1991;75:731–739.
45. Marion DW, Puccio A, Wisniewski SR, Kochanek P, Dixon CE, Bullian
L, Carlier P. Effect of hyperventilation on extracellular concentrations of
glutamate, lactate, pyruvate, and local cerebral blood flow in patients with
severe traumatic brain injury. Crit Care Med. 2002;30:2619 –2625.
46. Imberti R, Bellinzona G, Langer M. Cerebral tissue Po2 and SJVo2
changes during moderate hyperventilation in patients with severe
traumatic brain injury. J Neurosurg. 2002;96:97–102.
47. Christensen MS, Paulson OB. Prolonged artificial hyperventilation in
severe cerebral apoplexy. Clinical results and cerebrospinal fluid findings
in a controlled study. Eur Neurol. 1972;8:137–141.
48. Simard D, Paulson OB. Artifical hyperventilation in stroke. Trans Am
Neurol Assoc. 1973;98:309 –310.
49. Yoshida K, Marmarou A. Effects of tromethamine and hyperventilation
on brain injury in the cat. J Neurosurg. 1991;74:87–96.
50. Nagao S, Kitaoka T, Fujita K, Kuyama H, Ohkawa M. Effect of tris-
(hydroxymethyl)-aminomethane on experimental focal cerebral ischemia.
Exp Brain Res. 1996;111:51–56.
Downloaded from http://ahajournals.org by on December 15, 2018
51. Wolf AL, Levi L, Marmarou A, Ward JD, Muizelaar PJ, Choi S, Young
H, Rigamonti D, Robinson WL. Effect of THAM upon outcome in severe
head injury: A randomized prospective clinical trial. J Neurosurg. 1993;
78:54 –59.
52. Torres A, Serra-Batlles J, Ros E, Piera C, Puig de la Bellacasa J, Cobos
A, Lomena F, Rodriguez-Roisin R. Pulmonary aspiration of gastric
contents in patients receiving mechanical ventilation: The effect of body
position. Ann Intern Med. 1992;116:540 –543.
53. Rosner MJ, Coley IB. Cerebral perfusion pressure, intracranial pressure,
and head elevation. J Neurosurg. 1986;65:636 – 641.
54. Schwarz S, Georgiadis D, Aschoff A, Schwab S. Effects of body position
on intracranial pressure and cerebral perfusion in patients with large
hemispheric stroke. Stroke. 2002;33:497–501.
55. Jensen K, Ohrstrom J, Cold GE, Astrup J. The effects of indomethacin on
intracranial pressure, cerebral blood flow and cerebral metabolism in
patients with severe head injury and intracranial hypertension. Acta Neu-
rochir (Wien). 1991;108:116 –121.
56. Schwarz S, Bertram M, Aschoff A, Schwab S, Hacke W. Indomethacin
for brain edema following stroke. Cerebrovasc Dis. 1999;9:248 –250.
57. Qizilbash N, Lewington SL, Lopez-Arrieta JM Corticosteroids for acute
ischaemic stroke. Cochrane Database Syst Rev. 2002:CD000064.
58. Norris JW, Hachinski VC. High dose steroid treatment in cerebral
infarction. BMJ (Clin Res Ed). 1986;292:21–23.
59. Poungvarin N, Bhoopat W, Viriyavejakul A, Rodprasert P, Buranasiri P,
Sukondhabhant S, Hensley MJ, Strom BL. Effects of dexamethasone in
primary supratentorial intracerebral hemorrhage. N Engl J Med. 1987;
316:1229 –1233.
60. Mulley G, Wilcox RG, Mitchell JR. Dexamethasone in acute stroke. BMJ.
1978;2:994 –996.
61. Georgiadis D, Schwab S. Hypothermia in acute stroke. Curr Treat
Options Neurol. 2005;7:119 –127.
62. Maier CM, Ahern K, Cheng ML, Lee JE, Yenari MA, Steinberg GK.
Optimal depth and duration of mild hypothermia in a focal model of
transient cerebral ischemia: Effects on neurologic outcome, infarct size,
apoptosis, and inflammation. Stroke. 1998;29:2171–2180.
