Dilatation of the Virchow-Robin spaces as an indicator of unilateral carotid artery stenosis: correlation with white matter lesions

Abstract

Background

Virchow-Robin space (VRS) dilatation is related to many pathologic conditions, mostly associated with vascular abnormalities. White matter lesions (WMLs) are commonly seen on brain magnetic resonance imaging (MRI) with advancing age and generally considered as potential markers for vascular disease.

Purpose

To investigate if asymmetric dilatation of VRSs and WMLs are associated with unilateral internal carotid artery stenosis (ICAS) and to test the relationship between dilated VRSs and common vascular risk factors.

Material and Methods

Twenty-nine patients (18 men, 11 women; mean age, 68.62 years) with unilateral ICAS (≥70% carotid stenosis) undergoing carotid endarterectomy were identified for this Health Insurance Portability and Accountability Act (HIPAA) compliant prospective study and assessed with brain MRI. Two experienced radiologists scored VRSs and WMLs and evaluated old infarcts, chronic lacunar infarcts, and cerebral atrophy. Asymmetry of WML and VRS scores between two cerebral hemispheres was assessed and associations between VRS scores, WML scores, and explanatory variables (e.g. age, sex, vascular risk factors, and atrophy) were tested.

Results

In this study, WMLs and basal ganglia VRSs were significantly greater in the unilateral hemisphere with ICA stenosis than contralateral hemisphere. Basal ganglia VRSs were associated with WMLs and internal cerebral atrophy. No association between the severity of VRSs and vascular risk factors was found.

Conclusion

ICA stenosis may contribute as a factor in the development of WMLs and dilatation of VRSs by causing chronic hypoperfusion. VRS dilatation may be an additional MRI marker of ICAS.

Keywords

Carotid artery stenosis magnetic resonance imaging Virchow-Robin spaces white matter lesions

Introduction

Virchow-Robin spaces (VRSs) are perivascular spaces that accompany the perforating arteries as they course from the subarachnoid space through the brain parenchyma (1,2). VRSs are characteristically seen in three locations: (i) type I lenticulostriate VRSs – in the basal ganglia, the external and extreme capsule entering through the anterior perforated substance; (ii) type II high-convexity VRSs – in the centrum semiovale; (iii) type III pontomesencephalic VRSs – in the midbrain (1,3). These spaces can be easily identified by their typical location and imaging features. They are usually seen as well-defined rounded, oval, or linear structures with smooth margins, depending on the plane of the image. The signal intensity of VRSs is defined as isointense to cerebrospinal fluid (CSF) with all MR imaging sequences at visual analysis (1 –4). They are microscopic; however, a small number of dilated VRSs (usually ≤5 mm) may be normally identified on magnetic resonance imaging (MRI) in healthy individuals at any age. VRSs may be seen in larger sizes with increasing frequency in advancing age (1,3). Furthermore, recent studies showed that many pathologic conditions may result in abnormal dilatation of VRSs (5 –8). However, the etiology of VRS dilatation is still unclear. Pathologic dilatation of perivascular spaces is believed to be mostly related to microvascular abnormalities that arise due to ageing, hypercholesterolemia, diabetes, hypertension, smoking, and other vascular risk factors (3).

White matter lesions (WMLs), also termed “leukoaraiosis”, are commonly observed on brain MRI with advancing age. White matter lesions have been widely studied as potential markers for vascular disease (3,9 –12). They have been also reported to be associated with markers of carotid atheroma, such as intima media thickening and carotid plaques (13).

The aim of this study was to investigate if asymmetric dilatation of VRSs and WMLs are associated with unilateral internal carotid artery stenosis (ICAS) and to test the relationship between dilated VRSs and common vascular risk factors.

Material and Methods

We prospectively recruited patients with unilateral ICAS undergoing carotid endarterectomy (CEA). The research was approved by the Local Research Ethics Committee, and all patients (or their relatives) provided written informed consent.

