Neuroradiology DOI 10.1007/s00234-015-1491-y
DIAGNOSTIC NEURORADIOLOGY
Large anterior temporal Virchow-Robin spaces: unique MR imaging features Anthony T. Lim & Ronil V. Chandra & Nicholas M. Trost & Penelope A. McKelvie & Stephen L. Stuckey
Received: 2 December 2014 / Accepted: 7 January 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Introduction Large Virchow-Robin (VR) spaces may mimic cystic tumor. The anterior temporal subcortical white matter is a recently described preferential location, with only 18 reported cases. Our aim was to identify unique MR features that could increase prospective diagnostic confidence. Methods Thirty-nine cases were identified between November 2003 and February 2014. Demographic, clinical data and the initial radiological report were retrospectively reviewed. Two neuroradiologists reviewed all MR imaging; a neuropathologist reviewed histological data. Results Median age was 58 years (range 24–86 years); the majority (69 %) was female. There were no clinical symptoms that could be directly referable to the lesion. Two thirds were considered to be VR spaces on the initial radiological report. Mean maximal size was 9 mm (range 5–17 mm); majority (79 %) had perilesional T2 or fluid-attenuated inversion recovery (FLAIR) hyperintensity. The following were identified as potential unique MR features: focal cortical distortion by an adjacent branch of the middle cerebral artery (92 %), smaller A. T. Lim : R. V. Chandra (*) : S. L. Stuckey Neuroradiology Service, Monash Imaging, Monash Health, Monash University, 246 Clayton Road, Melbourne, Victoria 3168, Australia e-mail: [email protected] R. V. Chandra Department of Surgery, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia N. M. Trost Neuroradiology Service, St Vincent’s Hospital, Melbourne, Australia P. A. McKelvie Anatomical Pathology, St Vincent’s Hospital, Melbourne, Australia S. L. Stuckey Southern Clinical School, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia
adjacent VR spaces (26 %), and a contiguous cerebrospinal fluid (CSF) intensity tract (21 %). Surgery was performed in three asymptomatic patients; histopathology confirmed VR spaces. Unique MR features were retrospectively identified in all three patients. Conclusion Large anterior temporal lobe VR spaces commonly demonstrate perilesional T2 or FLAIR signal and can be misdiagnosed as cystic tumor. Potential unique MR features that could increase prospective diagnostic confidence include focal cortical distortion by an adjacent branch of the middle cerebral artery, smaller adjacent VR spaces, and a contiguous CSF intensity tract. Keywords Virchow-Robin space . Temporal lobe . Cystic lesion . MRI . Histology Abbreviations VR Virchow-Robin FLAIR Fluid-attenuated inversion recovery GRE Gradient-recalled echo SPACE Variable-flip-angle 3D TSE T2W (Siemens, Erlangen, Germany) CSF Cerebrospinal fluid
Introduction Virchow-Robin (VR) or dilated perivascular spaces are piallined spaces surrounding penetrating cerebral vessels as they course from the subarachnoid compartment into the brain parenchyma [1]. Small VR spaces can be identified on highresolution MRI in healthy patients of all ages [2, 3]. Enlarged VR spaces are typically 1–5 mm in size, with large and giant VR spaces having greater and more variable diameter [4]. Large and giant VR spaces can thus mimic other brain
Neuroradiology
parenchymal cysts such as neuroglial cysts, ependymal cysts, parasitic cysts, chronic infarction, and cystic tumor. Along with typical neuroimaging features, the anatomical site is an important criterion upon which the prospective diagnosis of a large or giant VR space is made. The most common reported sites are in the basal ganglia along the pathway of the lenticulostriate vessels that enter via the anterior perforated substance, the supratentorial convexity white matter along the pathway of the perforating medullary vessels, and the mesencephalon along the pathway of the perforating collicular vessels [5]. Other reported sites include the subinsular region, cingulate gyrus, corpus callosum, optic tracts, thalamus, pons, and cerebellum [1, 3, 6–8]. While small VR spaces are not uncommonly noted in the subcortical white matter of the anterior temporal lobe, the large (≥5 mm) anterior temporal white matter VR space is a recently described preferential location with only 18 cases published in the literature [9, 10]. We sought to provide further insight into this uncommon entity with a two-institution case series of 39 patients. Our aim was to identify unique MR features that could increase prospective diagnostic confidence, to avoid unnecessary invasive treatment.
Methods
flip-angle 3D TSE T2W (SPACE; Siemens, Erlangen, Germany), and gadolinium-enhanced imaging. Imaging analysis Two neuroradiologists with 3 and 17 years of clinical experience in neurological MR interpretation reviewed all MR imaging. The maximal dimension of the lesion in any plane was measured. Signal characteristics on T1, T2, PD/FLAIR, DWI, SWI/GRE, presence or absence of vascular contact, the presence or absence of focal cortical distortion or thinning, and the presence and extent of perilesional signal change were recorded. Perilesional signal change was classified as mild if there was only a thin rim, moderate for a larger focal area incompletely surrounding the lesion, and extensive if there was a diffuse abnormality surrounding the lesion. In addition, the presence or absence of adjacent smaller VR spaces or tracts, the presence of gadolinium enhancement, duration of imaging follow-up, and morphological changes during the follow-up period were also recorded. Histological analysis A neuropathologist with 27 years of clinical experience reviewed all histopathological slides obtained for patients undergoing surgery.
Patient selection We obtained institutional research ethics board approval for this study. Inclusion criteria were well-circumscribed round or ellipsoid cystic lesions within the white matter of the anterior temporal pole ≥5 mm with internal MR signal characteristics identical to cerebrospinal fluid (CSF). Between our two institutions, three neuroradiologists identified a series of 39 cases during normal clinical practice between November 2003 and February 2014. The patient clinical and imaging record was retrospectively reviewed to obtain patient demographic data, clinical symptoms, and diagnosis stated on the original radiologist’s report.
