Content uploaded by Jiří Lisy
Author content
All content in this area was uploaded by Jiří Lisy on Jul 12, 2019
Content may be subject to copyright.
Original article
Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis
in children with neurofibromatosis type 1
Marie Glombova
a,b,⇑
, Borivoj Petrak
a
, Jiri Lisy
c
, Josef Zamecnik
d
,
David Sumerauer
e
, Petr Liby
f
a
Department of Paediatric Neurology, Second Faculty of Medicine, Charles University in Prague and Motol University Hospital, Prague,
Czech Republic
b
Paediatric Department, District Hospital Kolin, Czech Republic
c
Department of Imaging Methods, Second Faculty of Medicine, Charles University in Prague and Motol University Hospital, Prague, Czech Republic
d
Department of Pathology and Molecular Medicine, Second Faculty of Medicine, Charles University in Prague and Motol University Hospital,
Prague, Czech Republic
e
Department of Haemato-oncology, Second Faculty of Medicine, Charles University in Prague and Motol University Hospital, Prague, Czech Republic
f
Department of Neurosurgery, Second Faculty of Medicine, Charles University in Prague and Motol University Hospital, Prague, Czech Republic
Received 29 September 2018; received in revised form 28 March 2019; accepted 2 April 2019
Abstract
Purpose: To evaluate the incidence and clinical importance of brain gliomas – optic pathway gliomas (OPGs) and especially glio-
mas outside the optic pathway (GOOP) for children with neurofibromatosis type 1 (NF1), additionally, to assess the causes of
obstructive hydrocephalus in NF1 children with an emphasis on cases caused by idiopathic aqueduct stenosis.
Subjects and methods: We analysed data from 285 NF1 children followed up on our department from 1990 to 2010 by the same
examination battery.
Results: We have found OPGs in 77/285 (27%) children and GOOPs in 29/285 (10,2%) of NF1 children, of who 19 had OPG and
GOOP together, so the total number of brain glioma was 87/285 (30,5%). GOOPs were significantly more often treated than OPGs
(p > 0.01). OPGs contain clinically important subgroup of 14/285 (4.9%) spreading to hypothalamus. Spontaneous regression was
documented in 4/285 (1.4%) gliomas and the same number of NF1 children died due to gliomas.
Obstructive hydrocephalus was found in 22/285 (7.7%) patients and 14/22 cases were due to glioma. Idiopathic aqueduct stenosis
caused hydrocephalus in 6/22 cases and was found in 2.1% of NF1 children. Two had other cause.
Conclusions: The total brain glioma number (OPGs and only GOOPs together) better reflected the overall brain tumour risk for
NF1 children. However, GOOPs occur less frequently than OPGs, they are more clinically relevant. The obstructive hydrocephalus
was severe and featuring frequent complication, especially those with GOOP. Idiopathic aqueduct stenosis shows an unpredictable
cause of hydrocephalus in comparison with glioma and is another reason for careful neurologic follow up.
Ó2019 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved.
Keywords: Neurofibromatosis type 1; Optic pathway glioma; Brain glioma; Hydrocephalus; Idiopathic aqueduct stenosis
https://doi.org/10.1016/j.braindev.2019.04.003
0387-7604/Ó2019 The Japanese Society of Child Neurology. Published by Elsevier B.V. All rights reserved.
Abbreviations: D2+H, Dodge 2 with hypothalamus involvement; ETV, endoscopic third ventriculostomy; FASI, Focal Areas of Signal Intensity;
GOOP, Glioma Outside Optic Pathway; NF1, Neurofibromatosis Type 1; OPG, Optic Pathway Glioma; VPS, Ventriculoperitoneal shunt
⇑
Corresponding author at: Department of Paediatric Neurology, Second Faculty of Medicine, Charles University and Motol University Hospital,
150 00 Prague 5, Czech Republic.
E-mail address: m.glombova@gmail.com (M. Glombova).
www.elsevier.com/locate/braindev
Brain & Development xxx (2019) xxx–xxx
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
1. Introduction
Neurofibromatosis type 1 (NF1) is an autosomal
dominant disorder, with complete penetrance, variable
expression and a high rate of new mutations. The inci-
dence is about 1: 2500–3000 individuals, the average glo-
bal prevalence 1 case per 3000 individuals [1]. The NF1
gene is located on the 17th chromosome (17q11.2) and
encodes neurofibromin. Neurofibromin is known as a
tumour suppressor and NF1 patients are at increased
risk for developing benign and malignant tumours.
The diagnosis is based on the National Institutes of
Health (NIH) Diagnostic Criteria for Neurofibromato-
sis Type 1 [2].
The most common NF1 brain gliomas are optic path-
way gliomas (OPG), usually with a presented incidence
of 15–20%, but in fact incidence differs between studies
from 4.8% to 29% [3,4]. The period of OPG manifesta-
tion is mostly up to six years of age, respectively during
first decade of life, but later manifestations have been
noted, too, and OPG appearing in older children or
adults could be more aggressive and more often progress
than in small children [5,6]. Histologically they are usu-
ally pilocytic astrocytomas grade I, they are in one half
to two thirds asymptomatic, and their biologic potential
is more favourable with a better prognosis than in non-
NF1 patients [7,8]. Identifying which lesions will become
aggressive is unpredictable in the beginning and also
spontaneous regression is described [8]. The most com-
mon OPG symptoms are ophthalmological, such as
vision loss or squinting, but also pubertas praecox or
small linear growth could appear too. According to Lis-
ternick et al., the actual incidence of symptomatic OPGs
in NF1 is probably 1.5–7.5% [9]. OPG are classified
according to modified Dodge criteria from 2008 into
four types involving: type 1 – optic nerve/s, type 2 – chi-
asma, type 3 – optic tracts and type 4 – posterior tracts.
