Neurosurg Rev
DOI 10.1007/s10143-017-0839-7
REVIEW
Vascular hyperpermeability as a hallmark of phacomatoses: is
the etiology angiogenesis comparable with mechanisms seen
in inflammatory pathways? Part I: historical observations
and clinical perspectives on the etiology of increased CSF protein
levels, CSF clotting, and communicating hydrocephalus:
a comprehensive review
Yosef Laviv 1,3
&
Burkhard S Kasper 2 & Ekkehard M. Kasper 1
Received: 6 December 2016 / Revised: 18 February 2017 / Accepted: 21 February 2017
# Springer-Verlag Berlin Heidelberg 2017
Abstract Phacomatoses are a special group of familial
hamartomatous syndromes with unique neuro-cutaneous
manifestations as well as disease characteristic tumors.
Neurofibromatosis 2 (NF2) and tuberous sclerosis complex (TSC) are representatives of this family. Vestibular
schwannoma (VS) and subependymal giant cell tumor
(SGCT) are two of the most common intracranial tumors
associated with NF2 and TSC, respectively. These tumors
can present with obstructive hydrocephalus due to their
location adjacent to or in the ventricles. However, both
tumors are also known to have a unique association with
an elevated protein concentration in the cerebrospinal fluid (CSF), sometimes in association with non-obstructive
(communicating) hydrocephalus (HCP), the causality of
which has been unclear. Furthermore, SGCTs have repeatedly been shown to have a predisposition for CSF
clotting, causing debilitating obstructions and recurrent
malfunctions in shunted patients. However, the exact relation between high protein levels and spontaneous
* Yosef Laviv
ylaviv@BIDMC.harvard.edu
1
Division of Neurosurgery, Beth Israel Deaconess Medical Center,
Harvard Medical School, Boston, MA, USA
2
Department of Neurology/Epilepsy Centre, University of Erlangen,
Erlangen, Germany
3
Department of Surgery, Division of Neurosurgery, Beth Israel
Deaconess Medical Center, Harvard Medical School, West Campus,
Lowry Medical Office Building, Suite 3B 110 Francis, St.
Boston, MA 02215, USA
clotting of the CSF is not clear, nor is the mechanism
understood by which CSF may clot in SGCTs. Elevated
protein levels in the CSF are thought to be caused by
increased vascular permeability and dysregulation of the
blood–brain barrier. The two presumed underlying pathophysiologic mechanisms for that, in the context of tumorigenesis, are angiogenesis and inflammation. Both mechanisms are correlated to the Pi3K/Akt/mTOR pathway
which is a major tumorigenesis pathway in nearly all
phacomatoses. In this review, we discuss the influence
of angiogenesis and inflammation on vascular permeability in VSs and SGCTs at the phenotypic level as well as
their possible genetic and molecular determinants. Part I
describes the historical perspectives and clinical aspects
of the relationship between vascular permeability, abnormal CSF protein levels, clotting of the CSF, and communicating HCP. Part II describes different cellular and molecular pathways involved in angiogenesis and inflammation in these two tumors and the correlation between inflammation and coagulation. Interestingly, while increased
angiogenesis can be observed in both VS and SGCT, inflammatory processes seem more prominent in SGCT.
Both pathologies are characterized by different subgroups
of tumor-associated macrophages (TAM): the pro-inflammatory, M1 type is predominating in SGCTs while proangiogenetic, M2 type is predominating in VSs. We suggest that lack of NF2 protein in VS and lack of TSC1/2
proteins in SGCT determine this fundamental difference
between the two tumor types, by defining the predominant
TAM type. Since inflammatory reactions and coagulation
processes are tightly connected, a Bpro-inflammatory
state^ of SGCT can be used to explain the observed
Neurosurg Rev
associated enhanced CSF clotting process. These distinct
cellular and molecular differences may have direct therapeutic implications on tumors that are unique to certain
phacomatoses or those with similar genetics.
Keywords Phacomatoses . Neurofibromatosis . Tuberous
sclerosis . Cerebrospinal fluid . Vascular permeability .
Angiogenesis . Inflammation . Protein . Giant cell tumor .
Schwannoma
Abbreviations
BBB
Blood–brain barrier
(c)HCP (Communicating) hydrocephalus
CNS
Central nervous system
CSF
Cerebrospinal fluid
EC
Endothelial cell
eNOS
Endothelial nitric oxide synthase
HGB
Hemangioblastoma
ICP
Intracranial pressure
mTOR Mammalian target of rapamycin
NF
Neurofibromatosis
NO
Nitric oxide
Pi3K
Phosphatidylinositol-3-kinase
SGCT
Subependymal giant cell tumor
TSC
Tuberous sclerosis complex
VEGF
Vascular endothelial growth factor
vHL
von Hippel–Lindau
VPS
Ventriculoperitoneal shunt
VS
Vestibular schwannoma
Introduction
In 1923, the European ophthalmologist Van der Hoeve introduced the term Bphacomatoses^ into the literature to describe
a hereditary group of diseases characterized by the presence of
disseminated hamartomas in neuro-ectodermal-derived tissues [99]. The technical term is derived from the Greek word
BPhakos,^ meaning Blentil,^ which was used to describe the
lens of the eye, an organ that is characteristically involved in
all phacomatoses. It also carries the meaning of Bspot^ or
Bbirth mark,^ emphasizing both the patchy (or Bspotty^) involvement of the skin and various other organs as well as the
hereditary nature of these syndromes. Although each original
description of any of the classic phacomatoses was based on
careful clinical and pathologic observations, we know now
that phacomatoses are bound together by complex genetics
and molecular pathways. These autosomal dominantly
inherited disorders include neurofibromatoses 1 and 2 (NF1/2); tuberous sclerosis complexes 1 and 2 (TSC-1/2) and von
Hippel–Lindau disease (vHL); and the less known Sturge–
Weber syndrome, ataxia telangiectasia, nevoid basal cell carcinoma syndrome, Cowden disease, Peutz–Jeghers syndrome,
familial adenomatous polyposis, and juvenile polyposis [98].
Each of the genetic mutations underlying a phacomatosis encodes a tumor suppressor protein that functions in signal transduction [98]. Some tumors may occur sporadically, without
the associated manifestations of the respective syndromes
[29], yet their molecular signature alterations and phenotype
presentation are similar to tumors encountered in the setting of
syndromic phacomatoses [29].
Subependymal giant cell tumors (SGCTs; characteristic of
TSC) and vestibular schwannomas (VSs; characteristic of
neurofibromatosis 2 (NF2)) are two benign, intracranial tumors related to phacomatoses. Both locate to distinct CSF
spaces, namely the lateral ventricle and the cerebellopontine
angle cistern, respectively. Interestingly, these two unique tumors are the only intracranial benign neoplasms repeatedly
associated with elevated CSF protein levels (=
proteinorhachia), sometimes in the context of symptomatic
non-obstructive hydrocephalus. However, of the two, only
SGCT is associated with a true clotting tendency of the CSF,
which frequently leads to refractory shunt obstructions [66].
Increased cerebrospinal fluid (CSF) protein level results as
a consequence of interrupted blood–brain barrier (BBB). The
BBB is a complex, highly selective anatomical and physiological barrier, which regulates the entry and egress of metabolites such as cerebral nutrients and other biological substances essential for cerebral metabolism and neuronal activity
[20]. The BBB is formed primarily by the layer of endothelial
cells lining the microvessels within the brain parenchyma
along with their closely associated astrocytic processes.
