The management of brain edema in brain tumors
Evert C.A. Kaal and Charles J. Vecht
This review focuses on pathophysiology, clinical signs, and
imaging of brain edema associated with intracranial tumors
and its treatment. Brain edema in brain tumors is the result of
leakage of plasma into the parenchyma through dysfunctional
cerebral capillaries. The latter type of edema (ie, vasogenic
edema) and the role of other types in brain tumors is
discussed. Vascular endothelial growth factor-induced
dysfunction of tight junction proteins probably plays an
important role in the formation of edema. Corticosteroids are
the mainstay of treatment of brain edema. When possible,
corticosteroids should be used in a low dose (eg, 4 mg
dexamethasone daily) to avoid serious side effects such as
myopathy or diabetes. Higher doses of dexamethasone (16
mg/day or more), sometimes together with osmotherapy
(mannitol, glycerol) or surgery, may be used in emergency
situations. On tapering, one should be aware of the possible
development of corticosteroid dependency or withdrawal
effects.
Novel therapies include vascular endothelial growth factor
receptor inhibitors and corticotropin releasing factor, which
should undergo further clinical testing before they can be
recommended in practice.
Keywords
brain tumor, edema, corticosteroids, toxicity, bloodbrain
barrier, imaging
Curr Opin Oncol 16:593600. 2004 Lippincott Williams & Wilkins.
Department of Neurology, Medical Centre Haaglanden, The Hague, Netherlands
Correspondence to Charles J. Vecht, MD, PhD, Department of Neurology, Medical
Centre Haaglanden, PO Box 432, 2501 CK The Hague, The Netherlands
Tel: 31 70 330 2067; fax: 31 70 330 3113; e-mail: c.vecht@mchaaglanden.nl
Current Opinion in Oncology 2004, 16:593600
Abbreviations
CRF corticotropin-releasing factor
VEGF vascular endothelial growth factor
VEGFR vascular endothelial growth factor receptor
2004 Lippincott Williams & Wilkins
1040-8746
Introduction
Brain edema is a prominent feature of brain cancer and
often contributes to neurologic dysfunction and impaired
quality of life. This review focuses on treatment of brain
edema in brain tumors. Clinical features of patients with
brain tumors are discussed, followed by pathologic and
pathophysiologic aspects of brain edema, and methods
for diagnosing its presence. The practice of corticoste-
roid administration, corticosteroid dependency, and
withdrawal effects are discussed, together with a number
of new agents that can influence the development of
brain edema.
Clinical presentation
Papilledema, occipital headache worsening in the morn-
ing, nausea and vomiting, abnormal eye movements, and
impaired consciousness are the classic signs of raised in-
tracranial pressure. Headache can be a prominent feature
of patients with brain edema and is probably caused by
traction or pressure on pain-sensitive structures such as
dural coverings and blood vessels. In one large series,
60% of patients with brain tumors reported headache.
The pain seems dependent on size and location of the
tumor (infra- vs supratentorial), presence of midline shift,
a prior history of headaches, and the amount of edema
surrounding the tumor [1,2]. Headache in patients with
cancer is an ominous sign: Intracranial metastases were
found in more than 30% of cancer patients with head-
ache as the presenting symptom. Headache duration of
10 weeks or less, emesis, and pain not of a tension type
are significant clinical predictors for the presence of brain
metastases in cancer patients [3].
An important, although less frequently found, feature is
papilledema. However, its absence would not exclude a
brain tumor: Papilledema was absent in more than two
thirds of patients with cerebellar metastases and head-
ache. In older patients with atrophic brains, the occur-
rence of brain edema would often not result in increased
intracranial pressure, probably because of a surplus of
intracranial space, and this may explain the more con-
spicuous absence of papilledema in the elderly. Vomit-
ing is more common in children than in adults, and is
more often associated with infratentorial lesions.
Hemiparesis, dysphasia, cognitive decline, and other fo-
cal neurologic signs may either be the result of brain
edema or of tumor growth. Thirty to 40% of patients with
593
594 Brain and nervous system
brain tumors present with focal neurologic deficits, and a
similar percentage develops epilepsy [4,5].
Infratentorial lesions need special attention given the
vulnerability of the brainstem. Under these conditions, a
small amount of edema may result in severe symptoms of
increased intracranial pressure such as impaired con-
sciousness, and emergency treatment is often necessary.
Pathology and pathophysiology
During the last decade, pathophysiologic mechanisms of
brain edema have been extensively studied. Brain
edema can be defined as an expansion of brain volume
resulting from an increase in water and sodium content.
Two major types of edema can be characterized: intra-
and extracellular edema.
Intracellular edema
Increased intracellular water content results in cellular
swelling. This type of edema is often the result of cyto-
toxic injury such as cerebral ischemia or trauma and is
therefore called cytotoxic edema. In this type of
edema, the primary targets are the ATP-dependent so-
dium pumps: Energy depletion-induced dysfunction of
these pumps results in increased intracellular sodium
levels and, as a consequence, results in the accumulation
of intracellular water. Cytotoxic edema is probably not an
important component of brain tumor edema, although it
may play a role in situations when (micro)circulation is
impeded (eg, after brain herniation).
Extracellular edema
In the extracellular compartment, water can either be
part of the cerebrospinal fluid or of the interstitial fluid.
The production of interstitial fluid is probably driven by
a pressure and osmotic gradient, is often called bulk
flow, and it may have a role in transporting nutrients
and metabolites. An increase in extracellular water leads
to brain edema. This type of edema is mostly located in
the white matter. Its properties, consisting of axons run-
ning parallel to one other and surrounded by an extra-
cellular space with a low cell density, would contribute to
a concentration of the edema within the white matter
and may also serve as a conduit for transporting fluid
[6].
