Key words
subependymom - subependymoma mri - subependymome review - MR-imaging
Background
The authors are unaware of a large case series on image-based morphological characteristics
of subependymomas. Although etiological, pathological and therapeutic characteristics
have been the subject of some publications for decades and have now been well studied,
the image-based morphological aspect has received much less attention [1]
[2]
[3].
Introduction
Subependymomas are a rare, benign, noninvasive entity of ependymal origin representing
approximately 0.2 – 0.7 % of all intracranial tumors worldwide [1]
[3]
[4]. The true incidence remains unclear, since subependymomas are generally asymptomatic
and usually appear as incidental findings in autopsies or imaging. Since Scheinker’s
initial description in 1945 [5], there have been few case series with generally small cohorts [2]
[6]
[7]
[8]
[9]. In addition to the more frequent occurrence of subependymomas of the fourth ventricle
(approx. 56 – 60 %), origin of tumor growth can be observed on the lateral ventricle
in about 30 – 40 % of cases [2], and more rarely on the spine [4]. The incidence is much more common among middle-aged and older men. Clinical symptoms
due to cerebrospinal fluid accumulation correspond to tumor localization; clinical
manifestation of tumors in the fourth ventricle is less common, compared with tumors
in the lateral ventricles, since CSF accumulation or seizures can occur here more
frequently [2]
[10]. Tumor recurrence is quite rare [11]; spinal drop metastases have not been described [11].
As initially described by Scheitauer in 1978, there is a seldom-observed special form
of a mixed tumor with cell clusters of a subependymoma as well as an ependymoma; that
is, morphologically “typical” subependymomas with atypical growth tendency [1].
Materials and Methods
Patients and techniques
The retrospective study was approved by the local ethics committee. After in-house
database research under the keyword “subependymoma” all relevant MRI and CT examinations
between January 2009 and January 2017 were included in which subependymomas were morphologically
suspected. This corresponded to 33 patients, of whom 22 were male and 11 female. The
mean age was 58.6 years (29 – 86 years). Based on clinical indication, and after providing
written consent, all patients were examined using 3.0 T magnetic resonance imaging
(MRI) equipment (Magnetom Verio, Siemens Healthcare, Erlangen, Germany) or a 1.5 T
MRI device (Intera Achieva, Phillips Medical Systems, Eindhoven, Netherlands). Examinations
followed a clinically-established protocol including at least axial T2-weighted (w),
FLAIR, T1w, T1w after intravenous contrast administration, T2*w, DWI/ADC and coronary
T1w after intravenous contrast administration. The contrast medium was administered
intravenously in a weight-adapted standard dose (0.1 ml/kg body weight of a 0.1 molar
gadolinium-based contrast agent) (Gadovist, Bayer Schering). Morphology, signal and
contrast behavior at initial diagnosis as well as the presence of possible growth
of the tumors in follow-up studies was consensually described by two experienced neuroradiologists.
Results
Tumor entities
The etiology of 10 tumors was confirmed histologically, of which 6 were subependymomas
and 4 intermediate forms between ependymoma and subependymoma.
Due to the incidental findings, wait-and-see behavior with imaging controls was selected
for 23 tumors. These remained stable over a period of a few months to 10 years, so
that the diagnosis of a subependymoma was made based on imaging.
Location and size
In most patients (n = 18), the tumor was located in the fourth ventricle, accompanied
in a few cases by tumor tissue growth into the foramen Magendie (n = 10) or the foramina
Luschka (n = 4). In 11 patients, the tumor was found in the lateral ventricles, 6
of them in the lateral ventricular anterior horn, 4 in the lateral ventricular posterior
horn, and in the cella media of one patient. Less common observed localizations included
the cisterna magna with a transition into the right foramen Luschka (n = 1) or exclusive
tumor location in the foramen Magendie (n = 3). One of the tumors located in the lateral
ventricle was classified as a mixed tumor whereas the other 3 tumors originated from
mixed tumor portions of a subependymoma and an ependymoma in the 4th ventricle with
transition to the foramina Luschka (n = 2) and the fourth ventricle with transition
into the foramen Magendie (n = 1). In addition, there was a wide range of the extent
of the tumors; 5 masses were smaller than 1 cm (15.15 %); 8 masses exhibited a width
between 1 and 2 cm (24.24 %); 10 had a maximum diameter greater than 2 cm (30.3 %).