Neighboring endothelial cells are joined via tight junctions
that effectively block free passage of molecules and ions via
the paracellular route across this layer. Any substance exchange with the brain across the barrier must hence be processed through the cells and must either pass through the lipid
portions of the plasma membranes (if hydrophilic) or be transferred via specific transporters localized in the apical or basilar
membranes of the cells [54]. Central nervous system (CNS)
pathologies, such as trauma, ischemia, inflammation, malignancy, and degenerative and demyelinating diseases, can
cause interruption and breakdown of the BBB [2], leading to
spillage of intravascular plasma proteins into the brain parenchyma and CSF. This pathological Bopening^ of the BBB can
also be referred to as vascular hyperpermeability.
Traditionally, the term vascular permeability plays a paramount role in basal vascular sieving of solutes and small molecules in an unstimulated setting [24] and is essential for normal tissue homoeostasis. When dysregulated though,
hyperpermeability becomes an important feature of many
pathological processes, including cancer [25]. Molecular regulators of vascular permeability include angiogenesis-
Neurosurg Rev
associated growth factors (such as vascular endothelial growth
factor (VEGF)), activating factors (such as platelet-activating
factor (PAF)), and inflammatory cytokines (such as histamine,
bradykinin, and nitric oxide) [24, 37].
In this first part of our manuscript, we will describe the
seminal historic observations made on the topic. We shall
review clinical aspects regarding changes in CSF protein
levels seen in association with CNS tumors, and we will correlate these parameters to the occurrence of communicating
hydrocephalus and clotting of the CSF that is characteristic for
SGCT and VS.
Communicating hydrocephalus and elevated CSF protein
levels
Hydrocephalus (HCP) is characterized by abnormal CSF dynamics that leads to CSF accumulation with dilatation of the
ventricles. HCP can be classified as either obstructive with
mechanical blockage of CSF circulation or communicating
(= non-obstructive) from presumed impaired CSF absorption.
This absorption is thought to occur via arachnoid granulations
into the venous sinuses [62]. Common etiologies of communicating hydrocephalus (cHCP) are idiopathic, post subarachnoidal hemorrhage, and post meningitis. cHCP may also be
associated with conditions causing increased CSF protein
levels, such as in, e.g., Guillain-Barré syndrome [7]. Other
scenarios include cHCP secondary to intraspinal tumors,
which occurs in about 1% of patients with benign spinal cord
tumors [7] and in as many as 65% of malignant spinal tumor
cases [23]. In patients with intracranial tumor, cHCP is most
frequently observed in the setting of leptomeningeal metastasis [78, 83], with an incidence of about 25% [58]. It can also
be associated with supratentorial gliomas in ca. 2.5% of cases
[31].
Interestingly, in the largest series published on this topic for
supratentorial gliomas, 95% of hydrocephalic cases occurred
in high-grade gliomas after at least one surgical resection [31].
The etiology of cHCP in this setting may be multifactorial and
related to post surgical changes, post radiation changes, and
leptomeningeal spread leading to high protein content of the
CSF [31, 55, 74]. Most of these factors, however, cannot explain the occurrence of cHCP in association with benign, single, small, previously untreated intracranial tumor. The most
likely explanation for this rare scenario is some form of de
novo increase in CSF protein concentration.
Almost all CSF protein is derived from serum [44]. There is
a 200-fold difference in the steady-state concentration between these two compartments (about 35 and 7000 mg/dl in
CSF and serum, respectively), and this difference is maintained because of the BBB [44]. Several aspects are of interest
here. Serum contains three main protein constituents: albumin, globulins, and fibrinogen [3]. If the BBB is intact and
undisturbed, all serum proteins with a molecular weight
greater than 160 kDa are largely excluded from the CSF
[42]. For example, fibrinogen (whose molecular weight is
340 kDa) has a normal CSF concentration of 0.6 mg/dl with
a plasma-to-CSF ratio of 4940. However, when the BBB is
dysfunctional, certain macromolecules from the serum may be
readily detected in the CSF [44]. In fact, five mechanisms for
pathological changes in distribution of proteins in the CSF
have been described [44]: (a) increased entry of plasma proteins due to disruption in BBB, (b) defective reabsorption of
CSF proteins via the arachnoid villi, (c) primary changes in
plasma protein (rare, example, elevated Bence–Jones protein
in CSF in multiple myeloma patients), (d) local immunoglobulin production due to various chronic and subacute inflammatory diseases of the CNS, and (e) degenerative patterns.
In their classic monograph (1937), Merritt and FremontSmith provided the most detailed description regarding CSF
findings in patients with brain tumors [71]. In their series of
106 patients, the second most common finding (after elevated
opening pressure) was an elevated total protein content. Total
CSF protein was >100 mg/dl (normal 15–45 mg/dl) in 29% of
glioma patients and was 45–100 mg/dl in another 37% of
cases. Furthermore, higher protein levels were associated with
higher pathological tumor grades. It was suggested that such
changes are due to increased passage of plasma proteins into
the CSF across an abnormal BBB or due to local inflammatory response and necrosis generated by the tumor itself [56,
75]. Importantly, Merrit and Fermont-Smith found VS to be
an exception of this rule; these lesions were commonly associated with a total CSF protein contents of 100–500 mg/dl in
the absence of any significant necrosis or inflammation [71].
In 1954, Gardner et al. described two cases in which small,
benign intracranial and intraspinal tumors were associated
with high CSF protein content, increased intracranial pressure,
and papilledema [46]. The first case was that of a cauda equina
ependymoma with a CSF protein level of 425 mg/dl, yellow
fluid, and no cells. In the second case, a 2-cm VS was associated with a CSF protein level of 400 mg/dl. After resection of
the tumor, protein levels dropped to 31 mg/dl and the associated HCP resolved. It was noted that high protein levels in the
CSF caused Bclotting of the CSF, with obstruction of the
Pacchionian bodies^ [46]. Interestingly, the authors have suggested that the chronicity of the case is more important than
the actual level of the protein (i.e., the cause of the malabsorption is the gradual blocking of CSF egress and not the increased osmotic tension of the protein-enriched fluid). In a
related experiment, the authors injected a dog with high levels
of albumin and gamma-globulin (two components of CSF’s
proteins) into its subarachnoid cisterns. This maneuver failed
to produce papilledema, and protein levels declined rapidly
from 1000 mg/dl immediately post injection to 100–200 mg/
dl at 5 h and back to normal levels at 24 h. Accordingly, the
authors concluded that in order to achieve and maintain high
protein levels in the CSF, the entry of proteins into it must be
Neurosurg Rev
constant and at high rate [46]. Their final conclusion was that
Ba small tumor located anywhere along the cerebrospinal axis
can cause increased intracranial pressure, papilledema and
communicating hydrocephalus^ through constantly elevated
levels of protein in the CSF and that these sequelae are reversible once the tumor is resected [46]. In a separate experiment,
an elevation of CSF total protein level by injection of serum
itself resulted in a decreased rate of absorption of CSF albumin. Thus, a vicious cycle exists, whereby a gradual increase
in protein further slows protein absorption [88].
It should be noted that it is still unclear whether high protein concentrations in the CSF contribute to cHCP formation
through a malabsorptive or a hyperosmotic mechanism.
Proteins are large macromolecules that are not diffusible
through the BBB and therefore contribute to larger osmotic
content in any fluid compartment. Higher protein content in a
fluid compartment will change the osmotic gradient in favor
of water transport into the respective compartment vs. the
level present at that time in blood [61]. Indeed, recent studies
suggest that chronically increased CSF osmolality is sufficient
to produce HCP by creating an osmotic gradient which draws
water into the ventricles (presumably through aquaporins) [62,
63]. Regardless of what pathophysiological mechanism is correct or dominant, the association between high levels of protein in the CSF and the subsequent development of cHCP has
been repeatedly demonstrated.
For over half a century, cHCP is being treated successfully
by way of ventricular shunting [35]. On the other hand, high
CSF protein levels are considered a relative contraindication
for shunt placement [82]. This conflict will be discussed in the
following section.
