Journal of Neuroradiology (2013) 40, 101—111
Available online at
www.sciencedirect.com
ORIGINAL ARTICLE
Central neurocytoma: Radiological and clinico-pathological findings in 18 patients and one additional MRS case Haleena Ramsahye a, Huijin He a,∗, Xiaoyuan Feng a, Siyao Li a, Ji Xiong b a b
Department of Radiology, Huashan Hospital, Fudan University, Shanghai 200040, People’s Republic of China Department of Pathology, Huashan Hospital, Fudan University, Shanghai 200040, People’s Republic of China
KEYWORDS Central neurocytoma (CN); Apparent diffusion coefficient (ADC); Normalised apparent diffusion coefficient (NADC); Axial fluid attenuated inversion recovery (FLAIR); Magnetic resonance spectroscopy (MRS)
∗
Summary Objectives: To evaluate clinical findings and radiological characteristics of central neurocytoma (CN) in 18 patients and magnetic resonance spectroscopy (MRS) features in one additional case. Materials and methods: Clinical and imaging findings of 18 patients (nine female and nine male; age range, 18—37 years old (27.8 ± 5.7)) with histopathological diagnosis of CN were evaluated retrospectively. Eight patients underwent CT and eight had MR imaging. Both MR and CT images were acquired for other two patients. We also assessed the tumour NADC values. Clinical data, such as presenting symptoms and medical histories were collected. MRS was also obtained for one additional case. Results: Clinical symptoms at the time of presentation were headaches (n = 11), dizziness (n = 6), visual disturbances (n = 2), etc. Eight lesions were unilateral ventricle (44%) and ten were located in both lateral ventricles. Three tumours continued towards the foramen of Monro and one to the third ventricle. The maximum diameter of the CNs varied from 3.4 to 9.2 cm (5.2 ± 1.5 cm). On CT, diffuse and diverse calcifications were observed in nine cases and cysts varying in sizes were revealed in all. On MRI, the solid parts of the tumours were mainly hypoto isointense on all T1WI and isointense to grey matter on T2WI. Clusters of cysts gave the tumours a ‘‘swiss cheese/soap bubble’’ inhomogeneous hyperintense appearance on T2WI and FLAIR images. Heterogeneous moderate enhancement (5/8) was present on T1 postcontrast images. On DWI, the tumours had heterogeneous hyperintense appearances and the tumour NADC values were 0.93 ± 0.21.On MRS, elevated Cho and Gly peaks and reduced Cr and NAA peaks were obtained.
Corresponding author. Tel.: +86 21 52 88 83 47. E-mail address:
[email protected] (H. He).
0150-9861/$ – see front matter © 2012 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.neurad.2012.05.007
102
H. Ramsahye et al. Conclusion: CN is almost exclusively located in the body of lateral ventricle in young adults. It is discovered due to symptoms of raised intracranial pressure. The distinct radiological features such as: (1) diffuse and diverse calcifications on CT images; (2) clusters of cysts of varying sizes resulting in the ‘‘swiss cheese/soap bubble’’ appearance on T2WI and heterogeneous moderate enhancement on MR images; (3) the incorporation of the septum pellucidum in bilateral tumours and abutting of the septum pellucidum in unilateral tumours together with the attachment of the wall of the ventricles can help in the diagnosis of preoperative central neurocytoma. © 2012 Elsevier Masson SAS. All rights reserved.
Introduction Central neurocytoma (CN) was originally described by Hassoun et al. in 1982, and became a well-defined clinical and pathological entity [1,2] by the early 1990s. The first report resulted in increased recognition of the tumor, and a number of reports have since been published in the literature. Nevertheless, CN remains a rare neoplasm of the central nervous system, representing 0.25—0.5% of primary intracranial tumors and nearly 10% of intraventricular tumors [3—5]. CN typically arises in young adults in their third decade of life. Almost 69% of described cases occurred in patients between 20 and 40 years of age, with an average age of 30 years [5,6]. However, it can, on rare occasions, arise in the first decade of life and after 50 years of age [7]. CN is a tumor of neuronal origin [8—10] that is preferentially located in the ventricular system, mainly in the frontal horn of the lateral ventricles and in the region of the foramen of Monro [11,12]. Histologically, CN corresponds to World Health Organization (WHO) grade II, and its International Classification of Diseases for Oncology (ICD-O) behavior grade is 1 (low or uncertain malignant potential, or borderline malignancy). Its diagnosis and management remain controversial, as it may be confused radiologically with oligodendroglioma, ependymoma, subependymal giant cell astrocytoma and choroid plexus papilloma. WHO grade II astrocytoma and oligodendroglioma have an ICD-O behaviour grade of 3 (malignant) and, thus, reinforce the fact that CN has a better prognosis compared with other intra-axial brain tumors presenting in the same age group. The present study represents the largest imaging series of CN so far, and was conducted to evaluate the clinical findings and to define the characteristic imaging features of the tumor to enable more accurate diagnosis of suspicious lesions before surgery.