Imaging acquisition All MR imaging was performed at either 1.5 T or 3 T with the following MRI scanners: 1.5 T Signa Excite or Signa Excite HDxt (GE Healthcare, Milwaukee, WI), 1.5 T Magnetom Symphony, 1.5 T Magnetom Vision, 1.5 T Magnetom Avanto, 3 T Magnetom Skyra, or 3 T Magnetom Verio (Siemens, Erlangen, Germany). Patients had the following sequences performed, which were reviewed if available: T1, T2, fluidattenuated inversion recovery (FLAIR) or proton density, diffusion-weighted imaging (DWI), susceptibility-weighted imaging (SWI) or gradient-recalled echo (GRE), variable-
Results Demographic, clinical, and initial diagnostic data Median age was 58 years (range 24–86 years); the majority (27 patients) was female. There were no clinical symptoms that could be directly referable to the lesion. Overall, two thirds were considered to be VR spaces on the initial radiological report. However, after excluding the cases reported in routine clinical practice by the study investigators, only a third of initial reports (6 of 19 patients) included VR spaces as a possible diagnosis. The range of differential diagnoses in the reports included low-grade glioma, ganglioglioma, dysembryoplastic neuroepithelial tumor, gangliocytoma, neuroepithelial cyst, choroid fissure cyst, arachnoid granulation, old trauma/infarct, gliotic change of microvascular origin, old sterile abscess, and cavernous angioma. MR imaging analysis Mean maximal size was 9 mm (range 5–17 mm). All had internal signal characteristics identical to CSF in all sequences performed (Fig. 1). Five patients did not have FLAIR sequences available. No enhancement was detected in the 30 patients in which gadolinium-enhanced scans were obtained;
Neuroradiology Fig. 1 Thirty-seven-year-old female. Coronal (a) and axial (b) T2WI demonstrates a 9-mm T2 hyperintense anterior temporal lesion (arrows) that is adjacent to a left MCA branch (arrowheads). Sagittal T1WI (c) confirms the elongated shape and demonstrates internal T1 hypointensity. Coronal FLAIR (d) reveals internal suppression of FLAIR signal, i.e., internal signal characteristics identical to CSF, with surrounding elevated FLAIR signal
the remaining 9 patients without gadolinium-enhanced scans had typical morphological features. Morphological data of the cases are summarized in Table 1. Thirty-one patients (79 %) had perilesional T2 or FLAIR hyperintensity. This was mild in 4 patients, moderate in 16, and extensive in 11 patients. There was no relationship between the size of the VR space to presence or severity of perilesional signal. Potential unique MR features were identified. Lesions were contacted by a branch of the middle cerebral artery with associated focal cortical distortion or thinning in all but three patients (92 %) (Fig. 2). Ten of the 39 patients (26 %) had smaller adjacent VR spaces (Fig. 3). Eight patients (21 %) had a contiguous CSF intensity tract (Fig. 2); five of which extended from the subarachnoid space to the VR space, and three extended posteriorly from the VR space toward the temporal horn. In the seven cases that had SPACE sequences performed, these adjacent VR spaces and tracts were more apparent, or only able to be visualized in this sequence.
Histological analysis Three patients (cases 25, 28, and 29) had surgical excision of asymptomatic cystic lesions. Mean size was 10 mm (range 9– 10 mm). Internal MR signal characteristics were identical to CSF; all had perilesional signal (moderate or extensive) with no enhancement. Overall MR findings were similar to the conservatively managed cohort. Histopathological findings were concordant with VR spaces in all three patients. In cases 25 and 29, the adjacent white matter showed mild reduction in myelin density and reactive astrocytic gliosis; in case 29 (Fig. 4), there were additional dilated perivascular spaces. In case 28, the resection was limited to the cyst and a thin rim of white matter. No surrounding dilated perivascular spaces were seen. Potential unique MR features were identified in retrospect in all three patients: Lesions were contacted by a branch of the middle cerebral artery with associated focal cortical distortion or thinning in all three patients (100 %); one of the three patients had a contiguous CSF intensity tract extending anteriorly to the subarachnoid space (33 %).
37/F 48/F
54/F
83/F
76/M
27/M
86/M 61/F 46/F
68/F
46/M 60/M
38/F
44/F 55/F 24/F 55/F 61/M 67/F 61/F 21/F
8 9
10
11
12
13
14 15 16
17
18 19
20
21 22 23 24 25 26 27 28
Seizure Seizure/syncope Unknowna Right optic neuritis Complex partial seizure Right-sided hearing loss Possible acromegaly Unknowna
Tonic clonic seizure, headache, nausea
Planning neurostimulator for Parkinson disease Migraines Unknowna
L R R R R R L R
R
L R
L
R L L
R
L
L
L
L L
75/F 57/M
6 7
R R
Unknowna Left sensorineural hearing loss Right sensorineural hearing loss Parkinson disease, increased falls Left optic neuropathy, disc swelling Intermittent nausea, leg weakness Metastatic cancer Unknowna Unknowna
56/F 62/F
4 5
R R R
L L
Syncopal episodes Unresponsive episodes Right internuclear ophthalmoplegia Left tinnitus, dizziness Left glomus jugulare
59/M 66/F 83/F
1 2 3
Yes
Yes
Yes Yes
Yes Yes
Yes Yes
Yes Yes Yes
12 5 6 5 10 12 12 9
9
5 15
10
17 10 12
8
Yes Yes Yes Yes Yes Yes Yesd Yesb
Yes
Yes Yes
Yes
Yes Yes Yes
Yes
6-5-2 Yes
5
15
9 7
8 1015 14 10
9 7 6
Yes Yes Yes
Yes Yes
Yes
Yes Yes Yes
Yes
Yes
Yes
Yes
Yes Yes
Yes Yes
Yes Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes Yes
At tract At tract
Yes Yes
Yes
Yes Yes Yes
Yes
Yes
Yes
Yes
Yes Yes
Yes Yes
At tract No Yes Yes
Yes Yes Yes
Extensive; reduced then increased Moderate Mild Moderate Moderate Moderate Moderate Extensive Extensiveb
Mild Extensive ModerateExtensive-mild Moderate; slightly reduced Moderate Moderate
ExtensiveModerate-none Mild
None
Moderate
Extensive Extensive
None None-extensiveModerate Extensive Extensive
Mild Extensive Moderate
Side Size Isointense Vessel Cortical Perilesional (mm) to CSF? contact? thinning? signal?
Headaches Unknowna
Indication for initial MRI
Demographic and MRI morphological data of the anterior temporal VR spaces
Case Age/ sex
Table 1
No No No No No Yesc No
No No No No Anterior Noc No
Noc Anterior and Posteriorc Anteriorc
Yesc Noc Yesc
No
No Posteriorc No No
No Yesc Yes
No
Noc
Noc No
No
No
No No
No No
Anterior No
No No Anterior
No
Yes
No No
Yes Yes
No Developed
No No No
Perilesional Tract? VR spaces?
0 54 12 0 11 23 23 0
73
50 4
25
34 36 94
0
57
0
0
100 22
28 33
0 50
14 0 66
Follow-up (months) No No No
Resected?
No
No
No No
No No
N/A Stable Stable N/A Stable Stable Stable
Stable
Stable Stable
Stable
Stable Stable Stable
N/A
No No No No Yes No No Yes
No
No No
No
No No No
No
Reduced No
N/A
N/A
Stable Stable
Stable Stable
N/a No Increased No
Stable N/A Stable
Change in size?
Neuroradiology
Left frontoparietal meningioma Recurrent right-sided headaches Right arm tremor
64/F
53/F
70/F
58/M 72/F
63/F
58/F 40/F 56/M 40/M
72/M
29
30
31
32 33
34
35 36 37 38
39
L
R L R R
R
R R
R
R
L
6
7 8 6 15
10
5 7
10
16
10
Yes
Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes
Yes Yes Yes Yes
Yes
No Yes
Yes
Yes
Yes
None
Moderate Extensive None Moderate
None Moderate around tract None
None
Moderate
Moderate
No
Posteriorc No No No
Noc No No No No
No
No Posterior
No
No
No
No
No No
Yes
No
No
Perilesional Tract? VR spaces?