Patients

Between June 2011 and July 2013, a total of 33 patients were diagnosed with unilateral severe ICAS by contrast-enhanced neck vessel computed tomography angiography (CTA), gadolinium-enhanced neck vessel MR angiography (MRA), and/or carotid digital subtraction angiography (DSA). Among them, two patients with confluent infarction were excluded because of inadequate analysis for VRSs and WMLs, and two patients were discarded due to motion artifacts. Finally, a total of 29 consecutive patients with unilateral ICAS were included in this study. Internal carotid artery (ICA) stenosis was evaluated by the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria. All patients met the following criteria: (i) more than 70% (significant) stenosis in unilateral ICA; (ii) less than 50% (non-significant) stenosis in contralateral ICA; and (iii) no history of interventional or surgical treatment for ICA stenosis. The demographics of the patients and medical data with vascular risk factors, including history of coronary artery disease, hypertension, diabetes, lower extremity peripheral arterial disease, hypercholesterolemia, and cerebrovascular disease (previous clinical presentation with transient ischemic attack or stroke) were recorded. Smoking status was categorized as never, former, or current. Vascular risk factors were defined according to standard clinical guidelines as described before (2,14).

MRI protocol

All patients were imaged using a 12-chanel phased array head coil on a 1.5 T clinical scanner (Espree; Siemens Medical Solutions, Erlangen, Germany) before surgery. The imaging protocol included the following axial sequences: (i) fluid-attenuated inversion recovery (FLAIR, TR/TE/TI, 11.000/140/2600; section thickness, 3.0 mm), (ii) T1-weighted (T1W) inversion recovery (TR/TE/TI, 6000/25/300; section thickness, 3.0 mm); (iii) variable-echo, fast spin echo (TR/TE1/TE2, 5500/20/90; section thickness, 3.0 mm); (iv) diffusion-weighted imaging (DWI) (TR/TE, 4200/114; section thickness, 5.0 mm; flip angle, 90°; b values, 0, 500, and 1000 s/mm²). For all sequences, the matrix was 256 × 256, and the field of view was 230 × 230 mm. T1W inversion recovery and variable-echo images were acquired with the same orientation as the FLAIR images.

Image analysis

All images were transferred to a commercial imaging workstation (Leonardo; Siemens Medical Solutions). The images were independently reviewed by two radiologists (NS, AS) who were blinded to clinical data with 10 and 18 years of experience in neuroradiology, respectively and old infarcts and chronic lacunar infarcts were recorded. The scores from both investigators were then compared, and disagreements were settled by consensus.

White matter lesions were rated on matched T1-weighted inversion recovery, proton density-weighted, T2-weighted (T2W), FLAIR, and DWI images by using a scoring system based on the Scheltens scale (15). This scale has four subscales, including evaluation of periventricular and deep white matter (WM), basal ganglia, and brain stem. The details are shown in Table 1. The modified version of this scale was used as in another study (3) in which deep WM hyperintensity of the lentiform nucleus were rated in a single group without further classification into lesions putamen and globus pallidus.

Table 1.

WML and VRS scoring system.

White matter (WM) lesions

1. Periventricular WM (frontal, parietal, and occipital) 0 = none, 1 = <5 mm thick, and 2 = >5 mm thick

2. Deep WM (frontal, parietal, temporal, occipital)

3. Basal ganglia (BG) (lentiform nucleus (LN), caudate nucleus (CN), thalamus, internal capsule)

4. Brain stem (cerebellar WM, midbrain and pontine reticular formations, medulla) For 2, 3, and 4 0 = none; 1 = 5 or fewer ≤3 mm, 2 = 6 or more ≤ 3 mm, 3 = 5 or fewer 4 –10 mm, 4 = 6 or more 4 –10 mm, 5 = ≥11 mm, and 6 = confluent

Virchow-Robin spaces (VRS)s

1. Centrum semiovale (0 = none, 1 = less than 5 per side, 2 = more than 5 on one or both sides)

2. Mesencephalon (0 = absent, 1 = present)

3. Subinsular region (0 = none, 1 = less than 5 on either side, 2 = more than 5 on one or both sides).

4. BG scheme 1 0 = VRS only in the substantia innominata (SI) and fewer than 5 VRSs on either side, 1 = more than 5 VRSs in the SI on either side or any VRS in the LN, 2 = VRSs in the CN on either side.