The various types of extracellular edema are as follows.
Vasogenic edema is the most common type in brain
tumors. As a result of increased brain capillary per-
meability and a pressure gradient from vascular to
extracellular compartments, plasma leaks into the
brain parenchyma and follows the pathways of bulk
flow. This type of edema is described in more de-
tail in the next section.
Hydrocephalic edema is the result of the obstruction
of cerebrospinal fluid flow. Edema is formed be-
cause of a hydrostatic pressure gradient between
the ventricles and the brain parenchyma. Brain
capillaries are not affected, and restitution of the
normal cerebrospinal fluid flow will therefore lead
to clearance of hydrocephalic edema.
Osmotic edema is the result of an altered osmotic
gradient between the plasma and the interstitial
fluid. Severe osmotic edema can be seen after wa-
ter intoxication, acute hyponatremia, or too rapid
reduction of hyperosmolarity [7].
Stasis, induced by tumor in venous drainage areas (eg,
compression of an adjacent cortical vein by the tu-
mor), with stasis at the site of the compression,
results in peritumoral edema [8].
An excretorysecretory mechanism in meningiomas in
which tumor-produced substances appear in the
peritumoral tissue was postulated in the 1980s.
Based on electron microscopic studies, a close cor-
relation was found between secretory activity and
production of edema [9].
Hydrodynamic processes, in which fluids originate
from the tumor itself, may contribute to the forma-
tion of edema. It was recently shown that contrast
agent effusion from the extracellular space of me-
ningiomas into the interstitium of the peritumoral
tissue was detectable 3 to 6 hours after contrast
administration [10].
Morphologic and molecular alterations in
the bloodbrain barrier
The bloodbrain barrier is a highly selective interface
separating the brain from the blood. Its most important
component is the capillary endothelial cell. In contrast to
extracerebral capillaries, cerebral endothelial cells are
nonfenestrated, lack intracellular clefts, contain low
numbers of pinocytotic vesicles, have a high mitochon-
drial content, and are enclosed by astrocytic foot pro-
cesses. These endothelial cells are connected by tight
junctions, which have both a high electric impedance
and a low permeability to polar solutes, and they con-
tribute to the selective barrier in this way. Opening of
tight junctions probably plays a key role in the formation
of vasogenic brain edema.
Edema accumulates around brain tumors at a rate of 14
to 78 mL/day [11]. Absorptive mechanisms help to main-
tain equilibrium between edema formation and edema
absorption. Edema is absorbed by transependymal flow
into the ventricles and by absorption into microvessels.
The resulting excess of extracellular protein is removed
by phagocytosis by astrocytes and microglia.
The proteins occludin, claudins, and the junctional ad-
hesion molecule are all part of the molecular composition
of tight junctions in the normal brain [1215,16,17].
These transmembrane proteins bind intracellular pro-
teins such as ZO-1 and ZO-2, and this binding results in
the coupling of tight junctions to the cytoskeleton of

endothelial cells. One has suggested that a decreased
expression or function of these tight junction proteins
leads to opening of the tight junction and to the forma-
tion of edema. Several studies support this hypothesis.
For example, microvessels in glioblastoma multiforme
express only low levels of claudin-1, and high-grade glio-
mas (grades III and IV) do not express functional occlu-
din [18,19]. A key mediator in these mechanisms is vas-
cular endothelial growth factor (VEGF).
Vascular endothelial growth factor
VEGF was originally described as vascular permeability
factor and it functions as a regulator of angiogenesis and
vascular permeability [20]. VEGF binds to endothelial
cells via interaction with the high-affinity tyrosine kinase
receptors flt-1 (VEGFR-1) and Flk-1/KDR (VEGFR2).
These receptors are expressed predominantly on endo-
thelial cells [21,22]. VEGF has a very strong vascular
permeability activity and is a thousand times more po-
tent than histamine [23,24], and probably has direct ef-
fects on the endothelial tight junction (which is de-
scribed later). Besides, VEGF induces edema formation
via the synthesis and release of nitric oxide, an activator
of cyclic GMP-dependent pathways [25]. VEGF may im-
pair the function of occludin: VEGF phosphorylates oc-
cludin, with the opening of tight junctions as a conse-
quence [18]. Others report that VEGF induces
fenestration of the endothelium and enhances capillary
permeability in this way [26]. The role and expression of
VEGF in meningiomas and gliomas is discussed in the
next section. VEGF-mediated vasogenic edema is pre-
sented in Figure 1.
Alternative hypotheses on the formation and regulation
of brain edema have been generated after the discovery
Figure 1. Mechanism of VEGF-mediated vasogenic edema formation
Brain edema Kaal and Vecht 595
of the aquaporin family. In the brain, aquaporin-4 is ex-
pressed in endothelial astrocytic foot processes. Aquapo-
rin-4 is highly upregulated in high-grade gliomas [27,28].
Whether this upregulation results in increased edema
formation or in enhanced clearance of edema is still un-
clear.
Meningiomas
Meningiomas are extraaxial tumors comprising approxi-
mately 20% of all primary intracranial tumors. Meningi-
omas are separated from the brain by the arachnoid, the
subarachnoid space, and the pia mater. In contrast to
intraaxial tumors such as gliomas, the relation between
VEGF expression and grade of malignancy is unclear.
Therefore, the presence of edema does not automatically
imply a high grade of malignancy. For example, secretory
meningiomas are benign meningiomas (World Health
Organization grade I) and display extensive edema sur-
rounding the tumor [2931]. These tumors do not dis-
play severe hypoxia, and other VEGF-regulating factors
may be involved. Upregulation of VEGF is probably
regulated by platelet-derived growth factor, epidermal
growth factor, and estrogens in these tumors [32,33,34].