CSF viscosity, CSF clotting, and fibrin Bsignature^
of tumors
BAmong the many changes which the cerebrospinal fluid may
undergo, certainly the most striking to the clinical observer is a
yellow coloration associated with the formation in the fluid,
soon after it is received into a test-tube, of a coagulum, which
may be so firm as to allow of the tube being turned upside
down without a drop of fluid escaping^ [Sir JC Greenfield,
1921].
The main objection for shunting hyperproteinorhachic patients has been that a high protein content in the CSF would
increase its viscosity and would thus impair its flow through
the catheter [82]. However, it has been shown that the mean
viscosity of CSF with a normal protein content is only 1.4%
greater than water at the same temperature, and there is little
increase in the mean viscosity of CSF specimens with increased protein content [22]. It means that in order to have
high protein levels resulting in blocking of a shunt catheter, it
probably will not be due to changes in viscosity. It therefore
must be due to specific proteins that promote coagulation and
adhesion and/or a strong inflammatory response.
Recurrent shunt obstructions can occur in the presence of a
highly viscous fluid or a coagulable fluid. However, these are
two different pathophysiological properties. Viscosity is the
one property of a fluid that will affect its flow through any
given length of tubing. Viscous CSF refers to CSF that flows
very slowly, dripping like glycerine from a tube. Truly viscous
CSF is extremely rare and, as mentioned, unrelated to absolute
levels of CSF protein. Fishman described two cases with truly
high CSF viscosity [44]: one case associated with large
amounts of mucin secreted by a metastatic mucinous adenocarcinoma originating from colon which presented with diffuse leptomeningeal carcinomatosis and a second case of
cryptococcal meningitis (in this case, high viscosity was attributed to the polysaccharide capsules of the yeasts). Viscous
CSF can also occur secondary to puncturing the annulus
fibrosus of a lumbar disc and the release of its liquid nucleus
pulposus [36]. However, these are all extremely rare
scenarios.
On the other hand, the observation of a CSF sample that
clots easily occurs more frequently. It occurs when sufficient
serum protein, including fibrinogen, is present in the CSF. The
conversion of fibrinogen to fibrin in the subarachnoid space
may induce arachnoiditis, fibrous adhesions, and eventually
obliteration of the subarachnoid spaces [96]. The same process will lead to obliteration of a ventricular catheter or any
other indwelling catheter located in a CSF compartment.
In healthy conditions, fibrinogen circulates through the
brain and the spinal cord vasculature without entering the
CNS parenchyma due to the elaborate architecture of the
BBB [30]. Several pathological conditions that involve either
acute hemorrhage, especially in the subarachnoid space (such
as brain or spinal cord injury, aneurysmal bleeding, and hemorrhagic stroke), or chronic disruption of the BBB (such as
multiple sclerosis, Alzheimer’s disease, brain glioblastoma,
HIV encephalitis, and bacterial meningitis) result in the deposition of fibrin/fibrinogen in the CNS [30, 44]. In regard to
tumors, acute or chronic subarachnoid hemorrhage or seeping
from tumor vessels is known as Fincher’s syndrome [43] and
has been described in various settings including in cases of VS
[10, 47, 50]. However, for the purpose of this review, we will
focus on non-hemorrhagic causes for increased CSF protein
levels and CSF clotting.
Of note, although non-hemorrhagic clotting of CSF is very
rare in the context of benign, intracranial neoplasms, deposition of fibrin in the stroma of tumor masses has long been
demonstrated. The Irish pathologist, R.A.Q. O’Meara, was
the first to propose in the late 1950s that fibrin was deposited
in solid tumors [81]. Further studies have shown that fibrin
provides a characteristic and reproducible signature for certain
types of tumors in much the same way as seen with other
histologic features [40]. Fibrin deposits must result from the
Neurosurg Rev
extravasation and extravascular clotting of plasma fibrinogen
because only a very few well-differentiated hepatocellular carcinomas can actually synthesize and secrete fibrinogen [40].
In tumors, as well as in other non-neoplastic pathologies, extravascular fibrin serves as a provisional matrix that favors and
supports the ingrowth of new blood vessels and other mesenchymal cells that generate mature, vascularized stroma [38].
The most important among these changes is the wellcoordinated inward migration of macrophages, fibroblasts,
and new blood vessels, followed by local synthesis and deposition of interstitial collagens and proteoglycans [40]. The extent of fibrin deposition and its persistence over time vary
extensively between different tumors. Apparently, these differences depend on quantitative balances among vascular
hyperpermeability, clotting, and fibrinolysis that are unique
to individual tumors [39].
Indeed, accumulation of fibrinogen and fibrin in the CSF
can be associated with increased level of protein in this media
in the absence of hemorrhage. Usually, this is the result of
hyperpermeability of the respective tumor vasculature and
was associated with CSF protein levels in excess of
1500 mg/dl [44]. Classically, this was observed in the presence of an intradural tumor in the lumbar region, a condition
known as Froin’s syndrome [13, 17, 79]. The association between Froin’s syndrome, high levels of CSF proteins, and
CSF clotting in relation to VS and SGCT is further discussed
in the following section.
Elevated CSF protein levels and hypercoagulable CSF
in VS and SGCT
Vestibular schwannomas (VSs or acoustic neuromas) are benign, Schwann cell-derived tumors that commonly arise from
the vestibular portion of the eighth cranial nerve. They are the
most common intracranial schwannomas and are also associated with NF2 [94]. The overall incidence of VS is about
1/100,000 person-years [89] whereas the incidence of NF2
is between 1/33,000 and 1/87,410 live births [94].
Stunningly, a correlation between VS, elevated levels of
CSF protein, and Froin’s syndrome can be dated back nearly
a century. In 1921, Sir Joseph Godwin Greenfield, Bthe father
of neuropathology,^ published his landmark review on Froin’s
syndrome [52]. Froin’s syndrome was also known as
Bsyndrome de coagulation massive et de xanthochromie^
and was originally (1903) described as hypercoagulable,
xantochromic CSF in the presence of chronic CNS infections
[45]. A few years later (1909), Blanchetiere and Lejonne were
the firsts to describe the full syndrome of Froin in association
with a spinal tumor (here Bsarcoma^) [52]. Over the years and
mainly due to the markedly decreased prevalence of chronic
spine infections such as leprosy and tuberculosis, Froin’s syndrome was mostly associated with the presence of a spinal
tumor blocking distal CSF flow [64, 67, 73]. Importantly, in
the same review, Greenfield mentioned that others have found
yellow and highly albuminous fluid in several cases of tumors
of the pontocerebellar angle [52]. Greenfield acknowledged
this observation by commenting that Bit is not impossible that
these changes might be present in the fluid of cases of solitary
eighth-nerve tumor, though I have not personally encountered
this^ [52]. Almost 100 years later, the association between
high CSF protein levels, cHCP, and VS has become a welldocumented phenomenon.
HCP occurs in 1.2 to 42% of VS patients [6, 11, 14, 19, 87,
91, 97, 101]. Due to its relative proximity to the fourth ventricle, large VS can cause obstructive HCP. Nevertheless, in
large published series, the majority (60–80%) of observed
HCP cases were of the communicating type [48, 91]. In other
series, the exact pathophysiology remained unclear and was
considered to be a combination of both types. For example, in
a study on 400 resected VSs, persistent HCP was found in up
to 22% of Bobstructive^ cases even after resection (compared
to 12.5% of Bcommunicating^ cases) suggesting a dual pathophysiology [48]. The presenting symptoms of HCP can either be increased intracranial pressure (ICP), which usually
indicates an obstructive pathology, or the persistence of a clinical triad typical of normal pressure hydrocephalus (NPH),
which may indicate a communicating pathology. In the series
by Pirouzmand et al., 90% (30/33) of patients with VS and
HCP clinically presented with NPH [87]. This observation
further supports the relatively high incidence of cHCP in VS
patients [18]. In our own series on 30 VSs treated by stereotactic radiosurgery, 4 were associated with cHCP (unpublished data). All four patients had signs and symptoms of
NPH, and two were treated with shunt placement. None of
these patients needed shunt revision over >5-year follow-up.