8
36 0 0 0
0
38 0
0
0
13
Follow-up (months)
Stable
Stable N/A N/A N/A
N/A
Stable N/A
N/A
N/A
Stable
Change in size?
No
No No No No
No
No No
No
No
Yes
Resected?
d
c
b
No FLAIR sequences performed in these cases
Volumetric SPACE sequences were performed in these cases, which better demonstrated the adjacent VR spaces and tracts
Case 28: A preoperative planning contrast enhanced T1 sequence was the only sequence available for review
The clinical indication for initial MRI is unknown in these cases. Scans were performed in our institutions for follow-up of the lesion. No clinical signs or symptoms referable to the lesion were identified at clinical review
a
No Yes
Yesd Yesd Yes
Yes
Yesd
Yes
Yes
Yes
Yes
Yes
Side Size Isointense Vessel Cortical Perilesional (mm) to CSF? contact? thinning? signal?
CSF cerebrospinal fluid, VR Virchow-Robin, N/A not available
Worsening headaches, adenocarcinoma Atypical visual field left eye Left paraesthesia Headaches Loss of sensation, headache Loss/distortion of vision
Sensorineural hearing loss Bilateral optic atrophy
Indication for initial MRI
Case Age/ sex
Table 1 (continued)
Neuroradiology
Neuroradiology
MR follow-up Imaging follow-up was available for 22 of the remaining 28 patients with increased perilesional T2 or FLAIR signal (mean 24 months; range 4–100). There was no change in lesion size or perilesional signal in 17 out of these 22 patients (77 %). One patient developed increased perilesional FLAIR signal that later regressed with development of surrounding smaller VR spaces (Fig. 5). One case developed increase in surrounding FLAIR signal before reducing back to below baseline level. Two patients demonstrated reduction in surrounding FLAIR hyperintensity, but size of the cystic spaces remained stable. One patient demonstrated a reduction in size and surrounding FLAIR signal change over time.
Discussion The large anterior temporal white matter VR space is a recently described entity that can mimic cystic tumor. The clinical experience from our large case series of 39 patients has enabled identification of potential unique MR features: focal cortical distortion by an adjacent branch of the middle cerebral artery, smaller adjacent VR spaces, and a contiguous CSF intensity tract. In our cohort, surgery was performed on three
Fig. 2 Thirty-eight-year-old female. Axial FLAIR (a) and sagittal SPACE (b) demonstrate cortical distortion and thinning adjacent to a right MCA branch (arrow). From this point, there is a CSF intensity tract (arrowheads) leading to a 9-mm cystic lesion in the temporal white matter with extensive surrounding T2 hyperintensity
asymptomatic patients with histological confirmation of VR spaces. At least one of these unique MR features was retrospectively identified in all three patients; prospective consideration of a large anterior temporal white matter VR space may have avoided unnecessary invasive treatment. To date, 18 similar cases have been described in the literature. Rawal et al. [9] described 15 cases, the majority with increased perilesional FLAIR signal (11 of 15 patients). This is concordant with our findings. However, in our cohort, a third of patients had extensive perilesional T2 or FLAIR signal change, which is in contrast to that reported by Rawal et al. [9] (1 of 15 patients). The cause of this surrounding FLAIR signal change is uncertain, with postulates including advanced chronic ischemic change [11, 12] or chronic mechanical stress from high blood pressure on the brain arterioles [13]. The mean age of patients in our cohort was slightly higher; the incidence of hypertension in the two cohorts is unknown. Nonetheless, in both of the cohorts, patients underwent surgical excision for asymptomatic lesions on the basis of extensive perilesional signal change. Histological analysis confirmed VR spaces in both cohorts’cases. The potential unique MR feature of focal cortical distortion by an adjacent branch of the middle cerebral arteryin all our cases is not surprising, as these would be the origin of the perforator vessels expected to traverse the perivascular space. Although not proven, we believe that in some patients, tortuous arterial branches may cause focal distortion of the overlying cortex (Figs. 2 and 3) that may in turn obstruct small tracts traversed by perforating vessels that connect the perivascular and subarachnoid spaces [7]. This mechanical obstruction
Fig. 3 Forty-six-year-old female. Axial FLAIR (a) and axial T2 SPACE (b) demonstrate a large VR space adjacent to the left MCA branch. Several smaller adjacent VR spaces (arrowheads) are better visualized on the SPACE sequence
Neuroradiology Fig. 4 Sixty-four-year-old female (case 29). a Axial T2WI: 10-mm cystic lesion in left temporal lobe with moderate adjacent signal hyperintensity. No adjacent VR spaces were apparent on MRI. This lesion was surgically excised with subsequent histopathology analysis. b X100 H&E: The cystic lesion is lined by multiple folded layers or pial membrane (arrowheads). c X100 H&E: enlarged perivascular spaces in white matter adjacent to the cystic lesion (arrows). d X100 Luxol fast blue stain: enlarged perivascular space (at the top) in white matter with myelin pallor and gliosis
could result in secondary dilation of the perivascular space. Reversible mechanical alteration of flow dynamics has also been previously suggested in additional reported cases where anterior temporal VR spaces regressed following resection of meningioma or pituitary shrinkage after apoplexy [10]. Smaller adjacent VR spaces have been suggested as a potential MR feature of anterior temporal VR spaces; Rawal et al. report this finding in 13 % (2 of 15 patients) and suggest that in addition to gliosis, the surrounding perilesional FLAIR signal may relate to smaller VR spaces. We concur with this observation (26 %; 10 of 39 patients); in our cohort, 7 patients had the SPACE sequence performed. Smaller adjacent VR spaces were more apparent, or only able to be visualized in this sequence, compared to FLAIR imaging. Moreover, one of our patients with 4 years of serial MR imaging developed new perilesional FLAIR signal. The perilesional FLAIR signal later regressed with replacement by smaller VR spaces and an overall increase in size of the large VR space. This development of perilesional FLAIR signal over time in our patient may be explained by development and coalescence of tiny surrounding VR spaces that were too small to be resolved by the initial MR imaging technique. Furthermore, dilated perivascular spaces were identified in histological specimens in both a previously reported [9] and our cohort.