H+/means hypothalamus involvement and LM +/
leptomeningeal dissemination. According to Taylor
et al., 98% of OPG involve the optic nerve – one or both,
and/or optic chiasma [10]. The MRI definition of the
OPG is an enlargement of the optic nerve beyond nor-
mal size, with or without contrast enhancement on brain
MR imaging [8].
Gliomas outside the optic pathway (GOOPs) in NF1
children are less commonly mentioned in literature, and
the incidence rate is not really known. Their biologic
potential, in comparison with non NF1 patient with
brain gliomas, is often less aggressive, but in comparison
with OPG it is more important in NF1 patients. The
described localisation is mostly in the brainstem and
cerebellum [5,11,12]. GOOPs have sometimes difficult
differential diagnosis with distinguishing from hyperin-
tense lesions on T2W images typical for NF1. These
findings are called Focal Areas of Signal Intensity
(FASI) [13]; they appear typically at about three years
of age, increase in number and size into adolescence,
and then spontaneously regress. They are typically
hyperintense on T2W and FLAIR MR images and
iso- to mildly hypointense on T1W images. Sometimes
they show slight T1 shortening, which has been related
to myelin clumping or microcalcification. Mass effect,
vasogenic oedema, and contrast enhancement are char-
acteristically absent [14], however, the lesions in the glo-
bus pallidus occasionally have a mild mass effect and
may be bright on T1W images [15]. The incidence of
FASI in the Czech NF1 child population is 86% [16].
Obstructive hydrocephalus is mostly caused by an
expansive lesion compressing the liquor pathway - espe-
cially a chiasmatic, hypothalamic, or brainstem tumour.
The incidence in NF1 patients is 1–5% [4,11,17,18].
Idiopathic aqueduct stenosis of the distal part of the
aqueduct is a rare condition connected with NF1 and
also another possible cause of obstructive hydro-
cephalus in NF1. Incidence is described in about 1.5–
2% of NF1 patients and the aetiology is unknown
[11,12,18]. Phase-contrast MR imaging is helpful for
the diagnosis of aqueduct stenosis [19]. Clinical signs
of increased intracranial pressure from this condition
are usually very inconspicuous, although patients could
have huge findings on brain imaging.
2. Subjects and methods
We undertook a retrospective analysis of 285 NF1
children according to the NIH diagnostic criteria for
NF1, followed up at the Department of Paediatric Neu-
rology in Motol Hospital (which is University Hospital
of Second Medical School of Charles University in Pra-
gue), between 1990 and 2010. This department examined
patients from the whole Czech Republic. Records were
collected for 154/285 (54%) boys and 131/285 (46%)
girls, ranging in age from birth to their nineteenth birth-
day. The cohort contains children followed up at our
Department, evaluated by the same scheme – neurologic
and ophthalmologic examinations, and all had also at
least one brain MR imaging. Children without brain
MR imaging (none or only CT) or with lack of clinical
information were excluded from the study. Neurologic
examination contained evaluation of muscle tonus, cra-
nial nerves function, deep tendon reflexes, cerebellar
function, in nursling evaluation of psychomotor devel-
opment, annually during follow up, at least once, and
in patients with neurologic symptoms/problems as fre-
quently as needed. Opthalmological evaluation included
visual acuity since 3 years old and evaluation of optic
disc (swelling or atrophy) each 4–6 months, in coopera-
tive children color vision and perimeter once a year. –
Some brain MRI examinations were recorded on a
0.5 T machine (14 patients) and the main part of the
cohort (271 patients) on 1.5 T MR equipment. Brain
MRI protocol contain T1W, T2W and FLAIR imaging,
2M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
coronal sequences for optic nerves evaluation, sagital
sequences, and imaging after contrast application. All
findings were evaluated at the Department of Radiology
in Motol Hospital and described by paediatric radiolo-
gists on MRIs with the same OPG diagnostic criteria.
Problematic findings, especially in identifying FASI
and suspected GOOP, were discussed on multidisci-
plinary seminars with paediatric specialists: neurolo-
gists, neurosurgeons, radiologists and oncologists.
The aim of the study was to emphasis especially
GOOPs and their importance for NF1 children although
they has been frequently missed out or outshined by
OPGs, additionally, to assess the causes of obstructive
hydrocephalus in NF1 children and show the rare cases
caused by idiopathic aqueduct stenosis.
2.1. Evaluated MR findings
OPGs were evaluated in all 285 NF1 patients and
were classified as a dilatation of the optic nerve more
than 4 mm on the coronal sequences, and in the chiasma
as a widening more than 4 10 mm (height width).
The measurements were based on Avery et al., Karim
et al., Kornreich et al. and Votruba et al. [8,20–22].