Materials and methods The present retrospective study was conducted using the pathology archives at our hospital with the approval of the review board and ethics committee of our institution. From January 2007 to July 2010, a series of 44 histopathologically confirmed CN cases was collected. The histological diagnosis was based on the tumor sample harvested intraoperatively and confirmed by immunohistochemistry. Of these 44 patients, only 18 underwent preoperative conventional radiological procedures at our hospital. The evaluated patients consisted of nine women and nine men, with an age range of 18 to 37 years (average: 27.8 ± 5.7 years) at the time of observation. Eight patients had
undergone computed tomography (CT) with and without intravenous injection of iodinated contrast, and eight had undergone magnetic resonance imaging (MRI). Both MRI and CT scans were acquired for two other patients. Of the 44 cases, one patient, a 24-year-old woman, also underwent magnetic resonance spectroscopy (MRS) at our hospital. MRI was performed on either a 1.5-T scanner (Signa EXCITE HD 1.5TTM TwinSpeed; GE Healthcare Bio-Sciences, Little Chalfont, Buckinghamshire, UK), a 3.0-T device (Signa VH/i 3 T/94) or a 3.0-T MAGNETOM Verio (Siemens AG, Erlangen, Germany). Axial and sagittal non-enhanced, T1weighted, spin-echo [SE] images were obtained in the 10 patients who underwent MRI. Additional unenhanced images were available in eight patients, and included axial T2-weighted and axial fluid-attenuated inversion recovery [FLAIR] sequences. Subsequently, axial Gd-DTPAenhanced (gadolinium-diethylenetriamine pentaacetic acid) T1-weighted SE images were also acquired in these eight patients. Furthermore, diffusion-weighted imaging (DWI) was performed using axial, multislice, single-shot, echoplanar SE sequences in seven patients. Apparent diffusion coefficient (ADC) maps were calculated on a pixel-by-pixel basis. For an exact demonstration of tumor heterogeneity, at least five uniform regions of interest 10—20 mm2 in size were selected from different areas of non-cystic/non-calcified parts of tumor with contrast enhancement and from normallooking parietal white matter. Minimum ADC values were taken into consideration. In addition, all measured tumor ADC values were divided by normal ADC values to obtain normalized ADC (NADC) values. For MRS, spectral and metabolite maps for each ‘slice’ along the third dimension were extracted using the FuncTool Display. Within the obtained volume of interest (VOI), separate voxels were individually placed in different parts of the tumor. The metabolite peaks used were: N-acetylaspartate (NAA) at 2.02 ppm; choline-containing compounds (Cho) at 3.22 ppm; (phospho-) creatine (Cr) at 3.01 ppm; and glycine (Gly) at 3.55 ppm. Metabolite ratios (maximum Cho/Cr and Cho/NAA ratios) were calculated from the metabolite maps using the relevant software. CT was performed in 10 patients, using a SOMATOM Emotion 6 CT scanner (Siemens). Routine head CT scans were done, resulting in 12 slices, each of which was 5-mm thick. Two experienced radiologists (H.R. and H.H.) reviewed all of the imaging data retrospectively for location, size, margins, density, signal intensity and enhancement characteristics, and meticulously examined the images for the presence of calcifications, cysts and necrotic changes in the lesions. The presence or absence of hemorrhage, vascular signal voids, ventricular dilatation and adjacent
Central neurocytoma: radiological and clinico-pathological findings
103
Table 1
Findings of central neurocytoma on magnetic resonance imaging (MRI) and clinical examination.
Patient no.
Gender/age (years)
Clinical presentation
Location
T1-weighted images
1
M/33
Right lateral ventricle
Isointensity to hyperintensity
2
M/28
Headache and dizziness for 3 months VI for 1 month
Heterogeneous hypointensity
3
F/22
Both lateral ventricles and third ventricle Both lateral ventricles
4
M/32
5
F/30
Both lateral ventricles Right lateral ventricle
Heterogeneous hypointensity Heterogeneous with slight hyperintensity
6
M/22
Right lateral ventricle
7
F/28
8
M/29
VI for 3 weeks
Both lateral ventricles
9
F/35
Headache for 2 years
Both lateral ventricles
10
F/23
Headache, vomiting and nausea for 1 week
Right lateral ventricle
Heterogeneous with hyperintensity Isointensity with patchy hypointensity Isointensity with patchy hypointensity Hypointensity with patchy hyperintensity Heterogeneous hypointensity
Dizziness and balance problems for 1 month Dizziness for 3 years Headache for 3 years and vomiting for 1 month Nausea and vomiting for 9 days Headache
Both lateral ventricles
Hypointensity with patchy isointensity
T2-weighted images
Hyperintensity
Isointensity with patchy hyperintensity Isointensity with patchy hyperintensity Isointensity with patchy hyperintensity
FLAIR
Enhancement
Hyperintensity with slight isointensity Heterogeneous hyperintensity
Heterogeneous
Moderate
Heterogeneous hyperintensity
No apparent enhancement
Heterogeneous hyperintensity
Mild heterogeneous
Heterogeneous hyperintensity
Moderate heterogeneous
Slight hyperintensity
Moderate heterogeneous
Hyperintensity
Moderate heterogeneous
Hyperintensity
Moderate heterogeneous
VI: visual impairment; M: male; F: female; FLAIR: fluid-attenuated inversion recovery sequence.