The identification of a tract leading from the subarachnoid space to the large VR space is also not unexpected. Occasionally, vessels can be identified on MRI within the dilated VR space. Moreover, histological studies have identified penetrating vessels passing along pial-lined canals from the subarachnoid space to dilated perivascular spaces [7, 14]. In the majority of patients, these tracts and penetrating vessels may be too small to be resolved by current clinical 1.5 and 3 T MR imaging techniques used. In eight of our cases, there was a clear CSF intensity tract either leading from the subarachnoid space to the large VR space or extending posteriorly toward the temporal horn. We believe that this is a previously unreported diagnostic feature of large anterior temporal white matter VR spaces. While the majority of large anterior temporal VR spaces remain stable, an occasional change in the extent of perilesional FLAIR signal, or increase or decrease in size may occur [9, 10]. Nearly a quarter of our patients (5 of 22 patients) demonstrated change in perilesional signal or overall size during MR follow-up (mean 24 months; range 4–100); Rawal et al. report no change in nine patients during MR follow-up (mean 38 months; range 6– 112 months) while Cerase et al. report a series of three patients in which anterior temporal VR spaces regressed.
Neuroradiology
We postulate that this alteration in perilesional signal and size relates to reversible mechanical obstruction or altered CSF flow dynamics [7, 10]. Our study has several limitations. Most importantly, only three cases had histopathological correlation. Nonetheless, all cases had identical internal MR imaging signal characteristics and were similar in location and morphology. The three cases in our study that were excised demonstrated perilesional signal similar to cases that were managed conservatively. We also cannot definitively confirm that large VR spaces conservatively managed are not low-grade cystic gliomas. However, in all cases with increased perilesional T2 or FLAIR signal and imaging surveillance, no cases have exhibited sustained growth making low-grade glioma extremely unlikely [15]. Further limitations are the small sample size, the retrospective nature, the variation in MR scanners and protocols, and the lack of MR imaging follow-up in all cases. However, the diagnosis of a large VR space is still strongly suspected in all our cases; the presence of our suggested unique
MR imaging features certainly increases our diagnostic confidence allowing us to continue conservative therapy. This management strategy has proved appropriate in all patients to date, but we would advocate for continuing MR surveillance until longer follow-up data are available.
Fig. 5 Sixty-two-year-old female. a Axial T2WI above, axial FLAIR below: Initial imaging demonstrates a large unilocular anterior temporal VR space with no surrounding FLAIR signal (arrow). b Axial T2WI above, axial FLAIR below: 35 months later, this mildly increased in size and developed extensive perilesional T2 signal. c Coronal T2WI
above, coronal T1 FLAIR below: Over the next 15 months, the perilesional FLAIR signal reduced in extent and was replaced by smaller adjacent VR spaces (arrowheads) with an overall increase of the size of the VR space from 10 to 15 mm over the course of 4 years
Conclusions Large anterior temporal lobe VR spaces commonly demonstrate perilesional T2 or FLAIR signal and can be misdiagnosed as cystic tumor. Potential unique MR features that could increase prospective diagnostic confidence include focal cortical distortion by an adjacent branch of the middle cerebral artery, smaller adjacent VR spaces, and a contiguous CSF intensity tract. While the majority remains stable during imaging surveillance, an uncommon change in perilesional signal or size can occur. Accurate prospective recognition of these benign “do-not-touch” abnormalities will avoid unnecessary invasive treatment.
Neuroradiology Ethical standards and patient consent We declare that all human and animal studies have been approved by the Human Research Ethics Committee and have therefore been performed in accordance with the ethical standards laid down in the National Statement on Ethical Conduct in Research involving Humans (1999) and its later amendments. Patient consent was waived due to the retrospective nature of this study. Conflict of interest We declare that we have no conflict of interest.
References 1. Song CJ, Kim JH, Kier EL, Bronen RA (2000) MR imaging and histologic features of subinsular bright spots on T2-weighted MR images: Virchow-Robin spaces of the extreme capsule and insular cortex. Radiology 214:671–677. doi:10.1148/radiology.214.3. r00mr17671 2. Zhu Y-C, Dufouil C, Mazoyer B et al (2011) Frequency and location of dilated Virchow-Robin spaces in elderly people: a populationbased 3D MR imaging study. AJNR Am J Neuroradiol 32:709– 713. doi:10.3174/ajnr.A2366 3. Groeschel S, Chong WK, Surtees R, Hanefeld F (2006) VirchowRobin spaces on magnetic resonance images: normative data, their dilatation, and a review of the literature. Neuroradiology 48:745–754. doi:10.1007/s00234-006-0112-1 4. del Hernández MCV, Piper RJ, Wang X et al (2013) Towards the automatic computational assessment of enlarged perivascular spaces on brain magnetic resonance images: a systematic review. J Magn Reson Imaging 38:774–785. doi:10.1002/jmri.24047 5. Kwee RM, Kwee TC (2007) Virchow-Robin spaces at MR imaging. Radiographics 27:1071–1086. doi:10.1148/rg.274065722
6. Osborn AG, Preece MT (2006) Intracranial cysts: radiologicpathologic correlation and imaging approach. Radiology 239:650– 664. doi:10.1148/radiol.2393050823 7. Saeki N, Nagai Y, Matsuura I et al (2004) Histologic characteristics of normal perivascular spaces along the optic tract: new pathogenetic mechanism for edema in tumors in the pituitary region. AJNR Am J Neuroradiol 25:1218–1222 8. Salzman KL, Osborn AG, House P et al (2005) Giant tumefactive perivascular spaces. AJNR Am J Neuroradiol 26:298–305 9. Rawal S, Croul SE, Willinsky RA et al (2014) Subcortical cystic lesions within the anterior superior temporal gyrus: a newly recognized characteristic location for dilated perivascular spaces. AJNR Am J Neuroradiol 35:317–322. doi:10.3174/ajnr.A3669 10. Cerase A, Vallone IM, Muccio CF et al (2010) Regression of dilated perivascular spaces of the brain. Surg Radiol Anat 32:555–561. doi: 10.1007/s00276-009-0603-y 11. Komiyama M, Yasui T, Izumi T (1998) Magnetic resonance imaging features of unusually dilated Virchow-Robin spaces—two case reports. Neurol Med Chir (Tokyo) 38:161–164. doi:10.2176/nmc.38. 161 12. Shiratori K, Mrowka M, Toussaint A et al (2002) Extreme, unilateral widening of Virchow-Robin spaces: case report. Neuroradiology 44: 990–992. doi:10.1007/s00234-002-0840-9 13. Sawada M, Nishi S, Hashimoto N (1999) Unilateral appearance of markedly dilated Virchow-Robin spaces. Clin Radiol 54:334–336. doi:10.1016/S0009-9260(99)90566-4 14. Zhang ET, Inman CB, Weller RO (1990) Interrelationships of the pia mater and the perivascular (Virchow-Robin) spaces in the human cerebrum. J Anat 170:111–123 15. Pallud J, Taillandier L, Capelle L et al (2012) Quantitative morphological magnetic resonance imaging follow-up of low-grade glioma: a plea for systematic measurement of growth rates. Neurosurgery 71: 729–739. doi:10.1227/NEU.0b013e31826213de, discussion 739–40