Three of the mixed-form tumors with subependymoma and ependymal cell clusters were
assigned to the last group with diameters greater than 2 cm. The fourth tumor with
mixed components exhibited a maximum dimension between 1 and 2 cm.
Imaging characteristics
In most cases (n = 21) the tumors showed a sharp boundary with the adjacent brain
tissue ([Fig. 1a]); less commonly there was an irregular sharp distinction (n = 11, see [Fig. 2b–d]), and in one case (n = 1, indicated in [Fig. 3f]), there was a fuzzy demarcation with respect to the neighboring brain parenchyma.
An irregular blurred boundary and in one case a sharp distinction could be observed
in the mixed special forms with ependymoma components. An inhomogeneous parenchymal
texture was observed in 25 cases, partly T2-hyperintense to cystic internal components,
and in the other 8 cases the internal signal was homogeneous. In the majority of cases
(n = 28), a T1w cortex isointense signal behavior of the tumor with respect to the
brain parenchyma was observed as a special signal characteristic in the individual
sequences; more rarely a comparatively hypointense internal signal (n = 4) was seen
in T1-weighting. The T2-weighted signal behavior was predominantly hyperintense (n = 32),
less frequently combined with iso- to hypointense components in the case of an overall
inhomogeneous signal. Diffusion restrictions were not observed. Contrast medium absorption
showed large fluctuations from completely absent to partially nodular or occasionally
homogeneously flat contrast accumulation of the tumor tissue ([Fig. 1f], [2], [3f]; [Table 1]). Small calcifications could be observed in all 6 patients who additionally had
undergone skull CT ([Fig. 1b]). Of the tumors with mixed histology, there was only one case of computed tomography
which, however, also exhibited calcification of the tumor.
Fig. 1 Infratentorial Subependymom a-f: infratentorial subependymoma in the 4th ventricle with tumor growth over the foramina
Lushkae on both sides (red arrows) and the foramen Magendie (green arrow). Susceptibility
effects in the T2* sequence indicating small calcifications or bleeding (blue arrows).
(a T2 TSE, b. T2*, c ADC map, d FLAIR, e T1 before KM and f T1 after i. v. gadolinium administration) The images available to us with an interval
of 2 years show constant size and shape.
Fig. 2 a-f Subependymoma in the central part of the left lateral ventricle (a-d of 2016: a T2 TSE, b FLAIR, d T1 post KM, e DWI and c of 2007: FLAIR) with a constant size and shape during the 9-year observation period.
There is no diffusion restriction (e+f) or contrast enhancement d.
Fig. 3 Mixed form of subependymoma with ependymal cell fractions f: preoperative imaging of an infra-temporal tumor in the 4th ventricle, which histopathologically
corresponds to a mixed form of subependymoma with ependymal cell fractions. (a T2 TSE, b FLAIR, c T1 SE native, d DWI, e ADC-map and f T1 SE after i. v. gadolinium administration).
Table 1
Overview of the patient population with morphological images with suspicion of subependymoma.
age
|
symptoms
|
location
|
side
|
size (mm)
|
boundary
|
texture
|
T1 signal
|
DWI
|
T2*
|
T2 signal
|
CM absorption
|
miscellaneous
|
progression
|
CSF accum.
|
CT
|
therapy
|
histo available?
|
77
|
incidental findings
|
foramen magendi 4th ventricle
|
|
4.6 × 8.7
|
sharp
|
homogeneous
|
iso
|
iso
|
hyper
|
hyper
|
none
|
|
n.a.
|
no
|
n.a.
|
|
|
71
|
incidental findings
|
lat. ventricle
|
left
|
7.5 × 14.3
|
indistinct
|
inhomogeneous with cystic components
|
iso
|
iso
|
hypo
|
hyper to iso
|
n.a.