Figures 1 and 2 illustrate two of the cases.
In our two shunted patients, pretreatment CSF protein
levels were 114 and 193 mg/dl, respectively. In a number of
series describing an association between cHCP and VS, varying levels of CSF protein have been reported with an increase
in concentration ranging from 1.5- to 15-fold [6, 14, 91].
Other etiological factors considered relevant for the development of cHCP in VS patients are recurrent tumor bleeding,
seeding of tumor cells through the CSF, and an age-related
decrease in CSF reabsorption capacity [6]. Of note, the incidence of HCP in patients with small VSs (<3 cm) was 12-fold
higher in older patients than in younger ones (25 vs. 2.1%,
respectively) [97], which again may reflect a more complex
pathophysiology.
As mentioned earlier, in order to eliminate potential selection bias secondary to post surgical or post radiation changes,
we shall focus on the prevalence of elevated liquor protein
levels and cHCP in untreated patients. In a review by
Edwards et al., 73/75 (97%) of patients with untreated VSs
were found to have high CSF protein levels (>70 mg/dl) [41].
On the other hand, none of the larger series in which shunt
Neurosurg Rev
Fig. 1 A 63-year-old male presents with mild hearing loss on the left
side, found to have a 13 mm × 16 mm, homogenously enhancing lesion in
the left cerebellopontine angle (CPA), extending into the internal auditory
canal, consisting with vestibular schwannoma (a MRI; post-contrast T1wi, coronal cuts). Patient had no signs or symptoms compatible with
increased intracranial pressure or NPH. Ventricles’ sizes were only mildly
enlarged (b MRI, fluid attenuation inversion recovery (FLAIR) sequence). Patient underwent SRS treatment (hypofractionated,
2500 cGy) 1 year after diagnosis due to gradual tumor’s enlargement.
Approximately 8 months later, he started having NPH-related symptoms
(NPH triad). Repeat MRI showed very mild change in tumor’s size since
the time of diagnosis (c MRI; post-contrast T1-wi, coronal cuts); however, the ventricles were significantly larger (d MRI, FLAIR sequence),
with signs of transependymal flow (black arrows). The fourth ventricle
was not compressed (not shown), and images were compatible with communicating hydrocephalus. Unfortunately, patient was lost to follow-up
before placement of ventricular shunt could be discussed
placement was the primary surgical intervention for VS have
reported complications of recurrent shunt malfunction secondary to elevated CSF protein levels or CSF clotting [48, 87, 97].
In the study by Al Hinai et al., VS patients presenting with
symptomatic cHCP were treated with ventriculoperitoneal
shunt (VPS) placement despite CSF protein concentrations
as high as 900 mg/dl and no subsequent shunt obstruction
was reported [6]. In fact, although an association between
VS, high CSF protein levels, and HCP has been inferred for
nearly a century, we did not come across a single description
of recurrent shunt malfunctions secondary to increased CSF
protein levels or due to hypercoagulable CSF in this setting.
This is in sharp contradiction to the reports on managing
SGCT under similar circumstances.
Tuberous sclerosis complex (TSC) affects approximately
one million individuals worldwide, with a birth incidence of
approximately 1 in 6000 [66]. Subependymal giant cell astrocytoma (SGCT) is a benign (WHO grade I) and slow-growing
intraventricular tumor. It is almost always encountered in the
setting of TSC. Up to 20% of individuals with TSC will develop SGCT, usually during the first two decades of life [4],
making it the most frequent intracranial tumor type in TSC.
SGCTs are known to present with signs and symptoms of
increased ICP secondary to obstructive HCP due to its most
common location at the foramen of Monroi [32]. In addition
and similar to VS, most symptomatic cases are treated by
surgical resection of the tumor [32]. In a review on 84 pediatric patients, 22 (26%) shunt surgeries were reported [53].
Neurosurg Rev
Fig. 2 A 69-year-old female patient presented with tinnitus and
gait instability. She was found to
have a wide-based, shuffling gait.
Brain MRI demonstrated a
20 mm × 27 mm, homogenously
enhancing lesion in the left CPA,
extending into the internal auditory canal, consisting with vestibular schwannoma (a MRI,
post-contrast T1-wi, axial cuts).
All four ventricles were enlarged
substantially, consisting with
communicating hydrocephalus (b
MRI, FLAIR sequence; arrows
indicate transependymal flow).
Patient’s gait was improved after
a lumbar puncture trial, and a
permanent ventriculoperitoneal
shunt was placed. Three months
later, she was treated with
hypofractionated SRS (2500 cGy)
to the lesion. On a 5-year followup, both the tumor and the ventricles are smaller and well controlled (c MRI, post-contrast T1wi, axial cuts. d MRI, T2-wi, axial cuts; arrow points at the ventricular catheter)
Comparable to VS, 86 % (19/22) of these shunts were inserted
after tumor resection [53]. Nevertheless, reports on
presurgical elevated protein content in the CSF in this setting
and an association with cHCP exist. In a case series by Di
Rocco and colleagues, all ten pediatric patients with tuberous
sclerosis and SGCT were found to have high CSF protein
levels (mean 600 mg/dl, range 270–1500 mg/dl) [33]. Three
out of those ten patients were treated with upfront VPS placement, and two out of those three suffered repeat episodes of
obstruction of the shunt system. Of note, the third patient who
has not had VPS malfunction also had the lowest CSF protein
level of the entire cohort (270 mg/dl) [33]. Since this report
was published, others have encountered similar problems of
recurrent shunt malfunctioning in the presence of protein-rich,
highly clotting CSF [66, 85] (Fig. 3).
Reported CSF protein levels were between 945 and
2250 mg/dl [66], and in all of these cases, all four ventricles were enlarged, consistent with cHCP. This phenomenon of increased CSF coagulability is a unique feature of
SGCT, making it an Bintracranial variant^ of the classic
observation made in Froin’s syndrome. One of the pathophysiological prerequisites for Froin’s syndrome to occur
is that spinal CSF must be contained in the lumbar Bculde-sac,^ which was shut off from communicating with
fluid around the upper part of the cord (the mechanism
being due to meningitis, tumor, or disease of the bones of
the spine) [52]. Similarly, the predilection of SGCT to
occur near the foramen of Monroi can cause the frontal
horn of the lateral ventricle to become an isolated
Bintracranial cul-de-sac,^ mimicking the pathological conditions of Froin’s syndrome. It is hard to decide whether
or not this is part of the pathogenesis of CSF clotting
observed in SGCT. However, it needs to be pointed out
that (a) no other known obstructing intraventricular lesions are associated with this phenomena and (b) this
was also reported in association with SGCTs not located
at the foramen of Monroi [66]. It is noteworthy that this
phenomenon has never been described in association with
VS or other benign intraventricular neoplasms.
Both angiogenesis and inflammation are characterized by
increased vascular permeability, leading to extravasation of
proteins from the serum [72, 77]. The presence of these two
pathophysiologic mechanisms in VS and SGCT may contribute to the mentioned phenomena in these tumors.
Neurosurg Rev
Fig. 3 Hypercoagulable CSF as was observed at the time of
ventriculoperitoneal shunt revision in TSC patient with SGCT
Angiogenesis, inflammation, and vascular
hyperpermeability in phacomatoses
BAs a rule the tumors are sparsely vascularized but in some
cases the vessels in certain areas are sufficiently numerous to
give to the tumor an angiomatous appearance.^ [H. Cushing,
Tumors of the nervus acusticus and the syndrome of the
cerebellopontine angle, 1917].