Gliomas
VEGF expression in gliomas is largely dependent on the
grade of malignancy. Low-grade gliomas express low lev-
els of VEGF, and progression of low-grade gliomas into
malignant gliomas is associated with an up to 50-fold
increase in VEGF mRNA [35]. The expression of VEGF
is largely restricted to the perinecrotic cells, suggesting
that hypoxia regulates its expression in high-grade glio-
mas [36]. An exception to this rule seems to be pilocytic
astrocytoma. This World Health Organization grade I
tumor has a good prognosis and displays high levels
596 Brain and nervous system
VEGF mRNA without necrosis [37]. In these tumors,
VEGF expression may be the result of other factors, such
as a loss of function of the P53 tumor suppresser gene
[38].
Other mediators of edema
Apart from VEGF, other vasoactive substances released
by the tumor itself or by the injured peritumoral tissue
may contribute to the formation of brain edema. Arachi-
donic acid metabolites may be important in this respect:
Elevated levels of leukotriene C4 (produced via the li-
poxygenase pathway) are found in glioblastoma multi-
forme, as well as in the surrounding edematous tissue.
More important, the concentration of leukotriene C4 cor-
relates significantly with the amount of peritumoral
edema [39,40]. Another interesting aspect is the obser-
vation that the degree of peritumoral macrophage infil-
tration is associated with the amount of edema. This has
prompted the suggestion that secretory products of mac-
rophages would contribute to the formation of edema.
The list of other potential vasogenic substances that may
promote the development of brain edema is long, includ-
ing serotonin, thromboxanes, and platelet-activating fac-
tor [4143].
Imaging
Brain edema can be identified with CT and MRI. On
CT and T1-weighted MRI, brain edema can be visual-
ized as a hypodense or hypointense lesion. Brain edema
and other structures with a high water (proton) content
such as cerebrospinal fluid are hyperintense on T2-
weighted MRI. Fluid-attenuated inversion-recovery MR
images provide additive information, because brain
edema is clearly visualized as a hyperintense lesion
against an iso- or hypointense background. Fluid-
attenuated inversion recovery MRI has been compared
with T2-weighted MRI in patients with meningiomas
and is superior to T2-weighted images with respect to
tumor delineation, contrast of tumor to cerebrospinal
fluid, and contrast of tumor to brain parenchyma [44].
Various methods can be used to display alterations in
biochemistry in peritumoral brain edema. Changes in
perfusion have recently been studied with dynamic per-
fusion-weighted MRI. Relative regional cerebral blood
volume and flow were almost 50% lower in peritumoral
brain edema than in the contralateral white matter in
patients with meningiomas [45]. Others found significant
reductions in blood flow, oxygen, and glucose in peritu-
moral edema using positron emission tomography [46].
Intraoperative studies have shown that peritumoral tis-
sue oxygen levels increases up to threefold after decom-
pression by opening the dura [47].
Treatment
Specific therapies such as surgery, osmotherapy, and cor-
ticosteroids contribute to the treatment of brain edema
and are described in the next sections. However, many
patients may benefit from simple general or pharmaco-
logic measures as well. Overhydration may lead to an
increase in interstitial fluid accumulation, especially
when the osmotic gradient is disturbed (eg, in hypona-
tremia). The observation that the diuretic furosemide
has beneficial effects in combination with corticosteroids
supports the importance of adequate hydration [48]. Fur-
thermore, prevention of severe hypertension may be im-
portant: Hypertension leads to increased hydrostatic
pressure and may lead to an increased generation rate of
edema. However, blood pressure reduction should be
done with care to prevent cerebral hypoperfusion and
ischemia. Symptomatic epilepsy should be treated: Sei-
zures create an increased metabolic need and blood flow,
potentially leading to edema [49].
Surgery
Surgery, for the purpose of edema treatment, should be
considered only when life is in immediate danger. Es-
pecially, edema-induced obstructive hydrocephalus can
be treated with shunting. Furthermore, surgery may be
necessary in patients with extensive brain edema and
impaired consciousness. However, surgery in patients
with extensive brain edema may be associated with in-
creased morbidity: It has been shown that intraoperative
increases in intracranial pressure mostly occur in patients
with large amounts of edema. Patients with extensive
preoperative edema may also be at higher risk for post-
operative edema and even brain herniation [5052]. In
practice, pharmacologic treatment with corticosteroids
(which is described later) results in therapeutic effects
within 24 to 48 hours and would help to avoid emergency
surgery.
Osmotherapy
Osmotherapy with mannitol, glycerol, or hypertonic sa-
line is often used in patients with severe brain edema
after stroke or neurotrauma. A typical dose of, for ex-
ample, mannitol is 1 g/kg (250 mL of a 20% solution in
an average adult), resulting in a reduction in intracranial
pressure of 30 to 60% for 2 to 4 hours [53]. Osmotherapy
results in massive osmotic diuresis, so fluid and electro-
lyte balance should be monitored carefully.
Evidence from randomized studies for a role of osmo-
therapy in brain tumor edema is lacking (the same holds
true for stroke and trauma). Osmotherapy may have
negative effects in brain cancer as well. As a result of a
disrupted bloodbrain barrier, the osmotic agent may
leak into the brain parenchyma, with further disturbance
of the osmotic gradient as a result. An alternative to man-
nitol is glycerol. The rebound phenomenon after ad-
ministration of an oral glycerol solution (40 mL 85% glyc-
erol) may be less prominent, because this drug is
metabolized intracellularly [54]. Interestingly, osmo-
therapy may even enhance disturbance of the blood
brain barrier: Mannitol-induced shrinkage of cerebral

capillaries results in the opening of endothelial tight
junctions and is used (in experimental conditions) to al-
low passage of, for example, chemotherapeutics that nor-
mally cannot enter the brain parenchyma [55]. Regarding
all the possible caveats of osmotherapy, we would not
routinely recommend osmotherapy in patients with brain
tumor edema.
