Abstract
Introduction:
Dilated perivascular spaces (PVS) are associated with small vessel disease in the aging population. We sought to investigate the incidence and dynamic evolution of MRI-detectable PVS progression in patients with cerebral amyloid angiopathy (CAA).
Methods:
Patients with symptomatic CAA who underwent baseline and follow-up MRI scans >2 years apart were included. The severity of both basal ganglia (BG) and centrum semiovale (CSO) PVS were rated. Multivariable logistic regression was used to determine the risk factors for PVS progression.
Results:
We included 90 patients with CAA (mean age 72.6 years, SD 8.0 years), of which 53 (58.9%) had intracerebral hemorrhage (ICH) at baseline. During a median follow-up of 4.8 years (IQR 3.6 – 6.6 years), PVS progression was observed in 24 patients (26.7%) at follow-up MRI. After adjusting for age, hypertension and time between baseline and follow-up MRI, cerebral microbleed (CMB) progression (OR 4.12, 95% CI 1.31 – 12.95; p=0.015) and presence of ICH at baseline (OR 8.61, 95% CI: 2.09 – 35.52; p=0.003) were independent predictors of PVS progression. In multivariable regression analysis, presence of ICH (OR 8.78, 95% CI 1.74 – 44.35; p=0.009) and hypertension (OR 5.73, 95% CI 1.25 – 26.29; p=0.025) were associated with BG-PVS progression. However, only CMB progression (OR 10.17, 95% CI 1.84 – 56.35; p=0.008) was associated with CSO-PVS progression.
Conclusion:
PVS progression occurs in a subset of CAA patients reimaged after a median of 4.8 years and is associated with CMB progression. PVS progression might be a useful neuroimaging marker for visualizing CAA-related vascular changes.
Keywords: perivascular space, progression, cerebral amyloid angiopathy, MRI
Introduction
Cerebral amyloid angiopathy (CAA) is a common neuropathological finding in the aging population that is caused by accumulation of amyloid-ß (Aß) in the leptomeningeal arteries and cortical arterioles.[1] It is a common cause of lobar hemorrhage in the elderly and may also occur in patients without intracerebral hemorrhage (ICH).[2] Accumulating evidence suggests that a high degree of dilated perivascular spaces (PVS) in centrum semiovale (CSO) is an imaging marker of CAA.[3,4] A neuroradiologic-neuropathological study suggested that using severe CSO-PVS as a diagnostic marker might improve the sensitivity of the in vivo diagnosis of CAA without compromising specificity.[5]
Cerebral small vessel disease (SVD) causes 25% of strokes and contributes to 45% of dementia cases. [6] SVD is diagnosed on the basis of brain imaging biomarkers, including recent small subcortical infarcts, white matter hyperintensities (WMH) of presumed vascular origin, lacunes, cerebral microbleeds (CMBs), cortical superficial siderosis (cSS), PVS, and cerebral atrophy.[7] PVS, also known as Virchow-Robin spaces, are pial-lined, fluid-filled structures found in characteristic locations throughout the brain.[8] Mechanisms underlying their dilation, possibly including reduced perivascular fluid drainage, are under investigation. PVS can become abnormally enlarged or dilated especially in patients with SVD. Dilated PVS in the basal ganglia (BG-PVS) are associated with hypertension and are more common in patients with hypertensive ICH while CSO-PVS are associated with CAA.[9–11] Previous studies suggest that high grades of PVS in either location are associated with increased risk of dementia, independently of other MRI markers of SVD.[12,13]
Although the dynamic evolution of other SVD markers has been investigated in patients with CAA, research on PVS in patients with CAA has been largely cross-sectional. Whether CSO-PVS reach their maximal severity prior to an individual’s first CAA diagnosis or whether they continue to progress after CAA becomes symptomatic remains unknown. The spatial pattern of MRI-detectable PVS progression also remains unknown. The aim of our study was to investigate the risk factors and spatial pattern of MRI-detectable PVS progression in CAA patients.
Methods
Study Population and Patient Selection
We analyzed prospectively collected data from a consecutive CAA patient cohort admitted to Massachusetts General Hospital between 1998 and 2020. Comprehensive clinical evaluation and neuroimaging acquisition data were gathered at the time of enrollment and follow-up period. Patients were screened for eligibility of the study if they fulfilled the following criteria: (1) symptomatic patients with CAA diagnosed according to the modified Boston criteria[14] (2) at least two interpretable brain MRIs with at least T1- and T2-weighted sequences, FLAIR and blood-sensitive T2*-weighted sequences. (3) Time interval >2 years between two MRIs, defined for the purposes of this study as the baseline and follow-up MRIs. The first and most recent follow-up MRIs were analyzed for the present study. This study was approved by the Mass General Brigham Institutional Review Board and in accordance with declaration of Helsinki. This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines, and the checklist is provided as supplementary material.
Neuroimaging Data Acquisition and Analysis
Baseline clinical information was obtained using standardized data collection forms as previously described.[15] The presence of WMH of presumed vascular origin, CMBs, cSS, and PVS were assessed according to STRIVE (Standards for Reporting Vascular Changes on Neuroimaging) criteria[7] by two trained observers in consensus reading without knowledge of clinical and follow-up imaging information.
MR images were obtained using 1.5T or 3.0T scanners with a standardized clinical protocol as previously described.[16] The blood sensitive imaging protocols for 3T MRI (Siemens Trio, Siemens Healthineers, Munich, Germany) were as follows: slice thickness 1.2 mm, TR 27 ms, TE 21 ms, flip angle 15°. The protocol for 1.5T MRI (Signa; General Electric Medical Systems, Milwaukee, WI) included T2* GRE sequence with slice thickness of 5 mm, TR 750 ms, TE 24 ms.