Accessory information as elongation of the optic nerve,
kinking, mass effect and enhancement after contrast
administration were also described. The tumour locali-
sation was defined according to MRI modified Dodge
criteria: type 1 – optic nerve/s, type 2 – chiasma, type
3 – optic tracts and type 4 – posterior tracts, H+/
means hypothalamus involvement and LM +/lep-
tomeningeal dissemination. [10].
GOOPs were evaluated in all 285 patients. The diag-
nosis of a tumour was considered in the presence of two
or more of the following radiological features: expansive
lesion, contrast enhancement and mass effect [5].MRS
was made in only some cases so we did not use it in
the study. The histology was reviewed in available cases
and classified according to the the 2016 World Health
Organisation (WHO) classification of tumours of the
central nervous system [23].
FASI has been defined as hyperintense on T2W and
FLAIR MR images and iso- to mildly hypointense on
T1W images, without mass effect or vasogenic oedema
[14], however, the lesions in the globus pallidus occa-
sionally have a mild mass effect and may be bright on
T1W images [15]. They do not enhance after gadolinium
administration and do not lead to focal neurological
symptoms. Problematic lesions were carefully followed
up and when change (and fullfit glioma definition, espe-
cially when became contrast enhancing) they were called
gliomas. FASI were evaluated in 271/285 (95.1%) cases.
FASI were not evaluated in 14 children with an incom-
plete description examined on 0.5 T MR equipment.
Obstructive hydrocephalus with its cause and idio-
pathic aqueduct stenosis were evaluated in all 290
patients.
2.2. Therapy
The glioma´s therapy means neurosurgery treatment,
actinotherapy or chemotherapy.
Neurosurgeons made partial or total tumour resec-
tion, evaluated cystic portion and/or solved hydro-
cephalus, mostly by ventriculoperitoneal shunt
implantation. In operated cases the histology was also
available. Neurosurgeons made also biopsy in indicated
cases (especially where was suspection to higher grade
glioma), but this was not count as neurosurgery ther-
apy. Actinotherapy was preferred in early 1990th, but
because of side effects and serious consequences was
later determinate for specific cases only. Localised
actinotherapy - gamma knife was used in some patients
too. Nowadays, respectively since 2000th, the
chemotherapy was preferred therapy for NF1 patients
with glioma, especially due to SIOP protocol for low
grade gliomas (SIOP LGG 2004 protocol). Some
patients needed combination of therapeutic methods.
Because the therapeutic strategy subsequently changed
during followed up period, we showed only the num-
bers of treated cases, without next specification. The
therapeutic strategy was made by paediatric oncolo-
gists in cooperation with neurosurgeons in Motol
Hospital.
The OPG treatment criteria were based on imaging
findings – hudge OPG or progression with optahmo-
logic problems as decrease or worsing visual acuity,
optic disc atrophy and neurologic symptoms as propto-
sis, ocular palsy and hydrocephalus development.
GOOP treatment was decided according to imaging
finding but also due to clinical findings – neurologic
symptoms. Neurosurgery has had still an important
position in GOOPs treatment - tumour resection or
hydrocephalus solution.
2.3. Statistical analysis
We compared the clinical importance in the necessity
of treatment in OPG subgroups Dodge 1 and Dodge 2,
and also in OPGs versus GOOPs. Differences were
tested by a x
2
test, with statistically significant P-
value < 0.05, and P-value < 0.01 was considered to be
statistically very significant.
3. Results
We evaluated 285 NF1 children, 131 (46%) girls, 154
(54%) boys.
M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx 3
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
3.1. Optic pathway gliomas
OPGs were found in 77/285 (27%) children, 37 girls
and 40 boys. We classified them according to modified
Dodge criteria: 35 gliomas were Dodge 1 and 42 were
Dodge 2. We did not find patients with Dodge 3 or 4
in our cohort (Table 1,Fig. 1).
OPGs Dodge 2 included 14 OPGs spreading to the
hypothalamus (Dodge 2 + H). Nine of the 14 developed
pubertas praecox and one had other endocrinopathy,
13/14 children had also visual problems. Only one
patient in this subgroup was not treated for an OPG.
We have found three patients with well documented
spontaneous OPG regression – one with Dodge 1
OPG and two had Dodge 2 OPG (none from Dodge
2 + H subgroup).
OPGs were diagnosed at the median age 6 years
(72 months) old (range from birth to 19 years old).
Twenty-nine/35 Dodge 1 OPG patients were only fol-
lowed up – 20/29 had unilateral OPG and 9/29 were
with bilateral OPGs. Twenty-three/29 had normal visus,
which got worse in only one patient, and was joined to
fast worsening of the clinical state, especially due to
the GOOP progression. Six/29 patient had visual prob-
lems, which were stable. Three patients had also some
endocrinologic problems. Sixteen/42 Dodge 2 OPG were
not treated. Forteen/16 had normal visus, 2/16 had
amblyopia and visual impairment, all patients were
without ophthalmologic progression during follow up.
Pubertas praecox was found in 4/16 cases and one/16
was treated with grow hormone (Table 2A).
Thirty-two/77 OPGs were treated – six/35 Dodge 1
and 26/42 Dodge 2 OPGs. Four/6 Dodge 1 OPGs were
unilateral gliomas, three patients had severe visual
impairment and underwent neurosurgery resection of
optic nerve with glioma, one patient had normal visus,
and chemotherapy was indicated due to MRI progres-
sion. Two/6 patients with bilateral OPGs were treated
with combination of treatment methods, because of clin-
ical progression after first therapy. Thirteen/26 Dodge 2
OPGs were from Dodge 2 + H subgroup, and initial
visus was normal in only one/13 case. Other ophthalmo-
logic symptoms were bulbus protrusion (2 cases) and
squinting (1 patient). Nine/13 children had pubertas
praecox and 1/13 another endocrinopathy. Monother-
apy was used in 8/13 cases, and five/13 children must
be treated with combination of treatment methods.