|
|
n.a.
|
no
|
smallest calcifications peripheral
|
|
|
SO
|
incidental findings
|
lat. ventricle
|
left
|
12.2 × 17.1
|
irregular sharp
|
inhomogeneous with T2 hyperintense components
|
iso
|
iso
|
n.a.
|
hyper to iso
|
none
|
|
since 2011 constant
|
no
|
n.a.
|
|
|
40
|
incidental findings
|
4th ventricle and foramina luschka
|
bilateral
|
20.5 × 48.8
|
irregular sharp
|
inhomogeneous with T2 hyperintense components
|
iso
|
iso
|
hyper
|
hyper to iso
|
partially nodular
|
|
postoperatively constant
|
no
|
partially calcified
|
part. resection
|
WHO I
|
78
|
incidental findings
|
lat. ventricle
|
right
|
11 × 12
|
irregular sharp
|
inhomogeneous
|
iso
|
iso
|
iso
|
hyper to iso
|
none
|
|
since 2011 constant
|
no
|
n.a.
|
|
|
63
|
incidental findings
|
4th ventricle
|
|
4.5 × 6.3
|
sharp
|
homogeneous
|
iso
|
iso
|
hyper
|
hyper
|
none
|
|
n.a.
|
no
|
n.a.
|
|
|
56
|
incidental findings
|
4th ventricle and foramen luschka
|
left
|
16 × 29
|
sharp
|
inhomogeneous
|
iso
|
iso
|
n.a.
|
hyper to iso
|
partially punctate
|
|
since 2011 constant
|
no
|
n.a.
|
|
|
83
|
unknown
|
4th ventricle and foramen luschka
|
left
|
30.5 × 34.7
|
irregular sharp
|
inhomogeneous
|
hypo
|
n.a.
|
hypo
|
hyper
|
inhomogeneous
|
|
|
no
|
n.a.
|
part. resection
|
WHO l/ll
|
30
|
double vision, flicker
|
lat. ventricle
|
right
|
20 × 23
|
sharp
|
homogeneous
|
hypo
|
n.a.
|
n.a.
|
hyper
|
n.a.
|
strong T2 hyper, FLAIR iso
|
postoperatively constant
|
yes
|
n.a.
|
full resection
|
WHO I
|
65
|
incidental findings
|
4th ventricle
|
|
7.1 × 9.9
|
sharp
|
homogeneous
|
iso
|
n.a.
|
n.a.
|
hyper
|
partially nodular
|
|
since 2011 constant
|
no
|
n.a.
|
full resection
|
WHO I
|
72
|
headaches, nausea
|
lat. ventricle
|
right
|
42 × 43
|
sharp
|
inhomogeneous
|
iso
|
iso
|
hypo
|
hyper to hypo
|
inhomogeneous
|
partial hyperperfusison
|
postoperatively constant
|
yes
|
partially calcified
|
full resection
|
WHO I/II
|
82
|
none
|
right foramen luschka/cisterna magna
|
right
|
16.5 × 17.9
|
sharp
|
inhomogeneous
|
iso
|
iso
|
n.a.
|
hyper
|
none
|
centrally punctate T2 hypo, T1 hyper
|
since 2011 constant
|
no
|
central calcifications
|
|
|
48
|
none
|
4th ventricle
|
|
8.4 × 13
|
sharp
|
inhomogeneous
|
iso
|
iso
|
n.a.
|
hyper
|
none
|
|
n.a.
|
no
|
n.a.
|
|
|
72
|
none
|
lat. ventricle
|
right
|
35 × 65
|
sharp
|
inhomogeneous
|
hypo
|
n.a.
|
partially hypo
|
hyper to hypo
|
inhomogeneous
|
|
postoperatively constant
|
yes
|
n.a.
|
full resection
|
WHO I
|
61
|
incidental findings
|
foramen magendi
|
|
5 × 6
|
irregular sharp
|
inhomogeneous
|
iso
|
iso
|
hyper
|
hyper
|
none
|
|
constant
|
no
|
n.a.