Most solid tumors, regardless of their type and origin, cannot grow beyond a certain size (~11 mm3) unless they establish a blood supply by inducing new vessels sprouting from
existing host capillaries in a process known as angiogenesis
[65]. In tumors, the process of angiogenesis is strongly correlated with increased vascular permeability [77]. Fluid leakage
in tumors takes place in such newly formed, highly abnormal
blood vessels post angiogenesis [76]. These neovessels usually display greatly enlarged sinusoids that arise from
preexisting normal venules by a process that involves pericyte
detachment, vascular basal lamina degradation, and a 4–5-fold
increase in lumen size that is accompanied by extensive endothelial cell thinning [76]. Microvascular extravasation has several clinical consequences, such as the formation of vasogenic
brain edema and the formation of tumor-related cysts [12]. It is
also a major determinant of radiographic appearance such as
prominent and diffuse contrast enhancement of the tumor
[27].
Interestingly, significant angiogenesis is a unifying feature
of most familial hamartoma syndromes [9, 21]. In this regard,
von Hippel–Lindau disease (vHL) is perhaps the most representative phacomatosis of all. The most common CNS tumor
associated with vHL is hemangioblastoma (HGB), benign
vascular tumors composed of tightly packed capillaries and
neoplastic stromal cells. Besides the obvious, highly
angiogenetic radiological and histological properties of these
tumors, they are characterized by a very high rate (ca. 80%) of
observed peritumoral edema and cysts, which are two characteristic features of increased vascular permeability [12, 49,
69]. Of note, nearly 30% of intracranial HGBs are also associated with HCP [57] though this is usually of the obstructive
type, secondary to their most frequent location in the cerebellum with resulting compression of the fourth ventricle. These
cases of HCP are usually resolved by surgical resection of the
tumor, and shunt placement in vHL patients occurs rarely. In a
large (n = 164) series on the management of cerebellar HGB in
vHL patients, Jagannathan et al. described only two patients
requiring VP shunt placement (one with craniospinal
hemangioblastomatosis and one who developed renal cell carcinoma with metastases to the cerebellum and leptomeningeal
carcinomatosis) [57]. In a pertinent review of the literature, we
have found that VPS placement for patients with HGB is
usually associated with the rare scenario of disseminated cases
[5]. As expected from disseminated neoplasms, some of these
cases were reported to have increased CSF protein levels [90,
102]. However, we could not find a single reported case describing an association between solitary HGB and high levels
of protein in the CSF, clotting of the CSF, recurrent shunt
malfunctions, or communicating HCP. We think that the reason for that is the intraaxial (i.e., intraparenchymal) location of
HGB, as opposed to the intracisternal and intraventricular location of VS and SGCT, respectively. Therefore, despite being
exemplary as a phacomatosis tumor with increased angiogenesis and hyperpermeability, we shall focus in this review on
VS and SGCT only.
The increased vascular permeability of schwannomas has
been known for decades. Studies from the early 1970s have
demonstrated that in human VS, numerous vessels are open to
the extracellular space by gaps between capillary endothelial
cells, lack of junctional substructure, and fenestrated membranes [68]. In addition, cyst formation as an expression of
vascular hyperpermeability is frequently observed in VS [86].
Histologically, VSs are often characterized by cystic or fatty
degeneration (aka degenerative changes or Antoni B morphology), focal accumulations of hyaline material around vessels,
infiltration of histiocytes and siderophages, and
hypercellularity (= Antoni A morphology) [8]. Although the
occurrence of BAntoni B^ patterns was initially thought to be
related to slow growth of schwannomas [28], such changes
were not observed in slow-growing meningiomas or neurofibromas [84]. Vilanova et al. pointed out that the degree of
degenerative changes correlates with tumor size and vascular
abnormalities in the tumor [100]. Papiez et al. have shown that
these degenerative changes are indeed related to the proliferation of vessels observed in schwannomas, along with functional deficits in the wall integrity causing increased permeability to proteins and blood. Interestingly, these vascular
Neurosurg Rev
changes were significantly more prominent in Antoni B areas
and areas with such degeneration were not demonstrated in
benign meningiomas or neurofibromas [84]. Most VSs contain numerous scattered vessels, often with thin walls. These
vessels may cause bleeds within the tumor, and the resulting
pigment deposit of hemosiderin in both macrophages and parenchyma is a characteristic feature of schwannoma [60]. Due
to their unique vascular texture, schwannomas have been conceptualized as structurally analogous to benign mixed mesenchymal tumors such as angiomyoma, angiolipoma, and angiomatous fibrohistiocytoma [60]. Strikingly, these are also tumors of the TSC [9].
Many, if not all, solid tumors contain inflammatory features
[26]. The effects of inflammation on vascular permeability
have been known for decades [103]. Under normal physiological conditions, the peripheral vasculature is not permeable to
macromolecules. However, in the presence of inflammation,
potent mediators (such as histamine, bradykinins, serotonin,
and prostaglandins) cause a significant increase in microvascular permeability of the peripheral vasculature [92]. The normal cerebral blood vessels appear mostly unable to respond to
these potent mediators of microvascular extravasation [27],
although it has been shown that brain endothelium permeability increases in response to as little as nanomolar concentrations of these agents [1]. Brain tumor vasculature loses BBB
properties during angiogenesis, causing their vessels to function similarly to peripheral vasculature. Therefore, in certain
cases, brain tumor vasculature can become responsive to mediators of vascular permeability associated with inflammation
[80]. The interplay of angiogenesis, increased vascular permeability, and inflammatory responses is of special importance in
TSC. Its associated tumors are highly vascularized [34] and
inflammatory markers, such as cytokines and chemokines,
have been found in specimen from patients with TSC [15,
16, 70, 93]. The characteristic histopathological feature of
SGCT is a rich vascular stroma [51] with a substantial component of inflammatory cells, including mast cells and T lymphocytes [93]. In part II of this review, we discuss the close
resemblance between pathways of inflammation and coagulation. We suggest that the predominance of the former in SGCT
might explain the unique association between SGCT and the
propensity of clotting of CSF in afflicted patients.
Both NF2 and TSC phacomatoses demonstrate abnormal
activation of the Pi3K/Akt/mTOR pathway (see part II, BNF2,
TSC, and the Pi3K/Akt/mTOR pathway^ section). This pathway is involved in angiogenesis [59] as well as in inflammation [95]. In the second part of this review, we describe both
aspects of this pathway in detail and emphasize their involvement in SGCT and VS at the cellular and molecular levels.
Correlation between these two mechanisms and two types of
tumor-associated macrophages is given as an explanation for
the phenotypical differences between the two tumors. Finally,
therapeutic implications of these differences are discussed.
Summary
VSs and SGCTs represent two of the most common intracranial tumors related to phacomatoses, recognized as NF2 and
TSC syndromes, respectively. Both VS and SGCT are known
to cause obstructive HCP due to their critical anatomical locations next to or inside the ventricles. Remarkably, both tumors are also uniquely known to be associated with elevated
concentrations of protein levels in the CSF, sometimes in association with communicating HCP. Furthermore, SGCTs
have been reported to be associated with CSF clotting, causing
debilitating clinical courses due to recurrent shunt obstructions. This phenomenon of CSF clotting in the presence of
small, benign tumor had been reported under the label of
BFroin’s syndrome^ in the past when observed in association
with spinal tumors. Thus, SGCT can be seen as representing a
unique, intracranial variant of Froin’s syndrome. Elevated
levels of CSF proteins and CSF clotting are caused by increased vascular permeability and dysfunction of the BBB.
The two main pathophysiologic mechanisms responsible for
that in association with tumorigenesis are angiogenesis and
inflammation. In this review, we were able to demonstrate that
these two mechanisms are actively present in VSs and SGCTs
at the phenotypic level. In part II of the review, we describe
these two pathological processes at the cellular and molecular
levels and discuss their therapeutic implications.