parenchymal edema were also evaluated. Cystic tumor components were defined as areas that appeared hypointense on T1-weighted images (T1WI), hyperintense on T2WI and hypointense on FLAIR sequences.
Results Clinical features The present cohort comprised 18 patients (nine female and nine male), ranging in age from 18 to 37 years, with an average age of 27.8 ± 5.7 years. Clinical symptoms at the time of presentation included headaches (n = 11), dizziness (n = 6), visual disturbances (reduced visual acuity, diplopia, blurred vision; n = 2), and nausea and vomiting (n = 4). Other observed signs included epilepsy, neuropsychological disorders and balance problems. The duration of clinical
symptoms prior to diagnosis ranged from a few weeks to 17 years.
Tumor size and location Tumor location was unilateral ventricular in eight cases (44%) and bilateral ventricular in ten cases, and almost all tumors were exclusively located in the anterior part of the lateral ventricle. Three extended into the foramen of Monro, and one reached the third ventricle. The maximum diameter of the CN lesions ranged from 3.4 cm to 9.2 cm (average: 5.2 ± 1.5 cm). The tumors involving both lateral ventricles were medium-sized to large, and incorporated the septum pellucidum. Of the eight unilateral ventricular cases, where the tumors were relatively smaller in size, the lesions appeared to have a broader base in relation to the wall of the lateral ventricle (subependymal layer). The contour of the septum pellucidum was preserved with
104 Table 2
H. Ramsahye et al. Findings of central neurocytoma on computed tomography (CT) and clinical examination.
Patient no.
Gender/age (years)
Clinical presentation
Location
CT attenuation of solid parts
Calcifications
Hydrocephalus
2
F/22
Dizziness and balance problems for 1 month
Hyperdensity
Patchy/circular
Present
4
M/32
Dizziness for 3 years
Present
M/29
Incidental finding
Slight hyperdensity Hyperdensity
Coarse/popcorn
11
Coarse/popcorn
Present
12
F/26
Epilepsy for 17 years
Present
M/33
Coarse/patchy
Present
14
F/33
Hyperdensity
Coarse
Present
15
F/26
Punctate
Present
F/17
Present
M/18
Slight hyperdensity Hyperdensity
Punctate
17
No calcification
Present
18
M/37
Left lateral ventricle Left lateral ventricle Both lateral ventricles Right lateral ventricle
Hyperdensity
16
Headache and dizziness for 20 days Headache, dizziness and vomiting for 2 weeks Headache for 3 months Headache and dizziness for 1 year Headache for 2 years
Slight hyperdensity Hyperdensity
Small circular
13
Both lateral ventricles and third ventricle Both lateral ventricles Both lateral ventricles Left lateral ventricle Both lateral ventricles Both lateral ventricles
Hyperdensity
Small circular/stripey
Present
Headache for 10 years
M: male; F: female.
partial adherence to the tumor, but was deviated to the contralateral side from the ipsilateral ventricle, which was more enlarged and where the main body of the tumor was located. The septum pellucidum delineated the medial margin of tumors located in the lateral ventricle.
CT imaging features In the ten cases with CT scans, diffuse and diverse calcifications were observed in nine cases, with scattered coarse and/or ‘popcorn’-like calcifications present in four patients, and smaller circular or punctate calcifications in five. Areas of hypodensity representing cystic degeneration of variable sizes were revealed in all ten cases. The solid parts of the tumors were hyperdense in all cases, and the presence of hemorrhage was evident in two.
revealed after enhancement. Out of all 18 cases, four also showed adjacent brain parenchymal edema, while hydrocephalus was present in all cases. The CT and MRI features of the lesions explored are summarized in Tables 1 and 2. On DWI, the tumors had heterogeneous hyperintense appearances. ADC maps demonstrated heterogeneous hypointensity and isointensity. When compared with the contralateral parietal lobe white matter, the tumor NADC value was 0.93 ± 0.21 (Table 3).