|
|
|
53
|
headache, blurred vision, deafness
|
foramen magendi
|
|
10 × 21
|
irregular sharp
|
inhomogeneous
|
iso
|
iso
|
n.a.
|
hyper to iso
|
partially punctate
|
|
since 2007 constant
|
no
|
n.a.
|
|
|
42
|
incidental findings
|
foramen magendi 4th ventricle
|
|
4.9 × 6.7
|
irregular sharp
|
homogeneous
|
iso
|
iso
|
n.a.
|
hyper
|
slightly flat
|
|
3 months constant
|
no
|
n.a.
|
|
|
31
|
incidental findings
|
foramen magendi 4th ventricle
|
|
18.6 × 15.9
|
sharp
|
inhomogeneous with cystic components
|
iso
|
iso
|
hyper, singly hypo
|
hyper to hypo
|
partially nodular
|
|
constant
|
no
|
n.a.
|
|
|
51
|
incidental findings
|
Foramen magendi 4th ventricle
|
|
15.5 × 17.1
|
irregular sharp
|
inhomogeneous
|
iso
|
iso
|
n.a.
|
hyper
|
partially nodular
|
|
postoperatively constant
|
no
|
n.a.
|
full resection
|
WHO I/II
|
52
|
incidental findings
|
lat. ventricle
|
right
|
4 × 5
|
sharp
|
homogeneous
|
iso
|
iso
|
n.a.
|
hyper
|
none
|
|
n.a.
|
no
|
n.a.
|
|
|
30
|
incidental findings
|
lat. ventricle
|
right
|
13.1 × 19.8
|
sharp
|
homogeneous
|
hypo
|
hypo
|
hypo
|
hyper
|
partially nodular
|
strong T2 hyper, FLAIR iso
|
n.a.
|
no
|
n.a.
|
|
|
85
|
dizziness, headaches
|
foramen magendi 4th ventricle
|
|
15.6 × 24.1
|
sharp
|
inhomogeneous
|
|
|
|
|
|
no MRI due to HSM
|
since 2011 constant
|
yes
|
partially calcified
|
|
|
52
|
incidental findings
|
lat. ventricle
|
right
|
9.9 × 17
|
sharp
|
inhomogeneous
|
iso
|
iso
|
hyper
|
hyper
|
partially nodular
|
|
since 2012 constant
|
no
|
no calcifications
|
|
|
73
|
incidental findings
|
lat. ventricle
|
right
|
9.7 × 12.7
|
sharp
|
inhomogeneous
|
iso
|
iso
|
hyper
|
hyper to iso
|
none
|
|
since 2011 constant
|
no
|
n.a.
|
|
|
55
|
incidental findings
|
foramen magendi 4th ventricle
|
|
26 × 34.9
|
irregular sharp
|
inhomogeneous
|
iso
|
iso
|
n.a.
|
hyper to iso
|
inhomogeneous
|
|
postoperatively constant
|
no
|
n.a.
|
part. resection
|
WHO I/II
|
55
|
incidental findings
|
4th ventricle foramen magendi and foramen luschka
|
right
|
10 × 27
|
irregular sharp
|
homogeneous
|
iso
|
iso
|
hypo
|
hyper
|
none
|
|
constant
|
no
|
n.a.
|
|
|
54
|
incidental findings
|
4th ventricle
|
left
|
15 × 15
|
sharp
|
inhomogeneous
|
iso
|
iso
|
centrally hypo
|
hyper, centrally hypo
|
none
|
|
n.a.
|
no
|
n.a.
|
|
|
84
|
incidental findings
|
lat. ventricle
|
right
|
7 × 10.3
|
sharp
|
inhomogeneous
|
iso
|
iso
|
hyper
|
hyper to iso
|
none
|
|
constant
|
no
|
n.a.