Other methods such as hyperventilation, elevation of the
head of the bed and fluid restriction may be applied,
although true evidence is not available.
Corticosteroids
Corticosteroids have been in use since the 1960s, and
although they are associated with substantial side effects,
they play a decisive role in the management of brain
edema associated with primary or secondary brain tumors
[56]. The introduction of corticosteroid therapy in the
early 1960s coincided with a remarkable decline in peri-
operative mortality rates and indicates the importance of
these drugs [57]. Several mechanisms for corticosteroid-
induced edema reduction have been proposed: inhibi-
tion of phospholipase A2, an enzyme of the arachidonic
acid cascade, stabilization of lysosomal membranes, and
improvement of peritumoral microcirculation [58].
Mechanisms of action
Corticosteroid treatment results in a decreased edema
formation rate without a noticeable effect on absorption.
Recent studies have shown that corticosteroids reduce
the expression of the edema-producing factor VEGF
[59,60]. The edema-reducing effect of corticosteroids is
rapid: Reductions in capillary permeability were ob-
served already 1 hour after a single dose of a corticoste-
roid [61]. Apart from an edema-reducing effect, a reduc-
tion in tumor size of 15% after corticosteroid treatment
has been observed, although this effect could not be
confirmed by others [62,63].
Dexamethasone is the most commonly used corticoste-
roid. It is approximately six times as potent as prednisone
(20 mg dexamethasone is equivalent to 130 mg predni-
sone) and reaches full effect within 24 to 72 hours. Dos-
ages of dexamethasone vary between 4 to 100 mg/day.
Dexamethasone has lower mineralocorticoid (salt-
retaining) effects than other corticosteroids. The latter
effect may lead to increased risk of hyponatremia, which
enhances the generation of edema. Therefore, some au-
Table 1. Dexamethasone compared with other corticosteroids
Dose equivalent to
Variable 20 mg cortisol, mg
Dexamethasone 0.75
Hydrocortisone or cortisol 20
Prednisone 5 1236
Methylprednisolone 4 1236
Brain edema Kaal and Vecht 597
thors prefer prednisone instead of dexamethasone [54].
A comparison of dexamethasone with other corticoste-
roids is presented in Table 1.
Randomized trial
One randomized study has addressed the therapeutic ef-
fects and side effects of corticosteroids. Ninety-six pa-
tients with brain metastases were randomized between
4, 8, and 16 mg/day dexamethasone. Administration of 4
mg/day or 16 mg/day dexamethasone resulted in the
same improvement in neurologic function. However, the
increase in side effects was dose dependent. Treatment
with 16 mg/day dexamethasone resulted in a an approxi-
mate 25% increase in the percentage of patients with
proximal muscle weakness 4 weeks after the start of ste-
roid treatment. Muscle weakness did not increase after 4
mg/day [64]. The percentage of cushingoid faces in-
creased significantly (32% vs 65%, 4 mg/day vs 16
mg/day; P = 0.03). These data indicate that many pa-
tients can be treated with 4 mg dexamethasone/day, re-
sulting in significantly less side effects. In patients with
impaired consciousness or signs of increased intracranial
pressure, we would recommend 10 mg intravenously, fol-
lowed by 4 4 mg/day orally.
Side effects
Apart from steroid myopathy, which may be the result of
catabolic effects of corticosteroids, and cushingoid faces
(resulting from the characteristic redistribution of fat
from peripheral to central parts of the body), other seri-
ous side effects are prominent after (high-dose) steroid
treatment [65].
Almost 50% of patients treated with corticosteroids over
a longer period develop a disturbed glucose metabolism,
and in nearly half of these patients, disturbances persist
despite reduction or even withdrawal of the drug [66].
Meta-analyses of the magnitude of the risk corticoste-
roid-induced peptic ulcers provide conflicting results
[67,68]. Treatment with H2 blockers or proton pump
inhibitors is recommended in patients receiving drugs
with potential gastrointestinal side effects, such as non-
steroidal antiinflammatory drugs, anticoagulants, and
high doses of dexamethasone. Also, in the elderly, peri-
operative conditions or in patients with a prior history of
peptic ulcer may benefit from peptic ulcer prophylactics
[69].
Antiinflammatory Mineralocorticoid
Biological activity activity
half-life, h (cortisol = 1) (cortisol = 1)
3654 2530 <0.2
812 1 1
3.5 0.8
5 0.5
598 Brain and nervous system
Psychiatric complications in patients receiving cortico-
steroids seem to be dose dependent as well. A total of
676 patients free of psychiatric disease were investigated
before corticosteroid treatment in the early 1970s. Severe
psychiatric illness (ie, psychosis, mania, or depression)
was uncommon (1.3%) after receiving less than 40
mg/day prednisone (equivalent to 3 mg/day dexametha-
sone), but increased to 18.5% at doses more than 80
mg/day prednisone (equivalent to 12 mg/day dexameth-
asone; as reviewed by Brown and Chandler [70]).
Patients with brain tumors receiving steroids are also at
risk for Pneumocystis carinii pneumonia. In one study of
brain tumors, the reported incidence was almost 2% [71].