Table 3 Apparent diffusion coefficient (ADC) values for central neurocytoma tumors and parietal white matter, and normalized ADC values (NADC). Patients Age (years)/gender
MR imaging features On MRI, the tumors were mainly non-homogeneously hypointense on T1WI, while solid parts of the tumors were isointense to gray matter on T2WI. Areas of heterogeneous intensity corresponded to cysts, intratumoral vessels and calcifications. The clusters of cysts of varying sizes gave the tumors a ‘‘swiss-cheese/soap-bubble’’ non-homogeneous appearance on T2WI and FLAIR images. Heterogeneous moderate enhancement (in five of eight patients) was present on contrast-enhanced T1WI and was related to the level of vascularity. Clearer images of the cystic clusters were also
33/M 28/M 22/F 30/F 28/F 35/F 23/F M: male; F: female.
Diffusion-weighted imaging Tumor (ADCT )
Normal white matter (ADCN )
NADC
0.000605 0.000722 0.000707 0.000662 0.000992 0.000732 0.000977
0.00087 0.000871 0.000775 0.000922 0.000767 0.000773 0.000856
0.695402 0.828932 0.912258 0.718004 1.293351 0.94696 1.141355
Central neurocytoma: radiological and clinico-pathological findings
105
Figure 1 Photomicrograph and immunohistochemical staining of CN: a) hematoxylin and eosin (H&E) staining shows acellular zones of neurophilic islands (black arrow; × 400); (b) synaptophysin staining is positive in both the cytoplasm and neurophil matrix (× 400); (c) neuronal nuclear antigen (NeuN) is positive (× 400); and (d) glial fibrillary acidic protein (GFAP) is also positive (white arrow). Calcification (black arrowhead) is also present.
Magnetic resonance spectroscopy features The MRS images (Fig. 1) exhibited an elevated Cho peak, and reduced Cr and NAA. On the other hand, increased Cho/Cr and Ch/NAA ratios were obtained. A Gly peak at 3.55 ppm was also noted in VOIs 2—4 (Fig. 1) of the tumor parenchyma.
Pathological and immunohistochemical features The gross pathology of CN was that of a lobulated, wellcircumscribed, gray friable mass. Necrosis, calcifications and cyst formation were also frequently seen. Immunohistochemical findings of synaptophysin and neuronal nuclear antigen (NeuN) were positive in all cases, whereas coexpression of glial fibrillary acidic protein (GFAP) was noted in five out of 11 patients. A photomicrograph and immunohistochemical staining of CN are shown in Fig. 2.
Discussion Central neurocytoma had long been overlooked, due to its macroscopic similarity to various other intraventricular lesions (such as oligodendroglioma, low-grade astrocytoma and ependymoma), until its first description in 1982 by Hassoun et al. [1]. Since then, more than 500 cases have been reported in the literature [13], and several studies are
ongoing to better understand its neuronal nature, biological behavior and variety of treatment modalities. In the present study, a series of 44 histopathologically confirmed CN cases was collected between January 2007 and July 2010. Such a large number during that period of time was related to the fact that our hospital is considered one the best hospitals for neurosurgery in the country and, as a consequence, many complicated and rare lesions, including CN, are dealt with at our hospital. Less complicated cases are sometimes referred to other institutions. In this series, the epidemiological features of the study population corresponded to those previously published [5—7,14]. In the main, patients were young adults with an average age of 27.8 ± 5.7 years, and both genders were equally affected [15,16]. CN is typically discovered because of symptoms of raised intracranial pressure secondary to obstructive hydrocephalus [10,15,16]. Other signs, such as decreased visual acuity with papillary edema, memory problems, walking disorders, hormonal dysfunction, neurological deficits and/or seizures may also be observed [3,10]. The latter two symptoms generally also present with signs of tumor aggressivity [10]. In the present study, one of our patients had a history of seizures over 17 years, while the others presented with symptoms of raised intracranial pressure. The histopathological appearances of CN can resemble those of an oligodendroglioma. Both neoplasms
106
H. Ramsahye et al.
Figure 2 An 18-year-old girl with central neurocytoma in the left lateral ventricle: (a) non-enhanced and (b) enhanced CT scans. Areas of hypodensity represent cystic degeneration.
Figure 3 Calcification patterns in central neurocytoma (CN): (a) coarse calcification; (b) ‘popcorn’ type of calcification (white arrow); and (c) small circular and punctate calcifications.