|
|
|
53
|
incidental findings
|
foramen magendi 4th ventricle
|
|
10 × 13.8
|
irregular sharp
|
inhomogeneous
|
iso
|
iso
|
hypo
|
hyper
|
none
|
|
postoperatively constant
|
no
|
n.a.
|
full resection
|
WHO I
|
48
|
incidental findings
|
foramen magendi 4th ventricle
|
bilateral
|
26.8 × 30.7
|
sharp
|
inhomogeneous
|
iso
|
iso
|
n.a.
|
hyper
|
partially punctate
|
|
postoperatively constant
|
no
|
n.a.
|
full resection
|
WHO I
|
63
|
incidental findings
|
foramen magendi 4th ventricle
|
|
11.3 × 14.2
|
sharp
|
inhomogeneous
|
iso
|
iso
|
n.a.
|
hyper
|
none
|
|
2 months constant
|
no
|
n.a.
|
|
|
87
|
incidental findings
|
foramen magendi
|
|
12.9 × 14
|
sharp
|
inhomogeneous
|
iso
|
iso
|
n.a.
|
hyper to iso
|
inhomogeneous
|
strong T2 hyper, FLAIR iso
|
since 2007 constant
|
no
|
n.a.
|
|
|
49
|
incidental findings
|
4th ventricle
|
|
9 × 12
|
sharp
|
inhomogeneous
|
iso
|
iso
|
hyper
|
hyper
|
none
|
|
constant
|
no
|
n.a.
|
–
|
|
On the whole, there were no MR or CT morphological differences between the “pure”
subependymomas and the tumors consisting of the mixed form of subependymoma and ependymal
cell clusters.
Clinical presentation
In our retrospective study, 4 patients presented with clinical symptoms of headache,
nausea and dizziness due to supratentorial cerebrospinal fluid accumulation caused
by the tumor. In these cases, the infratentorial tumor was located in the fourth ventricle
with transition into the foramen Magendie in only one instance; in 3 cases the mass
was located in the lateral ventricle. With regard to the clinical symptoms in the
other patients, we have only limited information based on neurosurgeon reports; in
29 cases, the lesion was discovered by chance. Indications for MRI here were a clarification
of various, mainly nonspecific, symptoms such as undirected dizziness, occasional
headache, paresthesia, hypoaesthesia and occasional visual disturbances (double vision,
flicker, blurred vision), and in another patient there was a gait disturbance without
evidence of CSF accumulation, so a connection between symptoms and tumor was not clear
on the whole. In the remaining 4 patients, the indication for imaging could not be
clearly determined in hindsight.
Discussion
Subependymomas are rare, low-grade (World Health Organization WHO grade 1) glial neoplasia
with an ependymal origin. Their rarity is reflected in the limited number of reports
in the scientific literature and low number of cases [3]
[12].
In this study, we present a relatively large group of 33 patients with typical imaging
subependymomas, of which 4 were histopathologically associated with the mixed special
form, and in addition we describe their imaging properties.
In the current WHO classification of 2016 [13], there is, in addition to subependymomas and ependymomas, a special form of mixed
subependymomas/ependymomas for which the ependymal components are evaluated according
to the WHO classification [1]
[3]. Prognostic differences between the subependymomas and mixed subependymomas/ependymomas
are still unknown and could not be determined in the previously largest cohort of
Rushing et al. [3].
Consistent with the available data in the literature, a hypo-isointense signal behavior
of the subependymomas was observed in T1 weighting for the brain parenchyma in all
cases of our patient data [3]
[10]
[14]. For example, in 12 cases of MRI examinations, Rushing et al. and Jain et al. described
iso- to hypointensity to the cortical brain parenchyma in T1w, hyperintensity on T2w
and a predominant gadolinium enrichment in 80 % of the reviewed cases [3]
[14]. However, this observation of the high number of contrast-enhancing subependymomas
does not correspond to the majority of the morphological properties of subependymomas
described in the literature. In this regard, a wide range of variations has been previously
described, ranging from slight [9]
[10]
[15]
[16]
[17] to strong and irregular enhancement [18] as well as individually described and controversially discussed incidence of accumulations
[19], in some cases also depending on tumor localization [12]. In our cases, partially nodular enhancement was predominantly evident; rarely an
inhomogeneous contrast enhancement was observed. Occasionally a small area enrichment
was seen. No contrast medium absorption occurred in 15 other tumors, and in 3 cases
there were no contrast-based sequences. There was no morphological difference between
the subependymomas and the histologically proven mixed subependymomas/ependymomas
[7]
[12], in particular there was no difference in opacification behavior.