Prophylactic treatment with trimethoprimsulfameth-
oxazole may prevent this potentially life-threaten-
ing pneumonia and is mainly applied to patients who
have additional reasons for being immune compromised.
Withdrawal of corticosteroids
After successful treatment (eg, gross tumor resection),
corticosteroids can be tapered. Corticosteroids should be
tapered within 2 to 3 weeks. Decreasing the dose by 50%
every 4 days is a reasonable approach in patients with a
good clinical condition. However, patients with poorly
controlled tumors or extensive brain edema may dete-
riorate after rapid tapering. Corticosteroids should be ta-
pered by 25% every 8 days in the latter group. Eventu-
ally, chronic treatment with a low dose of corticosteroids
(eg, 1 to 2 mg/day) may be needed to have an acceptable
quality of life. However, the risk of long-term side ef-
fects is increased in these patients.
Withdrawal of corticosteroids is associated with morbid-
ity as well. The so-called steroid withdrawal syndrome
was described in the 1960s and is characterized by ar-
thralgias, myalgias, and more general symptoms such as
headache, lethargy, and sometimes low-grade fever [72].
Especially, leg and joint pain can be severe and may
result in an inability to walk. The steroid withdrawal
syndrome usually follows rapid tapering of corticoste-
roids (particularly after prolonged periods of high-dose
use), although it may occur on slow tapering (2 mg
dexamethasone/week) [73].
Another side effect of corticosteroid withdrawal is adre-
nal insufficiency. Administration of corticosteroids in-
duces a negative feedback on the hypothalamic
pituitaryadrenal axis. Tapering that occurs too fast may
lead to secondary adrenal insufficiency with weakness,
pigmentation of skin, and weight loss as the most promi-
nent symptoms. Acute adrenal insufficiency is a feared
but probably rare complication and may lead to hypoten-
sion that is refractory to fluids and requires vasopressors
and intensive care unit admission.
Novel treatments
Corticotropin-releasing factor (CRF) is an endogenous
peptide responsible for the secretion and synthesis of
corticosteroids, and has been studied in animal models
and in patients.
Rats with intracerebral gliomas were treated systemically
with human CRF or dexamethasone. Both agents effec-
tively decreased bloodbrain barrier permeability mea-
sured with MRI. Increased levels of adrenal corticoste-
roids were not observed, and it has been concluded that
human CRF has a direct action on tumor microvascula-
ture, resulting in restoration of bloodbrain barrier integ-
rity [74].
Intravenously administered human CRF resulted in re-
covery of neurologic function in 10 of 17 patients with
brain metastases. Improvement in symptoms did not cor-
relate with changes in cortisol levels. Only a small change
in brain edema was detected with MRI after CRF treat-
ment. Dose-limiting toxicity (hypotension) was observed
at a dose of 4 g/kg/hour. Flushing and hot flashes were
also observed. Lower doses did not result in severe tox-
icity [75]. Although long-term side effects have not been
investigated, the data indicate that further studies are
necessary.
Selectively inhibiting VEGF may be effective, and data
from experimental studies are promising: SU5416, a
small, lipophilic synthetic molecule that selectively in-
hibits the tyrosine kinase activity of the VEGF receptor
Flk-1/KDR, was administrated systemically in rats with
intracerebral gliosarcoma. SU5416 prolonged survival
without any significant systemic adverse effect. Histo-
logic analysis of the treated tumors showed an increase in
necrosis and reduced vascularity after treatment with
SU5416 [76]. SU5416 is currently being tested in various
phase II studies and has shown biologic activity in pa-
tients with hematologic malignancies [77,78].
The cyclooxygenase-2 inhibitor SC-236 has been inves-
tigated recently. Rats with intracerebral gliosarcomas re-
ceived SC-236, dexamethasone, or no treatment. Median
survival increased almost 50% after SC-236 or dexameth-
asone treatment [79]. In another study, rats with intra-
cerebral gliosarcomas were treated with rofecoxib, an-
other selective cyclooxygenase-2 inhibitor. Rofecoxib
was as effective as dexamethasone in decreasing the dif-
fusion of contrast material into the brain parenchyma,
suggesting a reduction in bloodtumor barrier permeabil-
ity. The data from the aforementioned studies indicate
that selective cyclooxygenase-2 inhibitors appear to be as
effective as dexamethasone in preventing brain edema
and that survival is enhanced in these animal models. In
the past, other nonsteroidal antiinflammatory drugs have
shown variable results in animal models [80]. In a small
series, the classic cyclooxygenase-2 inhibitor ibuprofen

improved the Karnofsky performance score in 40% of
patients [81]. Novel cyclooxygenase-2 inhibitors need
further testing in humans with tumor-induced brain
edema.
Conclusion
Corticosteroids are the mainstay of treatment for brain
edema. Side effects are acceptable when using low doses
of steroids (eg, dexamethasone 2 2 mg/day orally). In
acute situations, a bolus of 10 mg dexamethasone intra-
venously may be necessary, followed by 16 mg dexa-
methasone in divided daily doses. Corticosteroids should
be tapered or maintained at the lowest effective dose
when possible. General therapeutic measures such as
regulation of fluid, electrolyte balance, and serum glu-
cose, and treatment of seizures and severe hypertension
may help to prevent further clinical deterioration. Proton
pump inhibitors or H2 blockers and PCP prophylactics
should be used on indication. Surgery as a means of
edema treatment only and osmotherapy are associated
with potentially serious side effects and should be used
in emergency situations only. Experimental treatment
with CRF, VEGF, and cyclooxygenase inhibitors seems
promising, but should undergo further clinical testing
before it can be recommended for use in routine practice.