Visus was stable in 5/13 cases, in 7/13 progress and in
only one/13 was little better after treatment. Two
patients developed moya-moya syndrome after
actinotherapy. Another 13/26 Dodge 2 OPGs were from
subgroup without hypothalamus involvement. Two/13
patients had initially normal visus, but both with later
progression, 11/13 patients showed decreased visus, with
next progression in 3 cases, stable in 7 cases, and one
patient was blind on affected eye after neurosurgery.
Other ophthalmologic symptoms were: exophthalmus
(1 patient), nystagmus (3 cases), squinting (1 case).
Five/13 children had pubertas praecox, one mild hyper-
prolactinemia. Ten/13 children were threated with
monotherapy, three/13 with combination (Table 2B).
Respectivelly, in conclusion, only eight symptomatic
OPGs were not treated, and their ophthalmologic func-
tions were stable during follow up. And from treated
symptomatic OPGs only twelve had stable visual func-
tions and even in one case was visus little better.
We compare the clinical importance of Dodge 1 and 2
groups statistically (according to necessity of treatment)
and found statistically very significant differences with
Dodge 2 being clinically more relevant than Dodge 1
OPGs (p < 0.001), because they more often needed
treatment.
3.2. Gliomas outside optic pathway
GOOPs were found in 29/285 (10,2%) of NF1
patients in our cohort. We divided GOOP into three
subgroups - supratentorial, infratentorial and patients
with more than one GOOP (Table 3).
Supratentorial gliomas were found in nine children.
Three were in the hypothalamus, without connection
to the chiasma (Figs. 2A–2C), and were only followed
up, while two of them spontaneously regressed – both
showed contrast enhancement which distinguished
them from FASI, but later the tumour regressed.
The regression in one boy was of both tumours:
OPG and hypothalamic GOOP. Three patients had
GOOP in the thalamus, all caused hydrocephalus,
although all were treated one patient died due to
tumour progression. Other one patient had treated
GOOP in the temporal lobe and last two patients
were treated and both GOOPs caused also hydro-
cephalus: one was located in basal ganglia and the
other in pineal gland.
Infratentorial gliomas were found in 12 children. Five
children had tumours in the cerebellum – three were
treated and one tumour caused hydrocephalus. Six
Table 1
Numbers of OPG in our cohort.
OPG Total F/M Treated Hydr. Died Regr.
Dodge 1 35 18/17 6 0 0 1
Left 11 6/5 3 1
Right 13 5/8 1
Bilateral 11 7/4 2
Dodge 2 42 19/23 26 2 1 2
Dodge 2 + H 14 4/10 13 2 1
Total 77 37/40 32 2 1 3
OPG – optic pathway glioma, F/M = female/male, Hydr. = hydro-
cephalus, Regr. = spontaneous regression, Dodge 2 + H = Dodge 2
+ hypothalamus involvement.
4M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
tumours were located in the brainstem – two were only
followed up, and four treated; one patient died. Hydro-
cephalus developed in three patients. One boy had a
huge tumour involving the brainstem and cerebellum,
and this naturally led to hydrocephalus, and this patient
died due to tumour progression.
Eight children had more than one GOOP, and all were
treated.Three patients did not have OPG together, two
patients developed hydrocephalus.
The median age of GOOP discovered was 9 years
and 10 months old (range from three years and three
months to 18 years old). Seven/29 patients were asymp-
tomatic, 22/29 were treated, included all with more
than two GOOPs. Six patients underwent neuro-
surgery, one chemotherapy and four actinotherapy
only. Eleven were treated by more than one modality.
The histology of the available cases included astrocy-
tomas grade I or II, only one patient had astrocytoma
grade II-III.
We compared the clinical significance of GOOPs and
OPGs in terms of treatment necessity and we discovered
that GOOPs were clinically, significantly more impor-
tant for NF1 children than OPGs (p < 0.01).
FASI were found in 229/271 (84.5%) cases – 106 girls
and 123 boys, in the typical localisation described in
NF1 patients (Figs. 3A and 3B).
Eighty-seven out of 285 (30.5%) patients had some
brain glioma.
Fig. 1. Glioma of prechiasmatic part of right optic nerve (white arrow), coronal T2-TSE.F/S (T2 – weighted turbo spin echo/fat saturation) image.
M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx 5
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
3.3. Obstructive hydrocephalus
Obstructive hydrocephalus was found in 22/285
(7.7%) patients, in the median age 10 years 1 month
old (range from three years and six months to 19 years
old). Fourteen cases were caused by glioma, respectively
two OPGs and 12 GOOPs were leading to hydro-
cephalus. The second most common cause was idio-
pathic aqueduct stenosis of distal part of aqueduct, in
six patients. The other two patients had hydrocephalus:
due to an expansive arachnoid cyst in one patient and
secondary aqueduct stenosis (after actinotherapy) in
the last one child.