5. BG scheme 2 0 = VRS only in the SI and fewer than 5 VRS on either side 1 = VRS only in the SI or more than 5 VRS on either side 2 = fewer than 5 VRS in the LN on either side 3 = five to 10 VRSs in the LN or fewer than 5 in the CN on either side 4 = more than 10 VRSs in the LN and fewer than 5 in the CN on either side 5 = more than 10 VRSs in LN and more than 5 in the CN on either side

Virchow-Robin spaces were essentially analyzed on T1W inversion recovery sequences by comparing proton density-weighted, T2W, FLAIR, and DWI sequences. VRSs were defined as rounded, oval, or linear structures with signal intensity isointense to that of CSF on all sequences with surrounding normal-appearing brain parenchyma and without diffusion restriction on DWI (1 –4). VRSs were scored according to the grading system used in another study (3), and scores were given by the number and extension of VRSs on the centrum semiovale (CS), mesencephalon (MC), subinsular region (SI), and basal ganglia (basal ganglia scheme 1 = BG1, basal ganglia scheme 2 = BG2). The details regarding the grading system are shown in Table 1. A single image was scored when individual VRSs were observed to extend through multiple images. However, it was not always possible to confidently differentiate enlarged VRSs and small lacunar infarcts, which were seen as small cystic spaces in severe microvascular disease. In such cases, scores of 2 and 5 were allocated in BG schemes 1 and 2, respectively.

Cerebral atrophy was measured by using two indices including the bicaudate ratio for internal cerebral atrophy and the sylvian-fissure ratio for external cerebral atrophy on T1W inversion recovery images (16).

Statistical analysis

Asymmetry of WML and VRS scores between the unilateral hemisphere with ICA stenosis and contralateral hemisphere without significant stenosis was assessed. The Wilcoxon signed ranks test was used to identify differences in scores of WML and VRS between two cerebral hemispheres. Fisher’s exact test was performed to determine associations between VRS scores and vascular risk factors including coronary artery disease, hypertension, diabetes, smoking, lower extremity peripheral arterial disease, hypercholesterolemia, cerebrovascular disease, old infarcts and chronic lacunar infarcts.

Spearman’s Rho correlation test was used to determine associations between VRS scores, WML scores, and explanatory variables (e.g. age, sex, vascular risk factors, and atrophy).

All probability values were two-tailed; P < 0.05 was considered to be statistically significant. All statistical analyses were performed using IBM Statistical Package for the Social Sciences (SPSS) Statistics for Windows, Version 20.0 (Armonk, NY, USA).

Results

Representative images from patients with different degrees of ICA stenosis are shown in Figs. 1 and 2.

Fig. 1.

A 62-year-old man with right ICA stenosis. Carotid MR angiography (a) shows 95% stenosis of the right ICA (arrow). Axial T1-weighted inversion recovery images (b, c) show severe dilatation of VRSs (arrows) greater on the right side than on the left side. FLAIR images (d, e) demonstrate more periventricular and deep WMLs (arrows) on the right side than on the left side.

Fig. 2.

A 63-year-old man with right ICA stenosis. Carotid DSA (a) shows 70% stenosis of the right ICA (arrow). Axial T1-weighted inversion recovery images (b, c) show VRS dilatation (arrows) greater on the right side than on the left side. FLAIR images (d, e) demonstrate more periventricular and deep WMLs (arrows) on the right side than on the left side. However, more WMLs and VRSs are seen in the patient with a higher degree of ICA stenosis, shown in Fig. 1.

Patients with unilateral ICA stenosis consisted of 29 patients (18 men, 11 women), with a mean age of 68.62 ( ± 9.22) years. Baseline characteristics of the study sample are shown in Table 2. Table 3 shows the statistical results for WML and VRS scores.

Table 2.

Baseline characteristics of patients (n = 29).