References and recommended reading
Papers of particular interest, published within the annual period of review,
are highlighted as:
Of special interest
Of outstanding interest
1 Pfund Z, Szapary L, Jaszberenyi O, et al.: Headache in intracranial tumors.
Cephalalgia 1999, 19:787790.
2 Forsyth PA, Posner JB: Headaches in patients with brain tumors: a study of
111 patients. Neurology 1993, 43:16781683.
3 Christiaans MH, Kelder JC, Arnoldus EP, et al.: Prediction of intracranial me-
tastases in cancer patients with headache. Cancer 2002, 94:20632068.
4 Posner JB: Overview. In Neurologic Complications of cancer. Edited by Pos-
ner JB. Philadelphia: FA Davis; 1995:114.
5 Posner JB: Intracranial metastases. In Neurologic Complications of Cancer.
Edited by Posner JB. Philadelphia: FA Davis; 1995:77110.
6 Thapar K, Taylor MD, Laws ER, et al: Brain edema, increased intracranial
pressure, and vascular effects of human brain tumors. In Brain Tumors. An
Encyclopedic Approach. Edited by Kaye AH, Laws ER. London: Churchill
Livingstone; 2001:189216.
A comprehensive review describing brain tumor edema in detail. Recently discov-
ered molecular mechanisms are obviously not included.
7 Milhorat TH: Classification of the cerebral edemas with reference to hydro-
cephalus and pseudotumor cerebri. Childs Nerv Syst 1992, 8:301306.
8 Hiyama H, Kubo O, Tajika Y, et al.: Meningiomas associated with peritumoural
venous stasis: three types on cerebral angiogram. Acta Neurochir (Wien)
1994, 129:3138.
9 Philippon J, Foncin JF, Grob R, et al.: Cerebral edema associated with me-
ningiomas: possible role of a secretoryexcretory phenomenon. Neurosur-
gery 1984, 14:295301.
10 Bitzer M, Nagele T, GeistBarth B, et al.: Role of hydrodynamic processes in
the pathogenesis of peritumoral brain edema in meningiomas. J Neurosurg
2000, 93:594604.
A study investigating the formation of edema in various types of meningiomas.
11 Reulen HJ, Huber P, Ito U, et al.: Peritumoral brain edema. A keynote address.
Adv Neurol 1990, 52:307315.
12 Furuse M, Hirase T, Itoh M, et al.: Occludin: a novel integral membrane protein
localizing at tight junctions. J Cell Biol 1993, 123:17771788.
Brain edema Kaal and Vecht 599
13 Furuse M, Fujita K, Hiiragi T, et al.: Claudin-1 and -2: novel integral membrane
proteins localizing at tight junctions with no sequence similarity to occludin. J
Cell Biol 1998, 141:15391550.
14 Hirase T, Staddon JM, Saitou M, et al.: Occludin as a possible determinant of
tight junction permeability in endothelial cells. J Cell Sci 1997, 110:1603
1613.
15 MartinPadura I, Lostaglio S, Schneemann M, et al.: Junctional adhesion mol-
ecule, a novel member of the immunoglobulin superfamily that distributes at
intercellular junctions and modulates monocyte transmigration. J Cell Biol
1998, 142:117127.
16 Davies DC: Bloodbrain barrier breakdown in septic encephalopathy and
brain tumours. J Anat 2002, 200:639646.
A useful review on molecular alterations in endothelial tight junctions in patients
with brain tumors.
17 Papadopoulos MC, Saadoun S, Davies DC, et al.: Emerging molecular
mechanisms of brain tumour oedema. Br J Neurosurg 2001, 15:101108.
A review on the role of occludins, claudins and aquaporins in the endothelial tight
junctions in brain tumors.
18 Papadopoulos MC, Saadoun S, Woodrow CJ, et al.: Occludin expression in
microvessels of neoplastic and non-neoplastic human brain. Neuropathol
Appl Neurobiol 2001, 27:384395.
19 Liebner S, Fischmann A, Rascher G, et al.: Claudin-1 and claudin-5 expres-
sion and tight junction morphology are altered in blood vessels of human
glioblastoma multiforme. Acta Neuropathol (Berl) 2000, 100:323331.
20 Senger DR, Galli SJ, Dvorak AM, et al.: Tumor cells secrete a vascular per-
meability factor that promotes accumulation of ascites fluid. Science 1983,
219:983985.
21 Millauer B, WizigmannVoos S, Schnurch H, et al.: High affinity VEGF bind-
ing and developmental expression suggest Flk-1 as a major regulator of vas-
culogenesis and angiogenesis. Cell 1993, 72:835846.
22 de Vries C, Escobedo JA, Ueno H, et al.: The fms-like tyrosine kinase, a re-
ceptor for vascular endothelial growth factor. Science 1992, 255:989991.
23 Connolly DT: Vascular permeability factor: a unique regulator of blood vessel
function. J Cell Biochem 1991, 47:219223.
24 Criscuolo GR: The genesis of peritumoral vasogenic brain edema and tumor
cysts: a hypothetical role for tumor-derived vascular permeability factor. Yale
J Biol Med 1993, 66:277314.
25 Mayhan WG: VEGF increases permeability of the bloodbrain barrier via a
nitric oxide synthase/cGMP-dependent pathway. Am J Physiol 1999,
276:C1148C1153.
26 Roberts WG, Palade GE: Neovasculature induced by vascular endothelial
growth factor is fenestrated. Cancer Res 1997, 57:765772.
27 Nielsen S, Nagelhus EA, AmiryMoghaddam M, et al.: Specialized membrane
domains for water transport in glial cells: high-resolution immunogold cyto-
chemistry of aquaporin-4 in rat brain. J Neurosci 1997, 17:171180.