DIAGNOSTIC NEURORADIOLOGY
Large anterior temporal Virchow-Robin spaces: unique MR imaging features Anthony T. Lim & Ronil V. Chandra & Nicholas M. Trost & Penelope A. McKelvie & Stephen L. Stuckey
Received: 2 December 2014 / Accepted: 7 January 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Introduction Large Virchow-Robin (VR) spaces may mimic cystic tumor. The anterior temporal subcortical white matter is a recently described preferential location, with only 18 reported cases. Our aim was to identify unique MR features that could increase prospective diagnostic confidence. Methods Thirty-nine cases were identified between November 2003 and February 2014. Demographic, clinical data and the initial radiological report were retrospectively reviewed. Two neuroradiologists reviewed all MR imaging; a neuropathologist reviewed histological data. Results Median age was 58 years (range 24–86 years); the majority (69 %) was female. There were no clinical symptoms that could be directly referable to the lesion. Two thirds were considered to be VR spaces on the initial radiological report. Mean maximal size was 9 mm (range 5–17 mm); majority (79 %) had perilesional T2 or fluid-attenuated inversion recovery (FLAIR) hyperintensity. The following were identified as potential unique MR features: focal cortical distortion by an adjacent branch of the middle cerebral artery (92 %), smaller A. T. Lim : R. V. Chandra (*) : S. L. Stuckey Neuroradiology Service, Monash Imaging, Monash Health, Monash University, 246 Clayton Road, Melbourne, Victoria 3168, Australia e-mail: [email protected] R. V. Chandra Department of Surgery, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia N. M. Trost Neuroradiology Service, St Vincent’s Hospital, Melbourne, Australia P. A. McKelvie Anatomical Pathology, St Vincent’s Hospital, Melbourne, Australia S. L. Stuckey Southern Clinical School, Faculty of Medicine, Nursing and Health Sciences, Monash University, Melbourne, Australia
adjacent VR spaces (26 %), and a contiguous cerebrospinal fluid (CSF) intensity tract (21 %). Surgery was performed in three asymptomatic patients; histopathology confirmed VR spaces. Unique MR features were retrospectively identified in all three patients. Conclusion Large anterior temporal lobe VR spaces commonly demonstrate perilesional T2 or FLAIR signal and can be misdiagnosed as cystic tumor. Potential unique MR features that could increase prospective diagnostic confidence include focal cortical distortion by an adjacent branch of the middle cerebral artery, smaller adjacent VR spaces, and a contiguous CSF intensity tract. Keywords Virchow-Robin space . Temporal lobe . Cystic lesion . MRI . Histology Abbreviations VR Virchow-Robin FLAIR Fluid-attenuated inversion recovery GRE Gradient-recalled echo SPACE Variable-flip-angle 3D TSE T2W (Siemens, Erlangen, Germany) CSF Cerebrospinal fluid
Introduction Virchow-Robin (VR) or dilated perivascular spaces are piallined spaces surrounding penetrating cerebral vessels as they course from the subarachnoid compartment into the brain parenchyma [1]. Small VR spaces can be identified on highresolution MRI in healthy patients of all ages [2, 3]. Enlarged VR spaces are typically 1–5 mm in size, with large and giant VR spaces having greater and more variable diameter [4]. Large and giant VR spaces can thus mimic other brain
Neuroradiology
parenchymal cysts such as neuroglial cysts, ependymal cysts, parasitic cysts, chronic infarction, and cystic tumor. Along with typical neuroimaging features, the anatomical site is an important criterion upon which the prospective diagnosis of a large or giant VR space is made. The most common reported sites are in the basal ganglia along the pathway of the lenticulostriate vessels that enter via the anterior perforated substance, the supratentorial convexity white matter along the pathway of the perforating medullary vessels, and the mesencephalon along the pathway of the perforating collicular vessels [5]. Other reported sites include the subinsular region, cingulate gyrus, corpus callosum, optic tracts, thalamus, pons, and cerebellum [1, 3, 6–8]. While small VR spaces are not uncommonly noted in the subcortical white matter of the anterior temporal lobe, the large (≥5 mm) anterior temporal white matter VR space is a recently described preferential location with only 18 cases published in the literature [9, 10]. We sought to provide further insight into this uncommon entity with a two-institution case series of 39 patients. Our aim was to identify unique MR features that could increase prospective diagnostic confidence, to avoid unnecessary invasive treatment.
Methods
flip-angle 3D TSE T2W (SPACE; Siemens, Erlangen, Germany), and gadolinium-enhanced imaging. Imaging analysis Two neuroradiologists with 3 and 17 years of clinical experience in neurological MR interpretation reviewed all MR imaging. The maximal dimension of the lesion in any plane was measured. Signal characteristics on T1, T2, PD/FLAIR, DWI, SWI/GRE, presence or absence of vascular contact, the presence or absence of focal cortical distortion or thinning, and the presence and extent of perilesional signal change were recorded. Perilesional signal change was classified as mild if there was only a thin rim, moderate for a larger focal area incompletely surrounding the lesion, and extensive if there was a diffuse abnormality surrounding the lesion. In addition, the presence or absence of adjacent smaller VR spaces or tracts, the presence of gadolinium enhancement, duration of imaging follow-up, and morphological changes during the follow-up period were also recorded. Histological analysis A neuropathologist with 27 years of clinical experience reviewed all histopathological slides obtained for patients undergoing surgery.
Patient selection We obtained institutional research ethics board approval for this study. Inclusion criteria were well-circumscribed round or ellipsoid cystic lesions within the white matter of the anterior temporal pole ≥5 mm with internal MR signal characteristics identical to cerebrospinal fluid (CSF). Between our two institutions, three neuroradiologists identified a series of 39 cases during normal clinical practice between November 2003 and February 2014. The patient clinical and imaging record was retrospectively reviewed to obtain patient demographic data, clinical symptoms, and diagnosis stated on the original radiologist’s report.