Central neurocytoma: radiological and clinico-pathological findings
107
Figure 4 A 22-year-old woman with central neurocytoma in both lateral ventricles with propagation into the third ventricle: (a) non-enhanced CT scan; (b) precontrast sagittal T1WI; (c) post-contrast sagittal T1WI; (d) precontrast axial T1WI; (e) FLAIR sequence; and (f) post-contrast axial T1WI. A slightly mixed-density mass with coarse calcifications (white arrow) can be seen on CT. The tumor exhibits a heterogeneous signal due to calcification and cystic lesions (black arrow), and mild heterogeneous enhancement with contrast.
comprise uniformly small cells with rounded nuclei and scant cytoplasm resembling perinuclear halos (‘fried-egg’ appearance). Many intraventricular tumors diagnosed as oligodendroglioma may, in fact, represent CN [7,16—18]. The definitive diagnosis is made by immunohistochemistry, which can confirm the neuronal origin of the tumor through the expression of synaptophysin, neuron-specific enolase (NSE), microtubule-associated protein 2 and GFAP [10,14,18]. Oligodendrocyte transcription factor 2 (Olig2) can differentiate oligodendroglioma and astrocytoma from CN. In the present series, NeuN appeared to be sensitive and specific for the diagnosis of CN (Fig. 2) [18]. Coexpression of GFAP was also found in five out of 11 patients, thereby supporting the theory that CN originates from bipotential (neuronal and astrocytic) progenitor cells in the subependymal plate of the lateral ventricle [19] or from circumventricular organs [19,20]. The neuroradiological features are crucial for revealing the various characteristics of CN. The tumors are usually well circumscribed, and typically located in the anterior half of the lateral ventricle [21] in close proximity to the region of the foramen of Monro [1,12,16]. As also reported in previously published papers, in our series, attachment to the septum pellucidum appeared to be a more frequent feature of these tumors, whereas their propagation into the third ventricle was less common (Fig. 3). Only 3% of CN are
found in the third ventricle as an isolated location [4,22], and the fourth ventricle is an even rarer location. In the present study, it was observed that both the attachment and incorporation of the septum pellucidum were usually associated with medium-sized to large bilateral tumors. On the other hand, of the eight unilateral ventricular cases, where the tumors were relatively smaller, the lesions appeared to have a broader base in relation to the wall of the lateral ventricle (subependymal layer). Radiological images revealed that the contour of the septum pellucidum was preserved with partial tumor adherence, but was deviated to the contralateral side from the more enlarged ipsilateral ventricle, where the main body of the tumor was located. The septum pellucidum delineated the medial margin of the tumor located in the lateral ventricle. It appeared that, as the CN increased in size and spread to both lateral ventricles, it adhered to and ultimately incorporated the septum pellucidum. This characteristic feature supports the idea that CN originates postnatally from remnants of the subependymal germinal plate of the lateral ventricles and from the bipotential progenitor cells present in the periventricular matrix of the mammalian brain [23—25]. CT imaging of CN typically demonstrates mixed solid—cystic components. Areas of hypodensity represent cystic degeneration, while solid parts of the tumors are isointense or hyperdense, and contain calcifications that
108
H. Ramsahye et al.
Figure 5 T2WI from different patients show the non-homogeneous ‘soap-bubble’ appearance of CN as a result of clusters of cysts. (a) a fluid level within one of the right-sided cysts (black arrow) probably indicates hemorrhage into the cyst. Parenchymal edema within the white matter adjacent to the tumor in the right cerebral hemisphere is also evident; (b) A signal void (thin white arrow) is present; (c) The FLAIR sequence demonstrates a non-homogeneous ‘swiss-cheese’ appearance of the lesion due to the presence of smaller scattered cysts.
Figure 6 Unilateral ventricular lesions: (a) post-contrast T1WI and (b) precontrast T1WI reveal a small central neurocytoma (CN) with a broader base in relation to the wall of the lateral ventricle. The well-preserved contour of the septum pellucidum can be seen deviated to the contralateral side from the ipsilateral ventricle, where the lesion is located. (c) A large CN has incorporated the septum pellucidum.
Central neurocytoma: radiological and clinico-pathological findings
109
Figure 7 A 22-year-old woman with CN in both lateral ventricles (b = 1000): (a) isotropic DWI shows a heterogeneous signal that is predominantly higher than that of the cerebral cortex; and (b) the ADC map confirms slightly restricted diffusion.