28 Saadoun S, Papadopoulos MC, Davies DC, et al.: Aquaporin-4 expression is
increased in oedematous human brain tumours. J Neurol Neurosurg Psychia-
try 2002, 72:262265.
29 Nishikawa R, Cheng SY, Nagashima R, et al.: Expression of vascular endo-
thelial growth factor in human brain tumors. Acta Neuropathol (Berl) 1998,
96:453462.
30 Pietsch T, Valter MM, Wolf HK, et al.: Expression and distribution of vascular
endothelial growth factor protein in human brain tumors. Acta Neuropathol
(Berl) 1997, 93:109117.
31 Samoto K, Ikezaki K, Ono M, et al.: Expression of vascular endothelial growth
factor and its possible relation with neovascularization in human brain tumors.
Cancer Res 1995, 55:11891193.
32 Maxwell M, Galanopoulos T, HedleyWhyte ET, et al.: Human meningiomas
co-express platelet-derived growth factor (PDGF) and PDGF-receptor genes
and their protein products. Int J Cancer 1990, 46:1621.
33 Speirs V, BoyleWalsh E, Fraser WD: Constitutive co-expression of estrogen
and progesterone receptor mRNA in human meningiomas by RT-PCR and
response of in vitro cell cultures to steroid hormones. Int J Cancer 1997,
72:714719.
34 Machein MR, Plate KH: VEGF in brain tumors. J Neurooncol 2000, 50:109
120.
A review describing the role of VEGF in gliomas, hemangioblastomas and menin-
giomas.
35 Plate KH, Breier G, Weich HA, et al.: Vascular endothelial growth factor is a
potential tumour angiogenesis factor in human gliomas in vivo. Nature 1992,
359:845848.
600 Brain and nervous system
36 Shweiki D, Itin A, Soffer D, et al.: Vascular endothelial growth factor induced
by hypoxia may mediate hypoxia-initiated angiogenesis. Nature 1992,
359:843845.
37 Leung SY, Chan AS, Wong MP, et al.: Expression of vascular endothelial
growth factor and its receptors in pilocytic astrocytoma. Am J Surg Pathol
1997, 21:941950.
38 Mukhopadhyay D, Tsiokas L, Sukhatme VP: Wild-type p53 and v-Src exert
opposing influences on human vascular endothelial growth factor gene ex-
pression. Cancer Res 1995, 55:61616165.
39 Black KL, Hoff JT, McGillicuddy JE: Increased leukotriene C4 and vasogenic
edema surrounding brain tumors in humans. Ann Neurol 1986, 19:592595.
40 Nathoo N, Barnett GH, Golubic M: The eicosanoid cascade: possible role in
gliomas and meningiomas. J Clin Pathol 2004, 57:613.
41 Gaetani P, Butti G, Chiabrando C, et al.: A study on the biological behavior of
human brain tumors. Part I. Arachidonic acid metabolism and DNA content. J
Neurooncol 1991, 10:233240.
42 Hirashima Y, Hayashi N, Fukuda O, et al.: Platelet-activating factor and edema
surrounding meningiomas. J Neurosurg 1998, 88:304307.
43 Forman A: Altered mental status. In Cancer in the Nervous System. Edited by
Levin VA. Oxford: University Press; 2002:543547.
44 Shimizu T, Miki H, Takeguchi T, et al.: Usefulness of turbo-fluid-attenuated
inversion-recovery (tFLAIR) sequence in diagnosing meningioma. Radiat Med
2003, 21:5561.
45 Uematsu H, Maeda M, Itoh H: Peritumoral brain edema in intracranial menin-
giomas evaluated by dynamic perfusion-weighted MR imaging: a preliminary
study. Eur Radiol 2003, 13:758762.
46 Hino A, Imahori Y, Tenjin H, et al.: Metabolic and hemodynamic aspects of
peritumoral low-density areas in human brain tumor. Neurosurgery 1990,
26:615621.
47 Pennings FA, Bouma GJ, Kedaria M, et al.: Intraoperative monitoring of brain
tissue oxygen and carbon dioxide pressures reveals low oxygenation in peri-
tumoral brain edema. J Neurosurg Anesthesiol 2003, 15:15.
48 Meinig G, Reulen HJ, Simon RS, et al.: Clinical, chemical, and CT evaluation
of short-term and long-term antiedema therapy with dexamethasone and di-
uretics. Adv Neurol 1980, 28:471489.
49 Cloughesy TF, Black KL: Peritumoral edema. In The Gliomas. Edited by
Berger MS, Wilson CB. Philadelphia: WB Saunders; 1999:107114.
50 Bedford RF, Morris L, Jane JA: Intracranial hypertension during surgery for
supratentorial tumor: correlation with preoperative computed tomography
scans. Anesth Analg 1982, 61:430433.
51 Ciric I, Ammirati M, Vick N, et al.: Supratentorial gliomas: surgical consider-
ations and immediate postoperative results. Gross total resection versus par-
tial resection. Neurosurgery 1987, 21:2126.
52 Fadul C, Wood J, Thaler HT, et al.: Morbidity and mortality of craniotomy for
excision of supratentorial gliomas. Neurology 1998, 38:13741379.
53 Weiss HD: Neoplasms. In Manual of Neurologic Therapeutics. Edited by
Samuels MA. Boston: Little, Brown and Company; 1995:224229.
54 Grisold W, Krauseneck P, Muller B: Praktische Neuroonkologie. Vienna:
Springer; 2000.
55 Kroll RA, Neuwelt EA: Outwitting the bloodbrain barrier for therapeutic pur-
poses: osmotic opening and other means. Neurosurgery 1998, 42:1083
1099.