Imaging acquisition All MR imaging was performed at either 1.5 T or 3 T with the following MRI scanners: 1.5 T Signa Excite or Signa Excite HDxt (GE Healthcare, Milwaukee, WI), 1.5 T Magnetom Symphony, 1.5 T Magnetom Vision, 1.5 T Magnetom Avanto, 3 T Magnetom Skyra, or 3 T Magnetom Verio (Siemens, Erlangen, Germany). Patients had the following sequences performed, which were reviewed if available: T1, T2, fluidattenuated inversion recovery (FLAIR) or proton density, diffusion-weighted imaging (DWI), susceptibility-weighted imaging (SWI) or gradient-recalled echo (GRE), variable-
Results Demographic, clinical, and initial diagnostic data Median age was 58 years (range 24–86 years); the majority (27 patients) was female. There were no clinical symptoms that could be directly referable to the lesion. Overall, two thirds were considered to be VR spaces on the initial radiological report. However, after excluding the cases reported in routine clinical practice by the study investigators, only a third of initial reports (6 of 19 patients) included VR spaces as a possible diagnosis. The range of differential diagnoses in the reports included low-grade glioma, ganglioglioma, dysembryoplastic neuroepithelial tumor, gangliocytoma, neuroepithelial cyst, choroid fissure cyst, arachnoid granulation, old trauma/infarct, gliotic change of microvascular origin, old sterile abscess, and cavernous angioma. MR imaging analysis Mean maximal size was 9 mm (range 5–17 mm). All had internal signal characteristics identical to CSF in all sequences performed (Fig. 1). Five patients did not have FLAIR sequences available. No enhancement was detected in the 30 patients in which gadolinium-enhanced scans were obtained;
Neuroradiology Fig. 1 Thirty-seven-year-old female. Coronal (a) and axial (b) T2WI demonstrates a 9-mm T2 hyperintense anterior temporal lesion (arrows) that is adjacent to a left MCA branch (arrowheads). Sagittal T1WI (c) confirms the elongated shape and demonstrates internal T1 hypointensity. Coronal FLAIR (d) reveals internal suppression of FLAIR signal, i.e., internal signal characteristics identical to CSF, with surrounding elevated FLAIR signal
the remaining 9 patients without gadolinium-enhanced scans had typical morphological features. Morphological data of the cases are summarized in Table 1. Thirty-one patients (79 %) had perilesional T2 or FLAIR hyperintensity. This was mild in 4 patients, moderate in 16, and extensive in 11 patients. There was no relationship between the size of the VR space to presence or severity of perilesional signal. Potential unique MR features were identified. Lesions were contacted by a branch of the middle cerebral artery with associated focal cortical distortion or thinning in all but three patients (92 %) (Fig. 2). Ten of the 39 patients (26 %) had smaller adjacent VR spaces (Fig. 3). Eight patients (21 %) had a contiguous CSF intensity tract (Fig. 2); five of which extended from the subarachnoid space to the VR space, and three extended posteriorly from the VR space toward the temporal horn. In the seven cases that had SPACE sequences performed, these adjacent VR spaces and tracts were more apparent, or only able to be visualized in this sequence.
Histological analysis Three patients (cases 25, 28, and 29) had surgical excision of asymptomatic cystic lesions. Mean size was 10 mm (range 9– 10 mm). Internal MR signal characteristics were identical to CSF; all had perilesional signal (moderate or extensive) with no enhancement. Overall MR findings were similar to the conservatively managed cohort. Histopathological findings were concordant with VR spaces in all three patients. In cases 25 and 29, the adjacent white matter showed mild reduction in myelin density and reactive astrocytic gliosis; in case 29 (Fig. 4), there were additional dilated perivascular spaces. In case 28, the resection was limited to the cyst and a thin rim of white matter. No surrounding dilated perivascular spaces were seen. Potential unique MR features were identified in retrospect in all three patients: Lesions were contacted by a branch of the middle cerebral artery with associated focal cortical distortion or thinning in all three patients (100 %); one of the three patients had a contiguous CSF intensity tract extending anteriorly to the subarachnoid space (33 %).
37/F 48/F
54/F
83/F
76/M
27/M
86/M 61/F 46/F
68/F
46/M 60/M
38/F
44/F 55/F 24/F 55/F 61/M 67/F 61/F 21/F
8 9
10
11
12
13
14 15 16
17
18 19
20
21 22 23 24 25 26 27 28
Seizure Seizure/syncope Unknowna Right optic neuritis Complex partial seizure Right-sided hearing loss Possible acromegaly Unknowna
Tonic clonic seizure, headache, nausea
Planning neurostimulator for Parkinson disease Migraines Unknowna
L R R R R R L R
R
L R
L
R L L
R
L
L
L
L L
75/F 57/M
6 7
R R
Unknowna Left sensorineural hearing loss Right sensorineural hearing loss Parkinson disease, increased falls Left optic neuropathy, disc swelling Intermittent nausea, leg weakness Metastatic cancer Unknowna Unknowna
56/F 62/F
4 5
R R R
L L
Syncopal episodes Unresponsive episodes Right internuclear ophthalmoplegia Left tinnitus, dizziness Left glomus jugulare
59/M 66/F 83/F
1 2 3
Yes
Yes
Yes Yes
Yes Yes
Yes Yes
Yes Yes Yes
12 5 6 5 10 12 12 9
9
5 15
10
17 10 12
8
Yes Yes Yes Yes Yes Yes Yesd Yesb
Yes
Yes Yes
Yes
Yes Yes Yes
Yes
6-5-2 Yes
5
15
9 7
8 1015 14 10
9 7 6
Yes Yes Yes
Yes Yes
Yes
Yes Yes Yes
Yes
Yes
Yes
Yes
Yes Yes
Yes Yes
Yes Yes Yes Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes Yes Yes Yes
At tract At tract
Yes Yes
Yes
Yes Yes Yes
Yes
Yes
Yes
Yes
Yes Yes
Yes Yes
At tract No Yes Yes
Yes Yes Yes
Extensive; reduced then increased Moderate Mild Moderate Moderate Moderate Moderate Extensive Extensiveb
Mild Extensive ModerateExtensive-mild Moderate; slightly reduced Moderate Moderate
ExtensiveModerate-none Mild
None
Moderate
Extensive Extensive
None None-extensiveModerate Extensive Extensive
Mild Extensive Moderate
Side Size Isointense Vessel Cortical Perilesional (mm) to CSF? contact? thinning? signal?
Headaches Unknowna
Indication for initial MRI
Demographic and MRI morphological data of the anterior temporal VR spaces
Case Age/ sex
Table 1
No No No No No Yesc No
No No No No Anterior Noc No
Noc Anterior and Posteriorc Anteriorc
Yesc Noc Yesc
No
No Posteriorc No No
No Yesc Yes
No
Noc
Noc No
No
No
No No
No No
Anterior No
No No Anterior
No
Yes
No No
Yes Yes
No Developed
No No No
Perilesional Tract? VR spaces?
0 54 12 0 11 23 23 0
73
50 4
25
34 36 94
0
57
0
0
100 22
28 33
0 50
14 0 66
Follow-up (months) No No No
Resected?
No
No
No No
No No
N/A Stable Stable N/A Stable Stable Stable
Stable
Stable Stable
Stable
Stable Stable Stable
N/A
No No No No Yes No No Yes
No
No No
No
No No No
No
Reduced No
N/A
N/A
Stable Stable
Stable Stable
N/a No Increased No
Stable N/A Stable
Change in size?
Neuroradiology
Left frontoparietal meningioma Recurrent right-sided headaches Right arm tremor
64/F
53/F
70/F
58/M 72/F
63/F
58/F 40/F 56/M 40/M
72/M
29
30
31
32 33
34
35 36 37 38
39
L
R L R R
R
R R
R
R
L
6
7 8 6 15
10
5 7
10
16
10
Yes
Yes Yes Yes Yes Yes
Yes Yes Yes Yes Yes
Yes Yes Yes Yes
Yes
No Yes
Yes
Yes
Yes
None
Moderate Extensive None Moderate
None Moderate around tract None
None
Moderate
Moderate
No
Posteriorc No No No
Noc No No No No
No
No Posterior
No
No
No
No
No No
Yes
No
No
Perilesional Tract? VR spaces?