63. Markarian GZ, Lee JH, Stein DJ, Hong SC Mild hypothermia: Thera-
peutic window after experimental cerebral ischemia. Neurosurgery. 1996;
38:542–550; discussion 551.
64. Kollmar R, Blank T, Han JL, Georgiadis D, Schwab S Different degrees
of hypothermia after experimental stroke. Short- and long-term outcome.
Stroke. 2007.
65. Georgiadis D, Schwarz S, Aschoff A, Schwab S. Hemicraniectomy and
moderate hypothermia in patients with severe ischemic stroke. Stroke.
2002;33:1584 –1588.
66. Georgiadis D, Schwarz S, Kollmar R, Schwab S. Endovascular cooling
for moderate hypothermia in patients with acute stroke: First results of a
novel approach. Stroke. 2001;32:2550 –2553.
67. Milhaud D, Thouvenot E, Heroum C, Escuret E. Prolonged moderate
hypothermia in massive hemispheric infarction: Clinical experience.
J Neurosurg Anesthesiol. 2005;17:49 –53.
68. Schwab S, Georgiadis D, Berrouschot J, Schellinger PD, Graffagnino C,
Mayer SA. Feasibility and safety of moderate hypothermia after massive
hemispheric infarction. Stroke. 2001;32:2033–2035.
69. Schwab S, Schwarz S, Spranger M, Keller E, Bertram M, Hacke W.
Moderate hypothermia in the treatment of patients with severe middle
cerebral artery infarction. Stroke. 1998;29:2461–2466.
70. De Georgia MA, Krieger DW, Abou-Chebl A, Devlin TG, Jauss M, Davis
SM, Koroshetz WJ, Rordorf G, Warach S. Cooling for acute ischemic
brain damage (cool aid): A feasibility trial of endovascular cooling.
Neurology. 2004;63:312–317.
71. Guluma KZ, Hemmen TM, Olsen SE, Rapp KS, Lyden PD. A trial of
therapeutic hypothermia via endovascular approach in awake patients
with acute ischemic stroke: Methodology. Acad Emerg Med. 2006;13:
820 – 827.
72. Bernard SA, Mac CJB, Buist M. Experience with prolonged induced
hypothermia in severe head injury. Crit Care (Lond). 1999;3:167–172.
73. Gupta R, Connolly ES, Mayer S, Elkind MS. Hemicraniectomy for
massive middle cerebral artery territory infarction: A systematic review.
Stroke. 2004;35:539 –543.
74. Vahedi K, Hofmeijer J, Juettler E, Vicaut E, George B, Algra A, Amelink
GJ, Schmiedeck P, Schwab S, Rothwell PM, Bousser MG, van der Worp
HB, Hacke W. Early decompressive surgery in malignant infarction of the
middle cerebral artery: A pooled analysis of three randomised controlled
trials. Lancet Neurol. 2007;6:215–222.
75. Juttler E, Schwab S, Schmiedek P, Unterberg A, Witte S, Hacke W.
Destiny: Decompressive surgery for the treatment of malignant infarction
of the middle cerebral artery— outcome results [abstract]. Int J Stroke.
2006;1:s38.
76. Vahedi K, Vicaut E, Mateau J, Kurtz A, Orabi M, Guichard JP, Boutron
C, Couvreur G, Touzé E, Rouanet F, Guillon B, Carpentier A, Yelnik A,
George B, Payen D, Bousser MG. Decimal trial: A sequential design,
multicenter, randomized, controlled trial of decompressive craniectomy
in malignant middle cerebral artery (MCA) infarction [abstract]. Int J
Stroke. 2006;1:s38.
77. Hofmeijer J, Amelink GJ, Algra A, van Gijn J, Macleod MR, Kappelle
LJ, van der Worp HB. Hemicraniectomy after middle cerebral artery
infarction with life-threatening edema trial (HAMLET). Protocol for a
randomised controlled trial of decompressive surgery in space-occupying
hemispheric infarction. Trials. 2006;7:29.