3.4. Idiopathic aqueduct stenosis
Idiopathic aqueduct stenosis was found in two girls
and four boys, in total 6/285 (2.1%) patients (Table 4).
Only one boy had OPG and none had GOOP. The med-
ian age of manifestation was 11 years 2 months old (a
range from seven years six months to 16 years
11 months). The clinical signs were very inconspicuous
Table 2
OPGs – therapy, visual outcome.
A - only followed OPGs
Dodge classification Initial visus Visus during follow up Other
opthalmologicsymptoms
Endocrinologic problems
Normal Impairment Stable Progression
Dodge
129
Unilateral
OPG 20
17 3
*
20 0 Bulbus protrusion 1 GH treatment 1
Puberta tarda 1
Bilateral
OPG 9
63
**
81
Dodge
216
14 2
+
16 0 Puberta praecox
4GH treatment 1
B - treated OPGs
Dodge classification Initial visus Visus during follow up Other
opthalmologicsymptoms
Endocrinologic problems Therapy
Normal Impairment Stable Progression NS ActT ChT Multi
Dodge
16
Unilateral
OPG4
13
§
Amaurosis
3
1 Amaurosis after NS 3 Puberta praecox 1 3 1
Bilateral
OPG 2
02
§§
02 2
Dodge
226
Dodge - 2
without H13
2 11 7 5 Exophthalmus
1Nystagmus 3, Squint 1
Puberta praecox
5Hyperprolactinemia 1
26 2 3
Dodge - 2 +
H13
1 12 5 7 Bulbus protrusion
2Squint 1
Puberta praecox 9Other
endocrinopathy 1
14 3 5
Better 1
GH = grow hormone, H = hypothalamus, NS = neurosurgery, ActT = actinotherapy, ChT = chemotherapy, Multi = multimodal treatment.
*
Hypermetropia with astigmatism 1, myopia 1, decreased visus 1.
**
Unilateral decreased visus 2, hypermetropia 1.
+
Amblyopia and decreased visus 2.
§
Severe visual impairment (nearly amaurosis).
§§
Severe visual impairment.
Table 3
Numbers of GOOPs in our cohort.
GOOP Total F/M Treated Hydr. Dead Regr. OPG – Dodge1 OPG – Dodge2
Supratentorial 9 6/3 6 5 1 2 1 4
Infratentorial 12 4/8 8 5 2 0 3 6
more than 1 8 3/5 8 2 0 0 3 2
-2 or multiple GOOP, no OPG 3 2/1 2 2
-multiple GOOP, with OPG 5 1/4 5 3 2
Total 29 13/16 22 12 3 2 7 12
GOOP = glioma outside optic pathway, F/M = female/male, Hydr. = hydrocephalus, Regr. = spontaneous regression, OPG = optic pathway
glioma.
6M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
for months and in most cases the first sign was a head-
ache. In two cases, vomiting was irregular and attached
importance to some gastrointestinal problems, similarly
to an increased frequency of seizures in another one
patient, which was regarded as inadequate drug therapy.
In two cases, a severe impairment to speech develop-
ment was described. In an asymptomatic case the hydro-
cephalus was found by routine MR imaging. All cases
were treated; four out of six patients underwent inter-
ventriculostomy with shunt placement from the third
to fourth ventricle. A ventriculoperitoneal shunt (VPS)
was implanted in two out of six cases. One girl, who
was asymptomatic, developed apallic syndrome after
VPS implantation, which lasted for a few months and
then slowly got better.
4. Discussion
NF1 is an illness with many complications, including
significantly increased tumour risks and a risk of idio-
pathic aqueduct stenosis and development of
hydrocephalus.
4.1. Optic pathway glioma
We have found 27% NF1 children with OPG in our
cohort, which is higher than the commonly stated 15–
20%. But, in fact, the data differs widely in literature
from 4.8%, in McGaughran et al., to the highest inci-
dence 28.6% described by Blazo et al. and 29% by Leisti
[3,4,24]. The reason should be in the lack of a strictly
Fig. 2A. Post contrast T1/SE in coronal plane, 10 yr. girl with NF1, hypothalamic GOOP above optic chiasma enhances after Gadolinium
application (white arrow).
M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx 7
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
defined pathology of the optic nerve and methods of
cohort definition and MRI indications. We consider a
normal width of the optic nerve as up to 4 mm, and
enlargement above this was evaluated as glioma and
the normal size of the chiasma was assessed as
(height width) 4x10 mm. But, in the literature only a
few papers defined the normal optic nerve diameter.
We based the limits on Avery et al. (3.9 mm), Karim
et al. (a mean optic nerve diameter 3.99 ± 0.04 mm, just
posterior to the globe, decreasing to 3.50 ± 0.04 mm
posteriorly), and Votruba et al. (3.5 ± 0.3 mm) [20–22].
Kornreich et al. defined OPG only as an enlargement
above the normal size and in chiasma greater than
1cm [8]. The other findings (e.g., abnormal optic nerve
elongation, kinking, presence of T2 hyperintensity, and
enhancement after contrast administration we consid-
ered as additional data, similarly to Avery et al. [20].