Characteristics
Mean age (years) (SD)	68.62 (9.22)
Male gender (n) (%)	18 (62.1)
ICA stenosis (%) (SD)	84.21 (10.9)
Coronary artery disease (n) (%)	21 (72.4)
Hypertension (n) (%)	22 (75.9)
Diabetes (n) (%)	11 (37.9)
Lower extremity arterial disease (n) (%)	3 (10)
Hypercholesterolemia (n) (%)	20 (69)
Cerebrovascular disease (n) (%)	9 (31)
Current smoker (n) (%)	10 (34.5)
Old infarct in the unilateral hemisphere (n) (%)	9 (31)
Chronic lacunar infarct in the unilateral hemisphere (n) (%)	9 (31)

Table 3.

Comparison of the WML and VRS scores between unilateral and contralateral hemispheres.

	Unilateral hemisphere Mean (SD)	Contralateral hemisphere Mean (SD)
Scheltens score
Deep WM	7.10 (4.20)	5.48 (4.17)
PV WM	4.14 (1.90)	3.55 (2.13)
BG	2.07 (2.68)	1.03 (1.74)
Infratentorial	0.69 (1.20)	0.52 (1.02)
VRS score
BG Scheme 1	1.97 (0.19)	1.86 (0.35)
BG Scheme 2	4.07 (1.00)	3.69 (1.04)
Centrum semiovale	1.93 (0.26)	1.90 (0.31)
Subinsular	1.69 (0.47)	1.52 (0.51)
Mesencephalic	1.03 (0.19)	0.97 (0.19)

Significant differences were observed in the components of Scheltens WML scores between unilateral and contralateral hemispheres. Periventricular, deep, and basal ganglia WML scores were significantly greater in the unilateral hemisphere.

Although significant differences were seen in BG2- and SI-VRS scores between unilateral and contralateral hemispheres, no significant differences were found in the scores of VRSs in the centrum semiovale. BG2-VRS scores were associated with scores of periventricular, deep, and BG WMLs, and internal cerebral atrophy. CS-VRSs were only associated with SI-VRSs. In addition, unilateral enlarged VRS scores were associated with contralateral VRS scores.

There was no association between the severity of VRSs and vascular risk factors including; chronic lacunar infarctions and old infarcts.

Discussion

In this study, BG2- and SI-VRSs and WMLs were significantly greater in the unilateral hemisphere than in the contralateral hemisphere in patients with unilateral ICAS. BG2-VRSs in the unilateral hemisphere were associated with WMLs, and internal cerebral atrophy.

VRSs are potential enclosed perivascular spaces covered by pia. These spaces are filled with interstitial fluid by pial layers and separated from CSF and the surrounding brain. Dilatation of VRSs occurs with the increase of fluid in perivascular spaces along the course of the penetrating arteries (1,3). Different theories for abnormal dilatation of perivascular spaces have been proposed, including mechanical trauma due to CSF pulsation or vascular ectasia and tortuosity, gradual leaking of the interstitial fluid from the intracellular compartment to the pial space around the metarteriole, permeability of the arterial wall, poorly compensated impedance of interstitial fluid drainage, brain atrophy, or ischemic perivascular tissue loss resulting in a secondary ex-vacuo effect (3,4,6,17,18).

The exact mechanism for VRS dilatation is unknown. Therefore, previous studies have suggested a relationship between abnormal VRS dilatation and variable disease entities with potential pathological mechanisms including: recent-onset multiple sclerosis (5); cognitive decline with ageing (6); Alzheimer’s disease (17); and diseases associated with small vessel disease, such as hypertension (7), vascular dementia (3), and diabetic retinopathy (8). As VRSs accompany the penetrating arterioles, one might expect that they would be related to cerebral small vessel disease. Furthermore, previous pathologic studies have reported that VRS dilatation is most commonly associated with arteriosclerotic microvascular disease, which forms a spectrum of severity graded from 1 to 3 on the basis of histologic appearances (3).

White matter lesions have been commonly used as potential imaging markers of vascular abnormality in the diagnosis of variable diseases with different scoring scales on computed tomography (CT) or MRI (3,15,19). Romero et al. (20) reported that global WML volume was related to carotid stenosis. Saba et al. (21) showed an association between WMLs and the carotid stenosis class. On the other hand, some investigators noted no association between WMLs and ipsilateral carotid stenosis (9,22), but relatively few patients (10–11%) had asymmetric stenosis in a study by Potter et al. (9). In addition, the scoring systems used by these investigators were different than the method of the current study.