8
36 0 0 0
0
38 0
0
0
13
Follow-up (months)
Stable
Stable N/A N/A N/A
N/A
Stable N/A
N/A
N/A
Stable
Change in size?
No
No No No No
No
No No
No
No
Yes
Resected?
d
c
b
No FLAIR sequences performed in these cases
Volumetric SPACE sequences were performed in these cases, which better demonstrated the adjacent VR spaces and tracts
Case 28: A preoperative planning contrast enhanced T1 sequence was the only sequence available for review
The clinical indication for initial MRI is unknown in these cases. Scans were performed in our institutions for follow-up of the lesion. No clinical signs or symptoms referable to the lesion were identified at clinical review
a
No Yes
Yesd Yesd Yes
Yes
Yesd
Yes
Yes
Yes
Yes
Yes
Side Size Isointense Vessel Cortical Perilesional (mm) to CSF? contact? thinning? signal?
CSF cerebrospinal fluid, VR Virchow-Robin, N/A not available
Worsening headaches, adenocarcinoma Atypical visual field left eye Left paraesthesia Headaches Loss of sensation, headache Loss/distortion of vision
Sensorineural hearing loss Bilateral optic atrophy
Indication for initial MRI
Case Age/ sex
Table 1 (continued)
Neuroradiology
Neuroradiology
MR follow-up Imaging follow-up was available for 22 of the remaining 28 patients with increased perilesional T2 or FLAIR signal (mean 24 months; range 4–100). There was no change in lesion size or perilesional signal in 17 out of these 22 patients (77 %). One patient developed increased perilesional FLAIR signal that later regressed with development of surrounding smaller VR spaces (Fig. 5). One case developed increase in surrounding FLAIR signal before reducing back to below baseline level. Two patients demonstrated reduction in surrounding FLAIR hyperintensity, but size of the cystic spaces remained stable. One patient demonstrated a reduction in size and surrounding FLAIR signal change over time.
Discussion The large anterior temporal white matter VR space is a recently described entity that can mimic cystic tumor. The clinical experience from our large case series of 39 patients has enabled identification of potential unique MR features: focal cortical distortion by an adjacent branch of the middle cerebral artery, smaller adjacent VR spaces, and a contiguous CSF intensity tract. In our cohort, surgery was performed on three
Fig. 2 Thirty-eight-year-old female. Axial FLAIR (a) and sagittal SPACE (b) demonstrate cortical distortion and thinning adjacent to a right MCA branch (arrow). From this point, there is a CSF intensity tract (arrowheads) leading to a 9-mm cystic lesion in the temporal white matter with extensive surrounding T2 hyperintensity
asymptomatic patients with histological confirmation of VR spaces. At least one of these unique MR features was retrospectively identified in all three patients; prospective consideration of a large anterior temporal white matter VR space may have avoided unnecessary invasive treatment. To date, 18 similar cases have been described in the literature. Rawal et al. [9] described 15 cases, the majority with increased perilesional FLAIR signal (11 of 15 patients). This is concordant with our findings. However, in our cohort, a third of patients had extensive perilesional T2 or FLAIR signal change, which is in contrast to that reported by Rawal et al. [9] (1 of 15 patients). The cause of this surrounding FLAIR signal change is uncertain, with postulates including advanced chronic ischemic change [11, 12] or chronic mechanical stress from high blood pressure on the brain arterioles [13]. The mean age of patients in our cohort was slightly higher; the incidence of hypertension in the two cohorts is unknown. Nonetheless, in both of the cohorts, patients underwent surgical excision for asymptomatic lesions on the basis of extensive perilesional signal change. Histological analysis confirmed VR spaces in both cohorts’cases. The potential unique MR feature of focal cortical distortion by an adjacent branch of the middle cerebral arteryin all our cases is not surprising, as these would be the origin of the perforator vessels expected to traverse the perivascular space. Although not proven, we believe that in some patients, tortuous arterial branches may cause focal distortion of the overlying cortex (Figs. 2 and 3) that may in turn obstruct small tracts traversed by perforating vessels that connect the perivascular and subarachnoid spaces [7]. This mechanical obstruction
Fig. 3 Forty-six-year-old female. Axial FLAIR (a) and axial T2 SPACE (b) demonstrate a large VR space adjacent to the left MCA branch. Several smaller adjacent VR spaces (arrowheads) are better visualized on the SPACE sequence
Neuroradiology Fig. 4 Sixty-four-year-old female (case 29). a Axial T2WI: 10-mm cystic lesion in left temporal lobe with moderate adjacent signal hyperintensity. No adjacent VR spaces were apparent on MRI. This lesion was surgically excised with subsequent histopathology analysis. b X100 H&E: The cystic lesion is lined by multiple folded layers or pial membrane (arrowheads). c X100 H&E: enlarged perivascular spaces in white matter adjacent to the cystic lesion (arrows). d X100 Luxol fast blue stain: enlarged perivascular space (at the top) in white matter with myelin pallor and gliosis
could result in secondary dilation of the perivascular space. Reversible mechanical alteration of flow dynamics has also been previously suggested in additional reported cases where anterior temporal VR spaces regressed following resection of meningioma or pituitary shrinkage after apoplexy [10]. Smaller adjacent VR spaces have been suggested as a potential MR feature of anterior temporal VR spaces; Rawal et al. report this finding in 13 % (2 of 15 patients) and suggest that in addition to gliosis, the surrounding perilesional FLAIR signal may relate to smaller VR spaces. We concur with this observation (26 %; 10 of 39 patients); in our cohort, 7 patients had the SPACE sequence performed. Smaller adjacent VR spaces were more apparent, or only able to be visualized in this sequence, compared to FLAIR imaging. Moreover, one of our patients with 4 years of serial MR imaging developed new perilesional FLAIR signal. The perilesional FLAIR signal later regressed with replacement by smaller VR spaces and an overall increase in size of the large VR space. This development of perilesional FLAIR signal over time in our patient may be explained by development and coalescence of tiny surrounding VR spaces that were too small to be resolved by the initial MR imaging technique. Furthermore, dilated perivascular spaces were identified in histological specimens in both a previously reported [9] and our cohort.