In 1997 Maiuri et al. described the slowly progressive growth of subependymomas in
40 % of tumors. For the most part, subependymomas are detected as incidental findings
in MRI; this was also true in our cohort. The presence of clinical symptoms is basically
dependent on tumor location. If clinical symptoms occur, this is almost always due
to cerebrospinal fluid accumulation; focal neurological abnormalities or epileptic
seizures due to the subependymoma were not described by Maiuri et al. and were also
not evident in our cohort. Localizations that more frequently lead to cerebrospinal
fluid involve tumors on the septum pellucidum in the lateral ventricular anterior
horn or posterior horn and interventricular foramen, in contrast to tumors in the
cella media of the lateral ventricle or in the fourth ventricle, which relatively
rarely lead to a corresponding symptomatology [8]
[9]. This could be based on the fact that subependymomas are usually small which in
fact may result in displacement of the interventricular foramen; in the fourth ventricle,
however, the three CSF drainage possibilities rarely lead to decompensation by blocking
all three foramina. Furthermore, in the course of our study, in some cases tumor size
constancy over a longer period was found in 25 of 33 cases, including 9 completely
or partially resected tumors without recurrence or progression of the residual tumor.
In 2 cases, observed retroactively since the first available study, size constancy
was evident over a period of 10 years; in 7 cases size remained consistent over the
entire recorded evaluation period of 8 years. In 16 other cases, constancy was observed
over shorter periods of a few years or months. In the remaining 8 cases there was
only a single examination; thus progression evaluation was not possible. Likewise
in tumors with mixed pathology of a subependymoma with ependymal component, only a
partial resection was performed in 2 cases, and no size constancy could be observed
in the postoperative course of several months. A macroscopically complete resection
is possible in most cases, due to the predominantly sharp delimitation of the subependymomas
in relation to the adjacent brain parenchyma and intraventricular localization [9]
[14]; in our cases of partial resection, this was restricted due to adhesions on the
adjacent brain stem or inaccessible location in the foramina Luschka. With respect
to patients with partial resections, we only had short-term follow-up intervals; Rushing
et al. however found no significant difference in survival based on the extent of
resection in their 83-patient cohort [3].
There is no consensus in the literature regarding growth tendency; most sources, particularly
older, describe mixed tumors as having a particularly strong growth tendency corresponding
to the ependymal cell component [7]
[12]
[18]
[20]
[21], whereas a more recent study by Rushing et al. [3] could not establish a correlation between survival rate and mixed tumors, so that
in typical image findings (see below) even though the presence of ependymal cell clusters
cannot be ruled out, confirmation of such a mixed tumor may not have any further consequence
than would be the case of a subependymoma. Morphologically, the lesions in our cohort
primarily exhibited sharp demarcations (96.9 %); 34.4 % demonstrated an irregularly
sharp delineation, similar to what has already been described [7]
[10]
[18]. Cystic interior elements could be observed in 4 cases of our population, just as
described in individual studies [18]. Partial T2* signal reduction as a sign of calcification or hemosiderin deposits
were evident in 9 of 33 cases (it should be noted that no T2* sequences of the tumor
region were available for 15 of the 33 patients); both calcifications and hemorrhages
have already be described for individual cases [10]
[12]
[16]
[18]
[22]. In addition, no calcifications could be determined in the 5 available CT examinations.