The therapeutic strategy of OPGs in NF1 subse-
quently changed during the last thirty years, to prefer
chemotherapy for OPGs and other low grade gliomas
in an effort to avoid neurosurgery interventions and
actinotherapy [25], and with knowledge about this
mostly benign and stable disease, most of patients are
only followed up on. The most jeopardised OPG sub-
group is Dodge 2 + H. These patients mostly need treat-
ment but usually had some additional clinical problems,
visual or endorinological. Moreover, these tumours
could also cause hydrocephalus. The treatment indica-
Fig. 2B. FLAIR in axial plane, hypothalamic GOOP has increased signal (white arrow), mesencephalic FASI (black arrows) have also increased
signal.
8M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
tion criteria assess not only clinical and imaging find-
ings, but also quicknes of symptoms/findings arising
and their progression.
The spontaneous regression of OPG had been
described in rare cases of NF1 patients [8]. This phe-
nomenon was described in case reports [14,26] and in
followed up on in NF1 cohorts too [5,8]. Shuper et al.
were the only ones noting a case of one NF1 patient
with OPG regressing significantly (about 50% of vol-
ume) during follow up, but later, after 6 years, regrowth
was found, and the patient had to be treated [27]. Lister-
nick et al. and other authors evaluate the development
of OPGs as unpredictable, while most OPGs remain
unchanged in the long term, a smaller part progressing
in size and/or clinical manifestations and a very small
part of OPG spontaneously regress [9]. A similar distri-
bution of clinical manifestations was seen in our cohort.
We described spontaneous regression in 4/285 (1.4%)
patients and none of these patients had glioma regrowth
during next follow up. In contrast to this, the same num-
ber of patients (4/285, 1.4%) died according to tumour
progression in our cohort.
4.2. Gliomas outside optic pathway
GOOPs are less commonly mentioned in literature,
although they are often clinically important. Ferner
et al. noted a group of gliomas outside the optic path-
way, mostly located in the brainstem and cerebellum,
with a frequency of 2–3% [12]. Noble et al. described
four patients with GOOP from the 121 patients evalu-
ated (3.3%), and Williams et al. reported gliomas located
Fig. 2C. Post contrast T1/SE in axial plane, hypothalamic GOOP enhances after Gadolinium application (white arrow) contrary to unenhancing
mesencephalic FASI (black arrow).
M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx 9
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
in the brainstem, diencephalus and cerebellum with a
frequency of 3.5% [28,29]. Blanchard et al. conducted
systemic MRIs in 306 children with NF1 younger than
six years old and found four patients with OPG and
GOOP (4/306, 1.3%) in total [7]. We have found GOOP
in 10.2% our patients. Histological findings in available
cases were astrocytomas grade I or II, only one was
grade II-III. The differential diagnosis of GOOP is
sometimes complicated by FASI, which are the most
common MR findings in NF1 children. But even in these
cases histological examination is not indicated because
benign character of these lesions in contrast with risks
and complications contained with biopsy. These patients
must be carefully long term followed up by neurologist
and also MRI should be made repeatedly. Histologic
examination is made in cases where neurosurgery treat-
ment is necessary especially when hydrocephalus
appears or tumour or some parts of tumour should be
removed, cystic portions drained etc. A common FASI
aetiology (due to NF1) appears to be a neurofibromin
disorder but the mechanism has not been elucidated
yet [1]. Our ambiguous cases were evaluated by paedi-
atric radiologists and widely discussed at multidisci-
plinary seminars, and patients were followed up in the
long-term.
Brain gliomas were found in our cohort, in total
87/285 (30.5%) NF1 children. The cumulative number
of brain gliomas better expresses the overall risk of brain
tumour manifestations in NF1 than the OPGs frequency
alone.
4.3. Obstructive hydrocephalus
The incidence of obstructive hydrocephalus in NF1
is described as 1–5% [4,11,17,18] and tumours are the
most common cause. We had a slightly higher num-
ber of obstructive hydrocephalus in our cohort, at
7.7%.
Fig. 3A. FLAIR in axial plane, 11 yr. Boy with NF1, FASI involve basal ganglia (black arrows).
10 M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
4.4. Idiopathic aqueduct stenosis
Idiopathic aqueduct stenosis of the distal part of the
aqueduct is considered very rare, but for NF1 character-
istics a possible cause of hydrocephalus in 1.2–2% of the
NF1 patients [3,11,12]. Idiopathic aqueduct stenosis
caused hydrocephalus in six out of 22, respectively six
out of 285 (2.1%), of our patients. Cre
´ange et al.
described four children in the evaluated group of
patients with idiopathic aqueduct stenosis, and one of
these patients was asymptomatic, without signs of
intracranial hypertension [11]. We had also one asymp-
tomatic patient with hydrocephalus in our cohort and
the others had only inconspicuous clinical signs without
significant signs of intracranial hypertension, despite a
large hydrocephalus found on the brain MRI.
All NF1 patients with hydrocephalus are recom-
mended for neurosurgery treatment – VPS implantation,
interventriculostomy, or nowadays endoscopic third
ventriculostomy (ETV) is preferred [11,17,30]. All of
our NF1 patients with idiopathic aqueduct stenosis
related hydrocephalus were treated years before ETV
was available in our hospital. Nowadays, ETV is pre-
ferred to resolve hydrocephalus in NF1 children in our
department.