are usually nodular and scattered in around 51% of cases [9,11,12]. In the present series, of the ten cases with CT scans, diffuse and diverse calcifications were observed in nine cases, with scattered coarse and/or popcorn-like calcifications seen in six and smaller circular or punctate calcifications noted in three (Fig. 4). Contrast enhancement was mild to moderate and heterogeneous for most CN lesions (Fig. 5). Also, hemorrhage, which is usually rare in CN, was observed in two cases. Hemorrhage is a useful sign for differentiating CN from other intraventricular tumors that have less of a tendency to bleed [12,26]. The depiction of the tumor on MRI is analogous and allows better delineation of the lesion. Signal intensities are usually non-homogeneously low on unenhanced T1WI. Solid parts of the tumors are isointense to gray matter on T2WI and FLAIR sequences, with areas of heterogeneous intensity corresponding to cysts, intratumoral vessels and calcification, thus resulting in the classical soap-bubble appearance [27,28]. After intravenous injection of gadolinium, variable low-to-medium enhancement can be seen related to the degree of vascularity [10,25]. The features observed in our present study corresponded to those mentioned above (Fig. 6). Clusters of cysts of varying sizes gave the tumor a non-homogeneous swiss-cheese/soap-bubble appearance on T2WI and FLAIR in 9/10 images (Fig. 7). Moderate heterogeneous enhancement was seen on post-contrast T1WI in 7/10 cases. Clearer images of the cystic clusters were also seen after enhancement. However, it was difficult to characterize intralesional calcifications on MRI because of its limited capacity. Indeed, intralesional calcification, considered a common finding and an important diagnostic sign [15], was more easily identified on CT scans. Sophisticated imaging methods for obtaining physiological and biological information can improve the diagnostic efficacy of MRI in brain tumor differentiation. DWI can help to assess their diffusion characteristics, thereby giving an indication of their malignancy. NADC values were significantly helpful for grading benign and malignant tumors as low and high grade, respectively (Fig. 8) [29]. A high degree
of malignancy is associated with decreased NADC values (more restricted diffusion), whereas low-grade tumors have increased NADC values [29,30]. In the present series, our NADC value of 0.93 ± 0.21 was lower than those for astrocytomas and non-astrocytic glioma [29,31]. These findings suggest that increased cellularity constitutes a relative barrier to water diffusion and is, therefore, responsible for the decreased ADC and NADC values. MRS may also be important for estimating changes in metabolites, including Cho-containing compounds, NAA, Cr, Gly, lactate and lipid, found in intracranial neoplasms [32—35]. According to previous reports using MRS, CN has typical patterns, including elevated Cho and Cho/Cr, and decreased Cr and NAA, patterns similar to those observed in our present MRS case. A few studies [32,34,36] have also reported an increase in Gly at 3.55 ppm; however, according to Chuang et al. [37], the presence of a Gly peak may suggest a diagnosis of CN, but the absence of Gly does not exclude a CN diagnosis. A Gly peak was also noted in our case, which suggests that the sensitivity and specificity of a Gly peak at 3.55 ppm need to be confirmed in larger study populations. The Cho peak is thought to represent increased membrane phospholipid biosynthesis: Cho originates from phosphocholine, glycerophosphocholine and acetylcholine, all of which are involved in membrane synthesis and degradation. Higher Cho levels are seen in more aggressive types of tumors [37]. A decreased NAA signal, which is localized in neurons, indicates the loss of intact neuronal cells [37]. The neuroradiological differential diagnosis for CN includes many tumors and, thus, is fairly complicated. Choroid plexus papilloma, for example, is more common in children than in young adults; its clinical manifestation starts early, with the diagnosis made before 5 years of age in 85% of patients. It usually has a rough, irregular surface and tends to involve the atria of the lateral ventricles. On MRI, it is heterogeneous with lobules, giving it a cauliflower-like appearance that differentiates it from CN. However, the tumor usually has marked homogeneous enhancement after intravenous administration of contrast [11,18]. In contrast,
110
H. Ramsahye et al.
Figure 8 A 24-year-old woman with central neurocytoma (CN) of both lateral ventricles. Magnetic resonance spectroscopy (MRS) demonstrated increased Cho resonance, reduced Cr and NAA in VOIs 1—4. Increased Gly in VOIs 2—4 was also observed, and increased Cho/Cr and Ch/NAA ratios were obtained.
intraventricular meningioma is more common in elderly patients, especially older women; it is typically found in either the trigone of the lateral ventricles, with no close spatial relationship to the septum pellucidum, or in the fourth ventricle. On CT, calcifications are commonly seen and strongly enhanced with contrast [11,38]. Ependymoma are mainly seen in the fourth ventricle. A supratentorial location is reported in one-third of patients, commonly with extension into the periventricular white matter. These tumors are found during childhood [39]. Subependymoma often has weak or no enhancement on post-contrast CT scans and MRI. It may be difficult to differentiate from CN, although one clue to a proper diagnosis may be its incidence in the fifth decade [40,41]. Subependymal giant cell astrocytoma frequently affects young patients with tuberous sclerosis. These subependymal nodules vary in size, and cortical tubers are easily found on CT and MRI, which is the key factor for differentiating them from CN [31,42]. These nodules may also be partially or completely calcified and may be enhanced with contrast. Intraventricular oligodendroglioma is usually located within the body of the lateral ventricle. These lesions are usually hyperattenuated on CT compared with normal brain parenchyma, and calcifications within the tumor are more common and larger than in CN. Oligodendroglioma can erode the inner table of the calvarium, which is also a distinguishing feature [43].