In our cases, we could not observe exclusive occurrence of calcifications in infratentorial
tumors as described by Chiechi et al.; however more infratentorial than supratentorial
tumors with calcifications were observed. Six of the cases with T2* signal voids concerned
infratentorial tumors, and 3 concerned supratentorial masses. There was decreasing
incidence of tumor localization in the fourth ventricle (54.5 %), in the lateral ventricles
(33.3 %), predominantly in the hind horns, and occasionally in the cisterna magna
(3 %) or exclusively the foramen Magendie (9 %) corresponding to the frequency distribution
in medical literature [4]
[12]; this was analogously similar to the frequency distribution noted by Ernestus et al.
with 58 % of tumors in the fourth ventricle, and 38 % in the lateral ventricle [2], or the frequency distribution observed by Smith et al. with tumor localization
in the fourth ventricle in 50 – 60 % of cases, and 30 – 40 % in the lateral ventricles
[18]. Similar to the more commonly observed infratentorial ependymomas, the mixed tumors
in our study with subependymoma and ependymoma components primarily demonstrated an
infratentorial location; occurrence was in the lateral ventricle in only a single
case.
Corresponding to data in the literature, our cohort reflected a distribution of two-thirds
males to one-third females affected by subependymomas [2]
[18]. The mean age of onset, 58.6 years of age, affects the middle-aged and elderly population,
which likewise corresponds to the distribution frequency among the elderly described
in the literature [2]
[3]
[9]
[10]
[12]; wide variations were observed in our cohort. We did not observe individual mixed
tumors among younger patients as described in the literature [20]
[21]
[23].
Essential differential diagnoses for intracranial subependymomas should include ependymomas,
medulloblastomas, astrocytomas, central neurocytomas and meningiomas. Central neurocytomas
and ependymomas should be considered in the area of the interventricular foramen;
whereas in the fourth ventricle differential diagnosis should take in to account medulloblastomas
and choroid plexus papillomas (CPP) [15]. Refer to [Table 2] for the main morphological imaging differentiation options [24]
[25].
Table 2
Overview of differential diagnoses.
ependymomas
|
-
younger patients
-
heterogeneous, CM enhanced lesions with edema
-
typically in 4th ventricle with hydrocephalus
-
supratentorial frequently with parenchymal expansion
|
medulloblastomas
|
-
slight to moderate CM enhancement
-
areas with cystic and necrotic degeneration (among children strong CM enhancement, no necrotic areas)
-
generally among patients < 10 years, second peak age, 20 – 40 years
|
subependymal giant cell astrocytomas
|
-
moderately CM enhanced lesion in interventricular foramen
-
frequently calcification
-
(patients with tuberous sclerosis, subependymal nodes, cortical glioneural hamartoma
and medullary layer lesions)
|
zentral neurocytomas
|
-
young patients
-
typical cystic appearance
-
lesion in lateral ventricle with relation to septum pellucidum or interventricular
foramen
-
slight to moderate CM enhancement
-
frequently calcification
|
meningiomas
|
|
choroid plexus papillomas (CPP)
|
-
typically pediatric tumors of the lateral ventricle
-
frequently in 4th ventricle among adults
-
CM enhanced papillary lesions
-
Frequently hydrocephalus
|
hemangioblastomas
|
|
metastases
|
-
primary tumor known as a rule
-
frequently numerous lesions in the borderline between the marrow and cortex
-
frequently intraventricular inclusion of the choroid plexus
|
cavernous malformations
|
|
-
Subependymomas are rare, benign, predominantly asymptomatic, intraventricularly localized
glial tumors of ependymal origin with no growth tendency with typical morphological
characteristics that distinguish them from tumors of other entities and with different
dynamics in the same localization.
-
It is of particular importance to know this fact, as the suspicion of the chance finding
of a subependymoma justifies a wait-and-see attitude and imaging follow-up at greater
intervals, whereas other tumors in the same location may require timely invasive investigation
and, if necessary, resection.
-
The even rarer special form of a histopathological mixed picture of subependymoma
and ependymal components does not differ with regard to the image-based morphological
criteria.
-
Since WHO II tumor components with a somewhat stronger growth tendency can be present,
imaging follow-up monitoring is in every case indispensable.