The identification of a tract leading from the subarachnoid space to the large VR space is also not unexpected. Occasionally, vessels can be identified on MRI within the dilated VR space. Moreover, histological studies have identified penetrating vessels passing along pial-lined canals from the subarachnoid space to dilated perivascular spaces [7, 14]. In the majority of patients, these tracts and penetrating vessels may be too small to be resolved by current clinical 1.5 and 3 T MR imaging techniques used. In eight of our cases, there was a clear CSF intensity tract either leading from the subarachnoid space to the large VR space or extending posteriorly toward the temporal horn. We believe that this is a previously unreported diagnostic feature of large anterior temporal white matter VR spaces. While the majority of large anterior temporal VR spaces remain stable, an occasional change in the extent of perilesional FLAIR signal, or increase or decrease in size may occur [9, 10]. Nearly a quarter of our patients (5 of 22 patients) demonstrated change in perilesional signal or overall size during MR follow-up (mean 24 months; range 4–100); Rawal et al. report no change in nine patients during MR follow-up (mean 38 months; range 6– 112 months) while Cerase et al. report a series of three patients in which anterior temporal VR spaces regressed.
Neuroradiology
We postulate that this alteration in perilesional signal and size relates to reversible mechanical obstruction or altered CSF flow dynamics [7, 10]. Our study has several limitations. Most importantly, only three cases had histopathological correlation. Nonetheless, all cases had identical internal MR imaging signal characteristics and were similar in location and morphology. The three cases in our study that were excised demonstrated perilesional signal similar to cases that were managed conservatively. We also cannot definitively confirm that large VR spaces conservatively managed are not low-grade cystic gliomas. However, in all cases with increased perilesional T2 or FLAIR signal and imaging surveillance, no cases have exhibited sustained growth making low-grade glioma extremely unlikely [15]. Further limitations are the small sample size, the retrospective nature, the variation in MR scanners and protocols, and the lack of MR imaging follow-up in all cases. However, the diagnosis of a large VR space is still strongly suspected in all our cases; the presence of our suggested unique
MR imaging features certainly increases our diagnostic confidence allowing us to continue conservative therapy. This management strategy has proved appropriate in all patients to date, but we would advocate for continuing MR surveillance until longer follow-up data are available.
Fig. 5 Sixty-two-year-old female. a Axial T2WI above, axial FLAIR below: Initial imaging demonstrates a large unilocular anterior temporal VR space with no surrounding FLAIR signal (arrow). b Axial T2WI above, axial FLAIR below: 35 months later, this mildly increased in size and developed extensive perilesional T2 signal. c Coronal T2WI
above, coronal T1 FLAIR below: Over the next 15 months, the perilesional FLAIR signal reduced in extent and was replaced by smaller adjacent VR spaces (arrowheads) with an overall increase of the size of the VR space from 10 to 15 mm over the course of 4 years
Conclusions Large anterior temporal lobe VR spaces commonly demonstrate perilesional T2 or FLAIR signal and can be misdiagnosed as cystic tumor. Potential unique MR features that could increase prospective diagnostic confidence include focal cortical distortion by an adjacent branch of the middle cerebral artery, smaller adjacent VR spaces, and a contiguous CSF intensity tract. While the majority remains stable during imaging surveillance, an uncommon change in perilesional signal or size can occur. Accurate prospective recognition of these benign “do-not-touch” abnormalities will avoid unnecessary invasive treatment.
Neuroradiology Ethical standards and patient consent We declare that all human and animal studies have been approved by the Human Research Ethics Committee and have therefore been performed in accordance with the ethical standards laid down in the National Statement on Ethical Conduct in Research involving Humans (1999) and its later amendments. Patient consent was waived due to the retrospective nature of this study. Conflict of interest We declare that we have no conflict of interest.
References 1. Song CJ, Kim JH, Kier EL, Bronen RA (2000) MR imaging and histologic features of subinsular bright spots on T2-weighted MR images: Virchow-Robin spaces of the extreme capsule and insular cortex. Radiology 214:671–677. doi:10.1148/radiology.214.3. r00mr17671 2. Zhu Y-C, Dufouil C, Mazoyer B et al (2011) Frequency and location of dilated Virchow-Robin spaces in elderly people: a populationbased 3D MR imaging study. AJNR Am J Neuroradiol 32:709– 713. doi:10.3174/ajnr.A2366 3. Groeschel S, Chong WK, Surtees R, Hanefeld F (2006) VirchowRobin spaces on magnetic resonance images: normative data, their dilatation, and a review of the literature. Neuroradiology 48:745–754. doi:10.1007/s00234-006-0112-1 4. del Hernández MCV, Piper RJ, Wang X et al (2013) Towards the automatic computational assessment of enlarged perivascular spaces on brain magnetic resonance images: a systematic review. J Magn Reson Imaging 38:774–785. doi:10.1002/jmri.24047 5. Kwee RM, Kwee TC (2007) Virchow-Robin spaces at MR imaging. Radiographics 27:1071–1086. doi:10.1148/rg.274065722
6. Osborn AG, Preece MT (2006) Intracranial cysts: radiologicpathologic correlation and imaging approach. Radiology 239:650– 664. doi:10.1148/radiol.2393050823 7. Saeki N, Nagai Y, Matsuura I et al (2004) Histologic characteristics of normal perivascular spaces along the optic tract: new pathogenetic mechanism for edema in tumors in the pituitary region. AJNR Am J Neuroradiol 25:1218–1222 8. Salzman KL, Osborn AG, House P et al (2005) Giant tumefactive perivascular spaces. AJNR Am J Neuroradiol 26:298–305 9. Rawal S, Croul SE, Willinsky RA et al (2014) Subcortical cystic lesions within the anterior superior temporal gyrus: a newly recognized characteristic location for dilated perivascular spaces. AJNR Am J Neuroradiol 35:317–322. doi:10.3174/ajnr.A3669 10. Cerase A, Vallone IM, Muccio CF et al (2010) Regression of dilated perivascular spaces of the brain. Surg Radiol Anat 32:555–561. doi: 10.1007/s00276-009-0603-y 11. Komiyama M, Yasui T, Izumi T (1998) Magnetic resonance imaging features of unusually dilated Virchow-Robin spaces—two case reports. Neurol Med Chir (Tokyo) 38:161–164. doi:10.2176/nmc.38. 161 12. Shiratori K, Mrowka M, Toussaint A et al (2002) Extreme, unilateral widening of Virchow-Robin spaces: case report. Neuroradiology 44: 990–992. doi:10.1007/s00234-002-0840-9 13. Sawada M, Nishi S, Hashimoto N (1999) Unilateral appearance of markedly dilated Virchow-Robin spaces. Clin Radiol 54:334–336. doi:10.1016/S0009-9260(99)90566-4 14. Zhang ET, Inman CB, Weller RO (1990) Interrelationships of the pia mater and the perivascular (Virchow-Robin) spaces in the human cerebrum. J Anat 170:111–123 15. Pallud J, Taillandier L, Capelle L et al (2012) Quantitative morphological magnetic resonance imaging follow-up of low-grade glioma: a plea for systematic measurement of growth rates. Neurosurgery 71: 729–739. doi:10.1227/NEU.0b013e31826213de, discussion 739–40