Compliance with ethical standards
Conflict of interest The authors declare that they have no conflict of
interest.
Funding This work was not funded or financially supported.
Informed consent Informed consent was obtained from all individual
participants included in the study.
Ethical approval For this type of study, formal consent is not required.
References
1.
2.
3.
4.
Abbott NJ (2000) Inflammatory mediators and modulation of
blood-brain barrier permeability. Cell Mol Neurobiol 20:131–147
Adams RA, Passino M, Sachs BD, Nuriel T, Akassoglou K (2004)
Fibrin mechanisms and functions in nervous system pathology.
Mol Interv 4:163–176
Adkins JN, Varnum SM, Auberry KJ, Moore RJ, Angell NH,
Smith RD, Springer DL, Pounds JG (2002) Toward a human
blood serum proteome: analysis by multidimensional separation
coupled with mass spectrometry. Mol Cell Proteomics 1:947–955
Adriaensen ME, Schaefer-Prokop CM, Stijnen T, Duyndam DA,
Zonnenberg BA, Prokop M (2009) Prevalence of subependymal
giant cell tumors in patients with tuberous sclerosis and a review
of the literature. Eur J Neurol 16:691–696. doi:10.1111/j.14681331.2009.02567.x
Neurosurg Rev
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Akimoto J, Fukuhara H, Suda T, Nagai K, Hashimoto R,
Michihiro K (2014) Disseminated cerebellar hemangioblastoma
in two patients without von Hippel-Lindau disease. Surg Neurol
Int 5:2152–7806
Al Hinai Q, Zeitouni A, Sirhan D, Sinclair D, Melancon D,
Richardson J, Leblanc R (2013) Communicating hydrocephalus
and vestibular schwannomas: etiology, treatment, and long-term
follow-up. J Neurol Surg B Skull Base 74:68–74. doi:10.1055/s0033-1333621
Amlashi SF, Riffaud L, Morandi X (2006) Communicating hydrocephalus and papilloedema associated with intraspinal tumours:
report of four cases and review of the mechanisms. Acta Neurol
Belg 106:31–36
Antonescu CR SB, Woodruff JM (2013) Schwannoma. Tumors of
the peripheral nervous system, AFIP atlas of tumor pathology
series 4. American Registry of Pathology, City, pp 129–210
Arbiser JL, Brat D, Hunter S, D’Armiento J, Henske EP, Arbiser
ZK, Bai X, Goldberg G, Cohen C, Weiss SW (2002) Tuberous
sclerosis-associated lesions of the kidney, brain, and skin are angiogenic neoplasms. J Am Acad Dermatol 46:376–380
Arienta C, Caroli M, Crotti FM (1988) Subarachnoid haemorrhage due to acoustic neurinoma. Case report and review of the
literature. Neurochirurgia (Stuttg) 31:162–165
Atlas MD, Perez de Tagle JR, Cook JA, Sheehy JP, Fagan PA
(1996) Evolution of the management of hydrocephalus associated
with acoustic neuroma. Laryngoscope 106:204–206
Baggenstos MA, Butman JA, Oldfield EH, Lonser RR (2007)
Role of edema in peritumoral cyst formation. Neurosurg Focus
22:E9
Bergquist BJ (1988) Intraspinal tumor with hydrocephalus.
Neurosurgery 22:969–970
Bloch J, Vernet O, Aube M, Villemure JG (2003) Non-obstructive
hydrocephalus associated with intracranial schwannomas:
hyperproteinorrhachia as an etiopathological factor? Acta
Neurochir 145:73–78
Boer K, Crino PB, Gorter JA, Nellist M, Jansen FE, Spliet WG,
van Rijen PC, Wittink FR, Breit TM, Troost D et al (2010) Gene
expression analysis of tuberous sclerosis complex cortical tubers
reveals increased expression of adhesion and inflammatory factors. Brain Pathol 20:704–719. doi:10.1111/j.1750-3639.2009.
00341.x
Boer K, Jansen F, Nellist M, Redeker S, van den Ouweland AM,
Spliet WG, van Nieuwenhuizen O, Troost D, Crino PB, Aronica E
(2008) Inflammatory processes in cortical tubers and
subependymal giant cell tumors of tuberous sclerosis complex.
Epilepsy Res 78:7–21.
Borgesen SE, Sorensen SC, Olesen J, Gjerris F (1977) Spinal
tumours associated with increased intracranial pressure. Report
of two cases and a discussion on the pathophysiology. Acta
Neurol Scand 56:263–268
Bradley WG (2000) Normal pressure hydrocephalus: new concepts on etiology and diagnosis. AJNR Am J Neuroradiol 21:
1586–1590
Briggs RJ, Shelton C, Kwartler JA, Hitselberger W (1993)
Management of hydrocephalus resulting from acoustic neuromas.
Otolaryngol Head Neck Surg 109:1020–1024
Brightman MW, Klatzo I, Olsson Y, Reese TS (1970) The bloodbrain barrier to proteins under normal and pathological conditions.
J Neurol Sci 10:215–239
Brugarolas J, Kaelin WG Jr (2004) Dysregulation of HIF and
VEGF is a unifying feature of the familial hamartoma syndromes.
Cancer Cell 6:7–10
Brydon HL, Hayward R, Harkness W, Bayston R (1995) Physical
properties of cerebrospinal fluid of relevance to shunt function. 1:
the effect of protein upon CSF viscosity. Br J Neurosurg 9:639–
644
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
Celli P, Cervoni L, Morselli E, Ferrante L (1993) Spinal
ependymomas and papilledema: report of 4 cases and review of
the literature. J Neurosurg Sci 37:97–102
Claesson-Welsh L (2015) Vascular permeability—the essentials.
Ups J Med Sci 120:135–143. doi:10.3109/03009734.2015.
1064501
Claesson-Welsh L, Welsh M (2013) VEGFA and tumour angiogenesis. J Intern Med 273:114–127. doi:10.1111/joim.12019
Colotta F, Allavena P, Sica A, Garlanda C, Mantovani A (2009)
Cancer-related inflammation, the seventh hallmark of cancer: links
to genetic instability. Carcinogenesis 30:1073–1081. doi:10.1093/
carcin/bgp127
Criscuolo GR (1993) The genesis of peritumoral vasogenic brain
edema and tumor cysts: a hypothetical role for tumor-derived vascular permeability factor. Yale J Biol Med 66:277–314
Dahl I (1977) Ancient neurilemmoma (schwannoma). Acta Pathol
Microbiol Scand A 85:812–818
Daras M, Kaley TJ (2015) Benign brain tumors and tumors associated with phakomatoses. Continuum (Minneap Minn) 21:397–
414. doi:10.1212/01.CON.0000464177.73440.44
Davalos D, Akassoglou K (2012) Fibrinogen as a key regulator of
inflammation in disease. Semin Immunopathol 34:43–62. doi:10.