Conclusion Central neurocytoma is a tumor that is almost exclusively located in the anterior part of the lateral ventricle and usually found in young adults. It is typically discovered because of symptoms of intracranial hypertension secondary to obstructive hydrocephalus. Its distinct radiological features include:
• diffuse and diverse calcifications on CT images; • clusters of cysts of variable sizes, resulting in a swisscheese/soap-bubble appearance on T2WI and moderate heterogeneous enhancement on MRI; • and incorporation of the septum pellucidum in bilateral tumors and abutting of the septum pellucidum in unilateral tumors, together with attachment of the ventricular wall, all of which can help in the preoperative diagnosis of CN.
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
References [1] Hassoun J, Gambarelli D, Grisoli F, Pellet W, Salamon G, Pellissier JF, et al. Central neurocytoma: an electron-microscopic study of two cases. Acta Neuropathol 1982;56:151—6. [2] Hassoun J, ¨ Soylemezoglu F, Gambarelli D, Figarella-Branger D, von Ammon K, Kleihues P. Central neurocytoma: a synopsis of clinical and histological features. Brain Pathol 1993;3:297—306. [3] Vinikoff L, Yettou H, Fomekong E, Scavarda D, Taillandier L, Baylac F. Central neurocytoma: a report of four cases. Neurochirurgie 1999;45:39—44. [4] Osborn AG. Central neurocytoma. Diagnostic Imaging—BRAIN, 6. Salt Lake City, Utah: Amirsys; 2004, p. 80—3. [5] Neuronal and mixed neuronal-glial tumours. Louis DN, Ohgaki H, Wiestler OD, Cavenee WK, editors. WHO Classification of tumours of the central nervous system. Lyon: IARC; 2007 [Chapter 6]. [6] Lehmann P, Bruniau A, Deramond H. Neurocytome central : à propos d’un cas. J Radiol 2003;84:855—6. [7] Maiuri F, Spaziante R, De Caro ML, Cappabianca P, Giamundo A, Iaconetta G. Central neurocytoma: clinico-pathological study
Central neurocytoma: radiological and clinico-pathological findings
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
of 5 cases and review of the literature. Clin Neurol Neurosurg 1995;97:219—28. Barbosa MD, Balsitis M, Jaspan TJ, Lowe J. Intraventricular neurocytoma: a clinical and pathological study of three cases and review of the literature. Neurosurgery 1990;26:1045—54. Chang KH, Han MH, Kim DG, Chi JG, Sun DC, Kim SG, et al. MR appearance of central neurocytoma. Acta Radiol 1993;34:520—6. Moussa R, Abadjian G, Nader M, Rizk T, Samaha E, Nohra G, et al. Neurocytome central : à propos de quatre cas. Neurochirurgie 2004;50:639—46. Zhang D, Wen L, Henning TD, Feng XY, Zhang YL, Zou LG, et al. Central neurocytoma: clinical, pathological and neuroradiological findings. Clin Radiol 2006;61:348—57. Georgen SK, Gonzales MF, McLean CA. Intraventricular neurocytoma: radiologic features and review of the literature. Radiology 1992;182:787—92. Leenstra JL, Rodriguez FJ, Frechette CM, et al. Central neurocytoma: management recommendations based on a 35-year experience. Int J Radiation Oncology Biol Phys 2001;67:1145—54. Shin JH, Lee HK, Khang SK, et al. Neuronal tumors of the central nervous system: radiologic findings and pathologic correlation. Radiographics 2002;22:1177—89. Zhang B, Luo B, Zhang Z, Sun G, Wen J. Central neurocytoma: a clinicopathological and neuroradiological study. Neuroradiology 2004;46:888—95. Schmidt MH, Gottfried ON, von Koch CS, Chang SM, McDermott MW. Central neurocytoma: a review. J Neurooncol 2004;66:377—84. Von Deimling A, Janzer R, Kleihues P, Wiestler OD. Patterns of differentiation in central neurocytoma. An immunohistochemical study of eleven biopsies. Acta Neuropathol 1990;79:473—9. Söylemezoglu F, Onder S, Guler Tezel G, Berker M. Neuronal Nuclear antigen (NeuN): A new tool in the diagnosis of central neurocytoma. Pathol Res Pract 2003;199:463—8. Jouvet A, Lellouch-Tubiana A, Boddaert N, Zerah M, Champier J, Fevre- Montange M. Fourth ventricle neurocytoma with lipomatous and ependymal differentiation. Acta Neuropathol 2005;109:346—51. Howard Lee S, Krishna CVG, Zimmerman RA, et al. Cranial MRI and CT. 4th edition Xi’an World Punblishing Corporation: McGraw-Hill, New York; 2000, p. 327—8. Gallina P, Mouchaty H, Buccoliero AM, Di Lorenzo N. Haemorrhagic central neurocytoma of the fourth ventricle. Acta Neurochir (Wien) 2005;147:1193—4. Koeller KK, Sandberg GD. From the archives of the AFIP. Cerebral intraventricular neuroplasms: radiologic-pathologic correlation. Radiographics 2002;22:1473—505. von Deimling A, Kleihues P, Saremaslani P, Yasargil MG, Spoerri O, Sudhof TC, et al. Histogenesis and differentiation potential of central neurocytomas. Lab Invest 1991;64:585—91. Ishiuchi S, Nakazato Y, Lino M, Ozawa S, Tamura M, Ohye C. In vitro neuronal and glial production and differentiation of human central neurocytoma cells. J Neurosci Res 1998;51:526—35. Ishiuchi S, Tamura M. Central neurocytoma: an immunohistochemical, ultrastructural and cell culture study. Acta Neuropathol 1997;94:425—35.