In this study, dilated VRSs were assessed in different regions of the brain, including the centrum semiovale, mesencephalon, subinsular region, and basal ganglia, where they are most commonly seen (1,3). We used a scoring system previously shown to have high inter-observer agreement (3). The VRS scoring was performed without knowledge of the clinical data to avoid expectation bias.

Pathological studies suggest that arteriolosclerosis causes WMLs as well as dilated VRSs (4,12). Patankar et al. (3) reported that VRS dilatation may be helpful to differentiate ischemic vascular dementias from degenerative dementias. Rouhl et al. (4) found associations between BG-VRSs and age, hypertension, asymptomatic lacunar infarcts, and WMLs. Potter et al. (3) further investigated associations between VRSs and WMLs, ischemic stroke subtype, and vascular risk factors and reported similar associations between VRSs and WMLs as Doubal et al. (23). These data indicate VRSs as a potential marker of cerebral small vessel disease.

Severe ICA stenosis has been reported to cause hemodynamic cerebral ischemia with physiological imaging, such as positron emission tomography (PET) and single-photon emission computed tomography (SPECT) (24,25). Several studies suggest cerebral hypoperfusion as an important mechanism for brain parenchymal atrophy (26,27). On the basis of these data, we considered that severe ICA stenosis may result in cerebral ischemia evidenced by severity of WMLs and the degree VRS dilatation on MRI. To our knowledge, the associations of dilated VRSs and WMLs with unilateral ICAS have not been studied.

In this study, BG2- and SI-VRSs and WMLs were statistically different in the unilateral hemisphere with ICA stenosis that may be related to cerebral hypoperfusion secondary to ICAS. Park et al. (18) reported that VRS dilatation in the ipsilateral hemisphere deep watershed zone was higher than that in the contralateral hemisphere in patients with unilateral ICAS. They suggested that ICA stenosis followed by chronic ischemia might be a contributing factor in VRS dilatation. However, no statistical difference was found in CS-VRSs between the two hemispheres in the current study. The method for the scoring of VRSs was different with an analysis of the different regions when compared with Park et al. (18), and scores of CS-VRSs were generally symmetric in image analysis. On the other hand, the results of this study showed that ipsilateral BG2-VRSs were associated with unilateral WMLs as well as contralateral WMLs and BG2-VRSs. In addition, ipsilateral VRSs were generally correlated with contralateral VRSs. Therefore, these findings suggest contributions of small vessel disease in addition to ICAS underlying VRSs and WMLs, as mentioned in previous studies (2 –4). The circle of Willis was complete in all patients except anterior cerebral artery A1 segment hypoplasia in four and posterior cerebral artery P1 segment hypoplasia in one of them. Although, no difference was observed between patients with variations and complete circle of Willis, the number of the patients with variations was small. The detailed analysis of the circle of Willis with cranial angiography and further evaluation of collateral ability may be helpful to determine whether the circle of Willis has a role in VRS dilatation and WMLs. Chronic vascular stenosis is a mostly systemic disease, which affects carotid arteries in addition to small vessels. Although the circle of Willis, contralateral carotid arteries and the aortic arch were visualized in diagnostic imaging and patients with unilateral carotid stenosis were involved, the effect of the possible microvascular disease on the contralateral hemisphere cannot be excluded.

Potter et al. (2) reported a stronger association of lacunar stroke with BG-VRSs than with CS-VRSs, which may be related to damage to the perforating arterioles, causing lacunar stroke. The results of this study also showed that BG2-VRSs were related to ICAS. These findings may suggest that the quantification of BG-VRSs may be sufficient in the evaluation of vascular disease.

VRS dilatation has also been attributed to volume loss and brain atrophy, which results in the increase of extracellular fluid in perivascular spaces (4,6,17). A previous study about the relationship between decreased cognitive function and enlarged perivascular spaces in healthy elderly men suggested that VRS dilatation is a common ageing phenomenon that is associated with WMLs and general atrophy (6). In this study ipsilateral BG2-VRSs were associated with internal cerebral atrophy in addition to WMLs. Chronic hypoperfusion due to ICAS may result in cerebral atrophy and WMLs, leading to VRS dilatation.