1007/s00281-00011-00290-00288
de la Fuente MI, DeAngelis LM (2014) The role of
ventriculoperitoneal shunting in patients with supratentorial glioma. Ann Clin Transl Neurol 1:45–48. doi:10.1002/acn1003.1017
de Ribaupierre S, Dorfmuller G, Bulteau C, Fohlen M, Pinard JM,
Chiron C, Delalande O (2007) Subependymal giant-cell astrocytomas in pediatric tuberous sclerosis disease: when should we
operate? Neurosurgery 60:83–89 discussion 89-90
Di Rocco C, Iannelli A, Marchese E (1995) On the treatment of
subependymal giant cell astrocytomas and associated hydrocephalus in tuberous sclerosis. Pediatr Neurosurg 23:115–121
Dodd KM, Yang J, Shen MH, Sampson JR, Tee AR (2015)
mTORC1 drives HIF-1alpha and VEGF-A signalling via multiple
mechanisms involving 4E-BP1, S6K1 and STAT3. Oncogene 34:
2239–2250. doi:10.1038/onc.2014.164
Drake JM, Kestle JR, Tuli S (2000) CSF shunts 50 years on—past,
present and future. Childs Nerv Syst 16:800–804
Dripps RD, Vandam LD (1951) Hazards of lumbar puncture. J Am
Med Assoc 147:1118–1121
Duran WN, Breslin JW, Sanchez FA (2010) The NO cascade,
eNOS location, and microvascular permeability. Cardiovasc Res
87:254–261. doi:10.1093/cvr/cvq139
Dvorak HF, Brown LF, Detmar M, Dvorak AM (1995) Vascular
permeability factor/vascular endothelial growth factor, microvascular hyperpermeability, and angiogenesis. Am J Pathol 146:
1029–1039
Dvorak HF, Harvey VS, McDonagh J (1984) Quantitation of fibrinogen influx and fibrin deposition and turnover in line 1 and
line 10 guinea pig carcinomas. Cancer Res 44:3348–3354
Dvorak HF, Nagy JA, Berse B, Brown LF, Yeo KT, Yeo TK,
Dvorak AM, van de Water L, Sioussat TM, Senger DR (1992)
Vascular permeability factor, fibrin, and the pathogenesis of tumor
stroma formation. Ann N Y Acad Sci 667:101–111
Edwards CH, Paterson JH (1951) A review of the symptoms and
signs of acoustic neurofibromata. Brain 74:144–190
Felgenhauer K (1974) Protein size and cerebrospinal fluid composition. Klin Wochenschr 52:1158–1164
Fincher EF (1951) Spontaneous subarachnoid hemorrhage in
intradural tumors of the lumbar sac; a clinical syndrome. J
Neurosurg 8:576–584
Fishman RA (1980) Cerebrospinal fluid in disease of the nervous
system. W.B Saunders City
Neurosurg Rev
F G (1903) Inflammations me ‘ninge’es avec chromatique,
fibrineuse et cytologique du liquide ce’phalo-rachidien. Gazette
des hoˆpitaux 76:1005–1006
46. Gardner WJ, Spitler DK, Whitten C (1954) Increased intracranial
pressure caused by increased protein content in the cerebrospinal
fluid; an explanation of papilledema in certain cases of small intracranial and intraspinal tumors, and in the Guillain-Barre syndrome. N Engl J Med 250:932–936
47. Gavra M, Thanos L, Pomoni M, Batakis N (2010) Spontaneous
subarachnoid haemorrhage due to acoustic neurinoma. Case report
and review of the literature. Br J Neurosurg 24:82–83. doi:10.
3109/02688690903506085
48. Gerganov VM, Pirayesh A, Nouri M, Hore N, Luedemann WO,
Oi S, Samii A, Samii M (2011) Hydrocephalus associated with
vestibular schwannomas: management options and factors
predicting the outcome. J Neurosurg 114:1209–1215. doi:10.
3171/2010.10.JNS1029
49. Glasker S, Vortmeyer AO, Lonser RR, Lubensky IA, Okamoto H,
Xia JB, Li J, Milne E, Kowalak JA, Oldfield EH et al (2006)
Proteomic analysis of hemangioblastoma cyst fluid. Cancer Biol
Ther 5:549–553
50. Gleeson RK, Butzer JF, Grin OD Jr (1978) Acoustic neurinoma
presenting as subarachnoid hemorrhage. Case report. J Neurosurg
49:602–604
51. Grajkowska W, Kotulska K, Jurkiewicz E, Matyja E (2010) Brain
lesions in tuberous sclerosis complex. Review. Folia Neuropathol
48:139–149
52. Greenfield JG (1921) Original papers: on Froin’s syndrome, and
its relation to allied conditions in the cerebrospinal fluid. J Neurol
Psychopathol 2:105–141
53. Harter DH, Bassani L, Rodgers SD, Roth J, Devinsky O, Carlson
C, Wisoff JH, Weiner HL (2014) A management strategy for intraventricular subependymal giant cell astrocytomas in tuberous
sclerosis complex. J Neurosurg Pediatr 13:21–28. doi:10.3171/
2013.3179.PEDS13193
54. Hladky SB, Barrand MA (2014) Mechanisms of fluid movement
into, through and out of the brain: evaluation of the evidence.
Fluids Barriers CNS 11:26. doi:10.1186/2045-8118-1111-1126
eCollection 2014
55. Inamasu J, Nakamura Y, Saito R, Kuroshima Y, Mayanagi K, Orii
M, Ichikizaki K (2003) Postoperative communicating hydrocephalus in patients with supratentorial malignant glioma. Clin Neurol
Neurosurg 106:9–15
56. Irani DN (2008) Cerebrospinal fluid in clinical practice. Saunders,
Elsevier Health Sciences, City
57. Jagannathan J, Lonser RR, Smith R, DeVroom HL, Oldfield EH
(2008) Surgical management of cerebellar hemangioblastomas in
patients with von Hippel-Lindau disease. J Neurosurg 108:210–
222. doi:10.3171/JNS/2008/108/2/0210
58. Jung TY, Chung WK, Oh IJ (2014) The prognostic significance of
surgically treated hydrocephalus in leptomeningeal metastases.
Clin Neurol Neurosurg 119:80–83
59. Karar J, Maity A (2011) PI3K/AKT/mTOR pathway in angiogenesis. Front Mol Neurosci 4
60. Kasantikul V, Netsky MG (1979) Combined neurilemmoma and
angioma. Tumor of ectomesenchyme and a source of bleeding. J
Neurosurg 50:81–87
61. Krishnamurthy S, Li J (2014) New concepts in the pathogenesis of
hydrocephalus. Transl Pediatr 3:185–194. doi:10.3978/j.issn.
2224-4336.2014.07.02
62. Krishnamurthy S, Li J, Schultz L, Jenrow KA (2012) Increased
CSF osmolarity reversibly induces hydrocephalus in the normal
rat brain. Fluids Barriers CNS 9:13. doi:10.1186/2045-8118-11891113
63. Krishnamurthy S, Li J, Schultz L, McAllister JP 2nd (2009)
Intraventricular infusion of hyperosmolar dextran induces
45.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
74.
75.
76.
77.
78.
79.
80.
81.
hydrocephalus: a novel animal model of hydrocephalus.
Cerebrospinal Fluid Res 6:1743–8454
Kwon SK, Kim MW (2014) Pseudo-Froin’s syndrome,
xanthochromia with high protein level of cerebrospinal fluid.
Korean J Anesthesiol 67:S58–S59. doi:10.4097/kjae.2014.4067.
S.S4058
Lane HA, Wood JM, McSheehy PM, Allegrini PR, Boulay A,
Brueggen J, Littlewood-Evans A, Maira SM, Martiny-Baron G,
Schnell CR et al (2009) mTOR inhibitor RAD001 (everolimus)
has antiangiogenic/vascular properties distinct from a VEGFR tyrosine kinase inhibitor. Clin Cancer Res 15:1612–1622. doi:10.