111
[26] Wichmann W, Schubiger O, von Deimling A, Schenker C, Valavanis A. Neuro-radiology of central neurocytoma. Neuroradiology 1991;33:143—8. [27] Cook DJ, Christie SD, Macaulay RJ, Rheaume DE, Holness RO. Fourth ventricle central neurocytoma: case report and review of the literature. Can J Neurol Sci 2004;31:558—64. [28] Bulakbasi N, Guvenc I, Onguru O, Erdogan E, Tayfun C, Ucoz T. The added value of the apparent diffusion coefficient calculation to magnetic resonance imaging in the differentiation and grading of malignant brain tumors. J Comput Assist Tomogr 2004;28:735—46. [29] Law M, Yang S, Wang H, et al. Glioma grading: sensitivity, specificity, and predictive values of perfusion MR imaging and proton MR spectroscopic imaging compared with conventional MR imaging. AJNR Am J Neuroradiol 2003;24: 1989—98. [30] Krabbe K, Gideon P, Wagn P, Hansen U, Thomsen C, Madsen F. MR diffusion imaging of human intracranial tumours. Neuroradiology 1997;39:483—9. [31] Kerkovsky M, Zitterbart K, Svoboda K, Zitterbartova J, Skotakova J. Central neurocytoma: the neuroradiological perspective. Child’s nervous system, 24. Germany: Springer Verlag; 2008, p. 1361—9. [ISSN 0256-7040]. [32] Kanamori M, Kumabe T, Shimizu H, et al. (201)Tl-SPECT, (1)HMRS, and MIB-1 labeling index of central neurocytomas: three case reports. Acta Neurochir (Wien) 2002;144:157—63. [33] Kim DG, Choe WJ, Chang KH, et al. In vivo proton magnetic resonance spectroscopy of central neurocytomas. Neurosurgery 2000;46:329—33. [34] Law M. MR spectroscopy of brain tumors. Top Magn Reson Imaging 2004;15:291—313. [35] Shah T, Jayasundar R, Singh VP, Sarkar C. MRS characterization of central neurocytomas using glycine. NMR Biomed 2011;10:1408—13 http://www.ncbi.nlm.nih.gov/ pubmed/21465595 [36] Chuang MT, Lin WC, Tsai HY, et al. 3T proton magnetic resonance spectroscopy of central neurocytoma : 3 case reports and review of literature. J Comput Assist Tomogr 2005;29:683—8. [37] Le Bihan D, Douek P, Argyropoulou M, Turner R, Patronas N, Fulham M. Diffusion and perfusion magnetic resonance imaging in brain tumors. Top Magn Reson Imaging 1993;5:25—31. [38] Jones BV. Ependymoma. In: Osborn AG, Blaser S, Salzman KL, editors. Diagnostic imaging: brain. Philadelphia: Elsevier; 2004, p. 16—52. [39] Spoto GP, Press GA, Hesselink JR, Solomon M. Intracranial ependymoma subependymoma: MR manifestations. AJR Am J Roentgenol 1990;154:837—45. [40] Chiechi MV, Smirniotopoulos JG, Jones RV. Intracranial subependymomas: CT and MR imaging features in 24 cases. AJR Am J Roentgenol 1995;165:1245—50. [41] Nishio S, Morioka T, Suzuki S, Fukui M. Tumours around the foramen of Monro: clinical and neuroimaging features and their differential diagnosis. J Clin Neurosci 2002;9:137—41. [42] Muzumdar DP, Goel A, Pakhmode CK. Oligodendroglioma causing calvarial erosion. Neurol India 2003;51:129—30. [43] Jayasundar R, Shah T, Vaishya S, et al. In vivo and in vitro MR spectroscopic profile of central neurocytomas. J Magn Reson Imaging 2003;17:256—60.