The asymptomatic hydrocephalus in one girl with
idiopathic aqueduct stenosis was found by a routine
MRI examination. She developed apallic syndrome after
a VPS implantation and she got better after nearly one
year. The clinical course demonstrated a slow increase
of intracranial hypertension with an adaptation to high
intracranial pressure and subsequent risks in fast pres-
sure compensation. Pivalliza et al. published a case
report of a patient with unexpected hydrocephalus due
to idiopathic aqueduct stenosis, who suddenly died at
21 years of age due to dramatic hydrocephalus decom-
pensation just after a banal surgery performed under
total anaesthesia [31]. The risk of idiopathic aqueduct
stenosis is one of the other reasons for a carefull neuro-
logic follow up and one of indication of brain imaging.
Fig. 3B. T2/TSE in axial plane, FASI involve periventicularly both cerebellar hemispheres (black arrows).
M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx 11
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
5. Conclusion
The prevalence of OPGs in our cohort was 27%. The
most important was the Dodge 2 + H subgroup, but
generally the clinical course of OPGs is unpredictable,
with the possibility of spontaneous regression but also
dramatic deterioration. GOOPs were found in 10.2%
of our patients, in median age 9 years 10 months old
(range from three years and three months to 18 years
old), and they proved to be a higher risk for NF1
patients, more often needing treatment and potentially
leading also to hydrocephalus. The total brain glioma
number (OPGs and only GOOPs together) better
reflected the overall brain tumour risk for NF1 children.
We would like to emphasise the 7.7% total occurrence of
obstructive hydrocephalus and 2.1% prevalence of
obstructive hydrocephalus due to idiopathic aqueduct
stenosis in NF1 children. The clinical signs of hydro-
cephalus according to idiopathic aqueduct stenosis were
inconspicious and the development of hydrocephalus
was unpredictable in comparison with hydrocephalus
due to tumour. The risk of developing hydrocephalus
according to idiopathic aqueduct stenosis is another
possibility to carefully follow up on in NF1 children.
Acknowledgements
We thank all the colleagues from our departments for
the years of great collaboration. We acknowledge the
Ministry of Health for supporting this Research Project
at Motol University Hospital, No. 00064203,6005.
Conflict of interest
All the authors claim no conflicts of interest.
Appendix A. Supplementary data
Supplementary data to this article can be found
online at https://doi.org/10.1016/j.braindev.2019.04.
003.
References
[1] Gutmann DH, Ferner RE, Listernick RH, Korf BR, Wolters PL,
Johnson KJ. Neurofibromatosis type 1. Nat Rev Dis Primers
2017;3:1–17.
[2] National Institute of Health Consensus Development Conference
Statement: Neurofibromatosis. Arch Neurol Chicago 1988; 45:
575-8.
[3] McGaughran JM, Harris DI, Donnai D, Teare D, MacLeod R,
Westerbeek R, et al. A clinical study of type 1 neurofibromatosis
in North West England. J Med Genet 1999;36:197–203.
[4] Leisti E-L. Radiologic findings of the head and spine in
neurofibromatosis I (NF I) in Northern Finland. Academic
Disertation. Oulu: Faculty of Medicine, University of Oulu; 2003.
[5] Guillamo JS, Cre
´ange A, Kalifa Ch, Grill J, Rodriguez D, Doz F,
et al. Prognostic factors of CNS tumours in neurofibromatosis 1
(NF1): a retrospective study of 104 patients. Brain
2003;126:152–60.
[6] Listernick R, Ferner RE, Liu GT, Gutmann DH. Optic pathway
gliomas in neurofibromatosis-1: controversies and recommenda-
tions. Ann Neurol 2007;61:189–98.
[7] Blanchard G, Lafforgue M-P, Lion-Franc¸ois L, Kemlin I,
Rodriguez D, Castelnau P, et al. Systematic MRI in NF1 children
under six years of age for the diagnosis of optic pathway gliomas.
Study and outcome of a French cohort. Eur J Paediatr Neurol
2016;20:275–81.
[8] Kornreich L, Blaser S, Schwartz M, Shuper A, Vishne TH, Cohen
IJ, et al. Optic pathway glioma: correlation of imaging findings
with the presence of neurofibromatosis. Am J Neuroradiol
2001;22:1963–9.
[9] Listernick R, Louis DN, Packer RJ, Gutmann DH. Optic
pathway gliomas in children with neurofibromatosis 1: consensus
statement from the NF1 optic pathway glioma task force. Ann
Neurol 1997;41:143–9.
[10] Taylor T, Jaspan T, Milano G, Gregson R, Parker T, Ritzmann
T, et al. Radiological classification of optic pathway gliomas:
experience of a modified functional classification system. Br J
Radiol 2008;81:761–6.
[11] Cre
´ange A, Zeller J, Rostaing-Rigattieri S, Brugire
´res P, Degos J-
D, Revuz J, et al. Neurological complications of neurofibromato-
sis type 1 in adulthood. Brain 1999;122:473–81.
[12] Ferner RE, Huson SM, Thomas N, Moss C, Willshaw H, Evans
DG, et al. Guidelines for the diagnosis and management of
individuals with neurofibromatosis 1. J Med Genet 2007;44:
81–8.
[13] Osborn AG. Neurofibromatosis type 1. Philadelphia: Elsevier;
2016. p. 84–7.
[14] Rossi LN, Triulzi F, Parazzini C, Maninetti MM. Spontaneous
improvement of optic pathway lessions in children with neurofi-
bromatosis type 1. Neuropediatrics 1999;30:205–9.