1158/1078-0432.CCR-08-2057
Laviv Y, Jackson S, Rappaport ZH (2015) Persistent communicating hydrocephalus in adult tuberous sclerosis patients: a possible
therapeutic role for everolimus. Acta Neurochir 157:241–245. doi:
10.1007/s00701-014-2309-0
Ljevak J, Poljakovic Z, Adamec I, Habek M (2014) Glioblastoma
multiforme presenting as Froin’s syndrome: a new face of an old
foe. Acta Neurol Belg 114:319–320. doi:10.1007/s13760-0140289-8
Long DM (1973) Vascular ultrastructure in human meningiomas
and schwannomas. J Neurosurg 38:409–419
Lonser RR, Vortmeyer AO, Butman JA, Glasker S, Finn MA,
Ammerman JM, Merrill MJ, Edwards NA, Zhuang Z, Oldfield
EH (2005) Edema is a precursor to central nervous system
peritumoral cyst formation. Ann Neurol 58:392–399
Maldonado M, Baybis M, Newman D, Kolson DL, Chen W,
McKhann G 2nd, Gutmann DH, Crino PB (2003) Expression of
ICAM-1, TNF-alpha, NF kappa B, and MAP kinase in tubers of
the tuberous sclerosis complex. Neurobiol Dis 14:279–290
Merritt H.H. F-SF (1937) The cerebrospinal fluid. W. B. Saunders
Co., City
Michel CC, Curry FE (1999) Microvascular permeability. Physiol
Rev 79:703–761
Mirza S, Adams WM, Corkhill RA (2008) Froin’s syndrome
revisited, 100 years on. Pseudo-Froin’s syndrome on MRI. Clin
Radiol 63:600–604. doi:10.1016/j.crad.2007.07.027
Montano N, D’Alessandris QG, Bianchi F, Lauretti L, Doglietto F,
Fernandez E, Maira G, Pallini R (2011) Communicating hydrocephalus following surgery and adjuvant radiochemotherapy for
glioblastoma. J Neurosurg 115:1126–1130. doi:10.3171/2011.8.
JNS11738
Murray KJ, Ausman JI, Chou SN, Douglas SD (1977)
Immunoproteins in human brain tumor cyst fluids. J Neurosurg
46:314–319
Nagy JA, Benjamin L, Zeng H, Dvorak AM, Dvorak HF (2008)
Vascular permeability, vascular hyperpermeability and angiogenesis. Angiogenesis 11:109–119. doi:10.1007/s10456-10008-19099-z
Nagy JA, Dvorak AM, Dvorak HF (2012) Vascular
hyperpermeability, angiogenesis, and stroma generation. Cold
Spring Harb Perspect Med 2:a006544. doi:10.1101/cshperspect.
a006544
Nigim F, Critchlow JF, Kasper EM (2015) Role of
ventriculoperitoneal shunting in patients with neoplasms of the
central nervous system: an analysis of 59 cases. Mol Clin Oncol
3:1381–1386
Nishida K, Ueda S, Matsumoto K, Kusaka K, Takeuchi R (1990)
Cauda equina neurinoma associated with normal pressure hydrocephalus—case report. Neurol Med Chir (Tokyo) 30:258–262
Nomura T, Ikezaki K, Natori Y, Fukui M (1993) Altered response
to histamine in brain tumor vessels: the selective increase of regional cerebral blood flow in transplanted rat brain tumor. J
Neurosurg 79:722–728
O’Meara RA (1958) Coagulative properties of cancers. Ir J Med
Sci 394:474–479
Neurosurg Rev
82.
83.
84.
85.
86.
87.
88.
89.
90.
91.
92.
Occhipinti ECC (1982) Shunt failure in hydrocephalus with high
protein fluid. Monog Neural Sci 8:220–222
Omuro AM, Lallana EC, Bilsky MH, DeAngelis LM (2005)
Ventriculoperitoneal shunt in patients with leptomeningeal metastasis. Neurology 64:1625–1627
Papiez J, Rojiani MV, Rojiani AM (2014) Vascular alterations in
schwannoma. Int J Clin Exp Pathol 7:4032–4038 eCollection
2014
Perek-Polnik M, Jozwiak S, Jurkiewicz E, Perek D, Kotulska K
(2012) Effective everolimus treatment of inoperable, lifethreatening subependymal giant cell astrocytoma and intractable
epilepsy in a patient with tuberous sclerosis complex. Eur J
Paediatr Neurol 16:83–85. doi:10.1016/j.ejpn.2011.1009.1006
Piccirillo E, Wiet MR, Flanagan S, Dispenza F, Giannuzzi A,
Mancini F, Sanna M (2009) Cystic vestibular schwannoma: classification, management, and facial nerve outcomes. Otol Neurotol
30:826–834. doi:10.1097/MAO.0b013e3181b04e18
Pirouzmand F, Tator CH, Rutka J (2001) Management of hydrocephalus associated with vestibular schwannoma and other
cerebellopontine angle tumors. Neurosurgery 48:1246–1253 discussion 1253-1244
Prockop LD, Fishman RA (1968) Experimental pneumococcal
meningitis. Permeability changes influencing the concentration
of sugars and macromolecules in cerebrospinal fluid. Arch
Neurol 19:449–463
Propp JM, McCarthy BJ, Davis FG, Preston-Martin S (2006)
Descriptive epidemiology of vestibular schwannomas. NeuroOncology 8:1–11
Rogers LR, LoRusso P, Nadler P, Malik G, Shields A, Kaelin W
(2011) Erlotinib therapy for central nervous system
hemangioblastomatosis associated with von Hippel-Lindau disease: a case report. J Neuro-Oncol 101:307–310. doi:10.1007/
s11060-010-0244-3
Rogg JM, Ahn SH, Tung GA, Reinert SE, Noren G (2005)
Prevalence of hydrocephalus in 157 patients with vestibular
schwannoma. Neuroradiology 47:344–351
Scallan J HV, Korthuis RJ. (2010) Capillary fluid exchange: regulation, functions, and pathology. Morgan & Claypool Life
Sciences, City
93.
Sharma M, Ralte A, Arora R, Santosh V, Shankar SK, Sarkar C
(2004) Subependymal giant cell astrocytoma: a clinicopathological study of 23 cases with special emphasis on proliferative
markers and expression of p53 and retinoblastoma gene proteins.
Pathology 36:139–144
94. Slattery WH (2015) Neurofibromatosis type 2. Otolaryngol Clin N
Am 48:443–460. doi:10.1016/j.otc.2015.02.005
95. Stark AK, Sriskantharajah S, Hessel EM, Okkenhaug K (2015)
PI3K inhibitors in inflammation, autoimmunity and cancer. Curr
Opin Pharmacol 23:82–91
96. Sullivan RL Jr, Reeves AG (1977) Normal cerebrospinal fluid
protein, increased intracranial pressure, and the Guillain-Barre
syndrome. Ann Neurol 1:108–109
97. Tanaka Y, Kobayashi S, Hongo K, Tada T, Sato A, Takasuna H
(2003) Clinical and neuroimaging characteristics of hydrocephalus associated with vestibular schwannoma. J Neurosurg 98:1188–
1193
98. Tucker M, Goldstein A, Dean M, Knudson A (2000) National
Cancer Institute Workshop Report: the phakomatoses revisited. J
Natl Cancer Inst 92:530–533
99. Van der Hoeve J (1923) Eye diseases in tuberous sclerosis of the
brain and in Recklinghausen’s disease Tran Ophthalmol Soc UK
43: 534–541
100. Vilanova JR, Burgos-Bretones JJ, Alvarez JA, Rivera-Pomar JM
(1982) Benign schwannomas: a histopathological and morphometric study. J Pathol 137:281–286
101. Wada K, Nawashiro H, Shimizu A, Shima K (2003) MRI analysis
of hydrocephalus associated with acoustic neurinoma. Acta
Neurochir Suppl 86:549–551
102. Weil RJ, Vortmeyer AO, Zhuang Z, Pack SD, Theodore N,
Erickson RK, Oldfield EH (2002) Clinical and molecular analysis
of disseminated hemangioblastomatosis of the central nervous
system in patients without von Hippel-Lindau disease. Report of
four cases J Neurosurg 96:775–787
103. Wilhelm DL (1973) Mechanisms responsible for increased vascular permeability in acute inflammation. Agents Actions 3:297–306