Enlarged VRSs have been considered to increase with age (2,3,14); however, no association was detected between the severity of VRSs and age as well as vascular risk factors in patients with unilateral ICAS in this study. The small number of the patients might underestimate the effects of ageing and other vascular risk factors in VRS numbers. In addition, the association between age and VRSs may not be demonstrated due to elderly participants.

This study had several limitations. The major limitation is the relatively small sample size because only patients with unilateral ICAS were included; thus, nearly half of the patients undergoing CEA were excluded due to bilateral ICAS and their history of interventional or surgical treatment for ICAS. In patients with ICAS, unilateral and contralateral hemispheres were compared to minimize the possible factors that can attribute to the VRS dilatation. However, the contralateral hemisphere as the control group may still have some limitations. Although we observed more BG2- and SI-VRSs in the unilateral hemisphere with ICA stenosis, the pattern of distribution of enlarged VRSs (i.e. subinsular region vs. basal ganglia) may characterize different etiologies. Another limitation of this study relates to the analysis of dilated VRSs. It is possible to be confident about scoring in cases with a small number of VRSs. However, in more severe cases, it is more difficult to identify VRSs individually as linear structures in several sections, which might be resulted in miscount. In addition, VRSs may not be accurately differentiated from lacunas based on MR imaging. Another limitation includes visual WML scores, instead of lesion volumes. Although volumes can measure WMLs more sensitively, infarcts with a similar signal to WMLs may also be included accidentally (28). On the other hand, visual WML scores are more specific because they are not related to problems caused by artifacts. In addition, scores and volumes have been reported to be closely related (29). Higher field strengths have the advantages of obtaining images with improved spatial resolution and increased image contrast; thus, they may provide better visualization of VRSs and WMLSs and differentiation of VRSs from lacunar infarcts. As mentioned before, the study sample was small, thus these data must be considered and interpreted with caution. However, our findings with a small sample of patients provide important data in patients with unilateral ICAS and seem to generate a need for more dedicated prospective studies with larger patient populations correlated with patients without ICAS and clinical outcomes.

In conclusion, the number of dilated VRSs in the basal ganglia are associated with the severity of WMLs in patients with unilateral ICAS. These findings suggest that ICA stenosis may contribute as a factor in the development of WMLs and dilatation of VRSs by causing chronic hypoperfusion. Virchow-Robin space dilatation appears to hold promise as an additional MRI marker in the diagnosis of ICAS.

Footnotes

Acknowledgements

The support of Dr. Timur Kose for statistical analysis and technologists Sevgi Dalkilic, Ozlem Dincer, Gunisik Karacan, Neslihan Karadeniz, and Serife Yildizlar is gratefully acknowledged.

Conflict of interest

None declared.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Kwee

. Virchow-Robin spaces at MR imaging. Radiographics 2007; 27: 1071–1086.

Potter GM, Doubal FN, Jackson CA, et al. Enlarged perivascular spaces and cerebral small vessel disease. Int J Stroke 2013. DOI: 10.1111/ijs.12054.

Patankar

Mitra

Varma

. Dilatation of the Virchow-Robin space is a sensitive indicator of cerebral microvascular disease: study in elderly patients with dementia. Am J Neuroradiol 2005; 26: 1512–1520.

Rouhl

van Oostenbrugge

Knottnerus

. Virchow-Robin spaces relate to cerebral small vessel disease severity. J Neurol 2008; 255: 692–696.

Achiron

Faibel

. Sandlike appearance of Virchow-Robin spaces in early multiple sclerosis: a novel neuroradiologic marker. Am J Neuroradiol 2002; 23: 376–380.

Maclullich

Wardlaw

Ferguson

. Enlarged perivascular spaces are associated with cognitive function in healthy elderly men. J Neurol Neurosurg Psychiatry 2004; 75: 1519–1523.

Hiroki

Miyashita

. Linear hyperintensity objects on magnetic resonance imaging related to hypertension. Cerebrovasc Dis 2001; 11: 164–168.