56 Ruderman N, Hall T: Use of glucocorticoids in the palliative treatment of meta-
static brain tumors. Cancer 1965, 18:298306.
57 Jelsma R, Bucy PC: The treatment of glioblastoma multiforme of the brain. J
Neurosurg 1967, 27:388400.
58 Yamada K, Ushio Y, Hayakawa T: Effects of steroids on the bloodbrain bar-
rier. In Implications of the BloodBrain Barrier and Its Manipulation. Edited by
Neuwelt EA. New York: Plenum Press; 1989:5376.
59 Heiss JD, Papavassiliou E, Merrill MJ, et al.: Mechanism of dexamethasone
suppression of brain tumor-associated vascular permeability in rats. Involve-
ment of the glucocorticoid receptor and vascular permeability factor. J Clin
Invest 1996, 98:14001408.
60 Machein MR, Kullmer J, Ronicke V, et al.: Differential downregulation of vas-
cular endothelial growth factor by dexamethasone in normoxic and hypoxic rat
glioma cells. Neuropathol Appl Neurobiol 1999, 25:104112.
61 Shapiro WR, Hiesiger EM, Cooney GA, et al.: Temporal effects of dexameth-
asone on blood-to-brain and blood-to-tumor transport of 14C-alpha-
aminoisobutyric acid in rat C6 glioma. J Neurooncol 1990, 8:197204.
62 Leiguarda R, Sierra J, Pardal C, et al.: Effect of large doses of methylprednis-
olone on supratentorial intracranial tumors. A clinical and CAT scan evalua-
tion. Eur Neurol 1985, 24:2332.
63 Hatam A, Bergstrom M, Yu ZY, et al.: Effect of dexamethasone treatment on
volume and contrast enhancement of intracranial neoplasms. J Comput Assist
Tomogr 1983, 7:295300.
64 Vecht CJ, Hovestadt A, Verbiest HB, et al.: Doseeffect relationship of dexa-
methasone on Karnofsky performance in metastatic brain tumors: a random-
ized study of doses of 4, 8, and 16 mg per day. Neurology 1994, 44:675
680.
The only randomised study on dose-dependent toxicity of corticosteroids in pa-
tients with brain tumors.
65 Batchelor TT, Taylor LP, Thaler HT, et al.: Steroid myopathy in cancer patients.
Neurology 1997, 48:12341238.
66 Meyer G, Badenhoop K: Glucocorticoid-induced insulin resistance and dia-
betes mellitus. Receptor-, postreceptor mechanisms, local cortisol action,
and new aspects of antidiabetic therapy. Med Klin 2003, 98:266270.
67 Gotzsche PC: Steroids and peptic ulcer: an end to the controversy? J Intern
Med 1994, 236:599601.
68 Conn HO, Poynard T: Corticosteroids and peptic ulcer: meta-analysis of ad-
verse events during steroid therapy. J Intern Med 1994, 236:619632.
69 Wen PY, Marks PW: Medical management of patients with brain tumors. Curr
Opin Oncol 2002, 14:299307.
A practical review focusing on management of patients with brain tumors including
steroids.
70 Brown ES, Chandler PA: Mood and cognitive changes during systemic cor-
ticosteroid therapy. Primary care companion. J Clin Psychiatry 2001,
3:1721.
71 Kovacs JA, Gill VJ, Meshnick S, et al.: New insights into transmission, diag-
nosis, and drug treatment of Pneumocystis carinii pneumonia. JAMA 2001,
286:24502460.
72 Amatruda TT, Hurst MH, DEsopo ND: Certain endocrine and metabolic fac-
ets of the steroid withdrawal syndrome. J Clin Endocrinol 1965, 25:1207
1217.
73 Posner JB: Supportive care of the neuro-oncology patient. In Management in
Neuro-oncology. Edited by Hildebrand J. Berlin: SpringerVerlag; 1992:89
103.
74 Tjuvajev J, Kolesnikov Y, Joshi R, et al.: Anti-neoplastic properties of human
corticotropin releasing factor: involvement of the nitric oxide pathway. In Vivo
1998, 12:110.
75 VillalonaCalero MA, Eckardt J, Burris H, et al.: A phase I trial of human cor-
ticotropin-releasing factor (hCRF) in patients with peritumoral brain edema.
Ann Oncol 1998, 9:7177.
76 Takamoto T, Sasaki M, Kuno T, et al.: Flk-1 specific kinase inhibitor (SU5416)
inhibited the growth of GS-9L glioma in rat brain and prolonged the survival.
Kobe J Med Sci 2001, 47:181191.
77 Zangari M, Anaissie E, Stopeck A, et al.: Phase II study of SU5416, a small
molecule vascular endothelial growth factor tyrosine kinase receptor inhibitor,
in patients with refractory multiple myeloma. Clin Cancer Res 2004,
10:8895.
78 Fiedler W, Mesters R, Tinnefeld H, et al.: A phase 2 clinical study of SU5416
in patients with refractory acute myeloid leukemia. Blood 2003, 102:2763
2767.
79 Portnow J, Suleman S, Grossman SA, et al.: A cyclooxygenase-2 (COX-2)
inhibitor compared with dexamethasone in a survival study of rats with intra-
cerebral 9L gliosarcomas. Neurooncology 2002, 4:2225.
80 Weissman DE, Stewart C: Experimental drug therapy of peritumoral brain
edema. J Neurooncol 1988, 6:339342.
81 Del Maestro RF, Megyesi JF, Farrell CL: Mechanisms of tumor-associated
edema: a review. Can J Neurol Sci 1990, 17:177183.