Table 4
Patients with idiopathic aqueduct stenosis and clinical signs of hydrocephalus.
Patient
No.
Gender Sign of hydrocephalus Years of age at time of
hydrocephalus finding
Therapy Other clinical data
1 F Asymptomatic 16 y 11 m VPS apallic syndrome after shunt implantation
2 F Headache 7 y 6 m Interventriculostomy severe speech development impairment,
mild mental retardation
3 M Increased seizure frequency, left side
hemiparesis, bilateral abducens palsy
8 y 2 m VPS seizures since 3 years, speech development
impairment, mild mental retardation
4 M Headache, intermittent vomiting for
long time
8 y 9 m interventriculostomy aortal stenosis - cardiology follow up since
3 month of age
5 M Headache, vomiting 13 y 6 m interventriculostomy no other clinical problems
6 M Headache 15 y 6 m interventriculostomy mild mental retardation
F = female, M = male, VPS = ventriculoperitoneal shunt, y = years, m = months.
12 M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
[15] DiPaolo DP, Zimmerman RA, Rorke LB, Zackai LH, Bilaniuk
LT, Yachnis AT. Neurofibromatosis type 1: pathologic substrate
of high-signal intensity foci in the brain. Radiology
1995;195:721–4.
[16] Petrak B, Lisy J, Kraus J, Kyncl M, Zatrapa T. The focal areas of
high-signal intensity on T2-weighted MRI of the brain are
significant for the diagnosis of neurofibromatosis von Reckling-
hausen type 1. Pediatrics 2008;121:S147.
[17] Garg P, Shruthi KM, Maheshwaran V, Devanand B. Rare case of
non-neoplastic aqueductal stenosis due to web in a patient with
neurofibromatosis type-1. Neurol. India 2016;64:1384–7.
[18] Van Es S, North KN, McHugh K, Silva de M. MRI findings in
children with neurofibromatosis type 1: a prospective study.
Pediatr Radiol 1996;26:478–87.
[19] Stoquart-El Sankari S, Lehmann P, Gondry-Jouet C, Fichten A,
Godefroy O, Meyer M-E, et al. Phase-contrast MR imaging
support for the diagnosis of aqueductal stenosis. Am J Neuro-
radiol 2009;30:209–14.
[20] Avery RA, Mansoor A, Idrees R, Biggs E, Alsharid MA,
Packer RJ, et al. Quantitative MRI criteria for optic pathway
enlargement in neurofibromatosis type 1. Neurology
2016;86:2264–70.
[21] Karim S, Clark RA, Poukens V, Demer JL. Demonstration of
systematic variation in human intraorbital optic nerve size by
quantitative magnetic resonance imaging and histology. Invest
Ophthalmol Vis Sci 2004;45:1047–51.
[22] Votruba M, Leary S, Losseff N, Bhattacharya SS, Moore AT,
Miller DH, et al. MRI of the intraorbital optic nerve in patiens
with autosomal dominant optic atrophy. Neuroradiology
2000;42:180–3.
[23] Louis DN, Perry A, Reifenberger G, Deimling A, Figarella-
Branger D, Cavenee WK, et al. The 2016 world health organi-
zation classification of tumors of the central nervous syste
´m: a
summary. Acta Neuropathol 2016;131:803–20.
[24] Blazo MA, Lewis RA, Chintagumpala MM, Frazier M, McClug-
gage C, Plon SE. Outcomes of systematic screening for optic
pathway tumors in children with neurofibromatosis type 1. Am J
Med Genet A 2004;127A:224–9.
[25] Sabatini C, Milani D, Menni F, Tadini G, Esposito S. Treatment
of neurofibromatosis type 1. Curr Treat Options Neurol
2015;17:26.
[26] Zuccoli G, Ferrozzi F, Sigorini M, Virdis R, Bassi P, Belloni M.
Early spontaneous regression of a hypothalamic/chiasmatic mass
in neurofibromatosis type 1: MR findings. Eur Radiol
2000;10:1076–8.
[27] Shuper A, Kornreich L, Michowitz S, Schwartz M, Yaniv I,
Cohen IJ. Visual pathway tumors and hydrocephalus. Pediatr
Hematol Oncol 2000;17:463–8.
[28] Noble F, Kornberg AJ, Elder JE, Delatycki MB. Retrospective
analysis of patiens attending a neurofibromatosis type 1 clinic. J
Paediatr Child Health 2007;43:55–9.
[29] Williams VC, Lucas J, Babcock MA, Gutmann DH, Korf B,
Maria BL. Neurofibromatosis type 1 revisited. Pediatrics
2009;123:124–33.
[30] Roth J, Ber R, Wisoff JH, Hidalgo ET, Limbrick DD, Berger DS,
et al. Endoscopic third ventriculostomy in patients with neurofi-
bromatosis type 1: a multicenter international experience. World
Neurosurg 2017;107:623–9.
[31] Pivalizza EG, Rabb MF, Johnson S. Fatal hydrocephalus in a
patient with neurofibromatosis. Anesthesiology 2000;92:630.
M. Glombova et al. / Brain & Development xxx (2019) xxx–xxx 13
Please cite this article in press as: Glombova M et al. Brain gliomas, hydrocephalus and idiopathic aqueduct stenosis in children with neurofi-
bromatosis type 1. Brain Dev (2019), https://doi.org/10.1016/j.braindev.2019.04.003