Ferguson

Blane

Perros

. Cognitive ability and brain structure in type 1 diabetes: relation to microangiopathy and preceding severe hypoglycemia. Diabetes 2003; 52: 149–156.

Potter

Doubal

Jackson

. Lack of association of white matter lesions with ipsilateral carotid artery stenosis. Cerebrovasc Dis 2012; 33: 378–384.

10.

Christiansen

Larsson

Thomsen

. Age dependent white matter lesions and brain volume changes in healthy volunteers. Acta Radiol 1994; 35: 117–122.

11.

Liou

Chen

Guo

. Cerebral white matter hyperintensities predict functional stroke outcome. Cerebrovasc Dis 2010; 29: 22–27.

12.

Pantoni

Garcia

. Pathogenesis of leukoaraiosis: a review. Stroke 1997; 28: 652–659.

13.

Bots

van Swieten

Breteler

. Cerebral white matter lesions and atherosclerosis in the Rotterdam Study. Lancet 1993; 341: 1232–1237.

14.

Zhu

Tzourio

Soumaré

. Severity of dilated Virchow-Robin spaces is associated with age, blood pressure, and MRI markers of small vessel disease: a population-based study. Stroke 2010; 41: 2483–2490.

15.

Scheltens

Barkhof

Leys

. A semiquantative rating scale for the assessment of signal hyperintensities on magnetic resonance imaging. J Neurol Sci 1993; 114: 7–12.

16.

van Zagten

Kessels

Boiten

. Interobserver agreement in the assessment of cerebral atrophy on CT using bicaudate and sylvian-fissure ratios. Neuroradiology 1999; 41: 261–264.

17.

Chen

Song

Zhang

. Assessment of the Virchow-Robin Spaces in Alzheimer disease, mild cognitive impairment, and normal aging, using high-field MR imaging. Am J Neuroradiol 2011; 32: 1490–1495.

18.

Park

Chung

Suh

. Asymmetric dilatation of virchow-robin space in unilateral internal carotid artery steno-occlusive disease. J Comput Assist Tomogr 2011; 35: 298–302.

19.

Fazekas

Chawluk

Alavi

. MR signal abnormalities at 1.5 T in Alzheimer's dementia and normal aging. Am J Roentgenol 1987; 149: 351–356.

20.

Romero

Beiser

Seshadri

. Carotid artery atherosclerosis, MRI indices of brain ischemia, aging, and cognitive impairment: the Framingham study. Stroke 2009; 40: 1590–1596.

21.

Saba

Sanfilippo

Pascalis

. Carotid artery abnormalities and leukoaraiosis in elderly patients: evaluation with MDCT. Am J Roentgenol 2009; 192: W63–W70.

22.

Altaf

Morgan

Moody

. Brain white matter hyperintensities are associated with carotid intraplaque hemorrhage. Radiology 2008; 248: 202–209.

23.

Doubal

MacLullich

Ferguson

. Enlarged perivascular spaces on MRI are a feature of cerebral small vessel disease. Stroke 2010; 41: 450–454.

24.

Gibbs

Wise

Leenders

. Evaluation of cerebral perfusion reserve in patients with carotid-artery occlusion. Lancet 1984; 1: 310–314.

25.

Powers

Raichle

Grubb

Jr . Positron emission tomography to assess cerebral perfusion. Lancet 1985; 1: 102–103.

26.

Wakita

Tomimoto

Akiguchi

. Axonal damage and demyelination in the white matter after chronic cerebral hypoperfusion in the rat. Brain Res 2002; 924: 63–70.

27.

Yamauchi

Kudoh

Kishibe

. Selective neuronal damage and chronic hemodynamic cerebral ischemia. Ann Neurol 2007; 61: 454–465.

28.

Enzinger

Ropele

Gattringer

. High-grade internal carotid artery stenosis and chronic brain damage: a volumetric magnetic resonance imaging study. Cerebrovasc Dis 2010; 30: 540–546.

29.

van Straaten

Fazekas

Rostrup

. Impact of white matter hyperintensities scoring method on correlations with clinical data: the LADIS study. Stroke 2006; 37: 836–840.