The distribution and severity of PVS were assessed according to STRIVE definitions and rated on axial T2-weighted MRI with 5 mm slice thickness, using a validated 4-point visual rating scale (1 = ≤ 10 PVS, 2 = 11–20 PVS, 3 = 20–40 PVS and 4 = ≥40 PVS) in the BG and CSO as previously described.[5] The severity of PVS was dichotomized as high (PVS>2) or low (PVS≤2) according to our previous studies.[10,11]
The BG-PVS severity and CSO-PVS severity were compared using baseline and follow-up MRI scans. The presence and severity of PVS were rated by two readers, blinded to MRI timepoint. MRI-detectable PVS progression was defined as an increase of 1 point or more on the validated 4-point visual rating scale in the BG or CSO, or arbitrarily >5 newly occurring PVS between baseline and follow-up MRI when the baseline grade was 4. (Fig. 1). The interobserver reliabilities for judging PVS progression, as well as specifically in the BG and CSO, were determined by calculating kappa statistics. The specific point changes were recorded for each participant.
Fig. 1. Representative example of a CAA patient with PVS progression.
Baseline MRI T2-weighted image showing basal ganglia grade 1 PVS (A) and centrum semiovale grade 2 PVS (B). Axial view of follow-up MRI performed 9 years later showed progression of PVS in the basal ganglia (grade 2) (C) and centrum semiovale (grade 3) (D).
The presence and number of lobar CMBs were evaluated on axial T2*-GRE or SWI according to current consensus criteria and categorized as lobar, deep, or cerebellar.[17,18] CMB progression was defined as newly occurring CMB on follow-up MRI. Significant CMB progression was defined as 5 or more new CMBs.[16] cSS was defined as curvilinear residues of blood products following the superficial layers of the cerebral cortex showing hypointense signal on T2*-GRE/SWI. The distribution and severity of cSS was classified as focal (≤3 sulci) or disseminated (≥4 sulci).[19] cSS progression was defined as cSS extension within an already present focus of cSS at the baseline MR or appearance of ≥1 new cSS focus.[16] Periventricular and deep WMHs were visually assessed on axial FLAIR images on the 4-point Fazekas rating scale.[20]
Statistical Analysis
Baseline demographic, clinical, and neuroimaging characteristics of patients with CAA-related ICH vs CAA patients without ICH were compared in univariate analyses by t-test, Wilcoxon rank-sum test, chi-square test, or Fisher exact test as appropriate. The primary goal of the analysis was to explore pathophysiologically relevant predictors of PVS progression. Univariable analyses were initially conducted to screen for variables potentially associated with PVS progression. Variables with a p-value < 0.05 in the univariable analyses, as well as well-established clinically relevant variables reported in previous studies, were entered into the multivariable model to investigate factors associated with PVS progression on follow-up MRI.[21] Multivariable logistic regression analyses were also performed to identify predictors associated with PVS progression in the BG and CSO, respectively. Multicollinearity was evaluated using variance inflation factors (VIFs). Variables with VIF exceeding 5 were considered excluded from the model, and no variable violated this criterion. All statistical analysis were performed using SPSS version 23.0. Significance level was set at 0.05 for the current study.
Results
A total of 90 patients with CAA (mean age: 72.6 years, SD: 8.0 years at baseline) were eligible and included in the current analysis. Two were diagnosed with definite CAA, 54 with probable CAA, and 34 with possible CAA based on the Boston criteria. Of the 90 patients with CAA, 53 (58.9%) presented with ICH at baseline, 37 (41.1%) patients presented with cognitive decline. The baseline characteristics in the CAA-related ICH and CAA without ICH patients are shown in Table S1. CAA patients without ICH demonstrated higher prevalence of hypertension (86.5% vs 58.5%, p=0.004) and lobar CMB (97.3% vs 60.4%, p<0.001) at baseline. The baseline lobar CMB count was significantly higher in CAA without ICH as compared to CAA with ICH (p=0.008). In contrast, cSS progression was more common in patients with CAA-related ICH than those without ICH (56.6% vs 29.7%, p=0.012).
PVS Progression Prevalence and Risk Factors
Among the 90 CAA patients, the time interval between the baseline and follow-up MRI scans was 4.8 years (IQR, 3.6–6.6 years). PVS progression in either CSO or BG (defined as above) on the follow-up MRI scan was observed in 24 patients (26.7%). Of those, 13 (14.4%) patients had MRI-detectable CSO-PVS progression, 17 (18.9%) had BG-PVS progression, and 6 (6.7%) had both CSO- and BG-PVS progression. The inter-rater agreement for presence versus absence of PVS progression was excellent (κ=0.914). Similar levels of agreement were observed when separately assessing PVS progression in the BG (κ=0.927) and CSO (κ=0.904).
The demographic, radiological, and clinical characteristics of patients with MRI-detectable PVS progression and those without progression are shown in Table 1. Patients who had MRI-detectable PVS progression were more likely to present with ICH at baseline (87.5% vs 48.5%, p=0.001) and show CMB progression (70.8% vs 47.0%, p=0.045). The median time between baseline and follow-up MRI was longer in patients with PVS progression than those without (6.2 years vs 4.8 years, p=0.034). There were no differences in baseline presence of lobar CMBs (66.7% vs 78.8%, p=0.237) or cSS (29.2% vs 19.7%, p=0.339). The baseline characteristics between patients with and without PVS progression in the BG and CSO were summarized in Table S2 and Table S3, respectively.
Table 1.
Comparison of demographic, clinical and imaging characteristics of CAA patients according to PVS progression.
| Variables | Whole CAA cohort (N=90) |
Patients with PVS progression (N=24) |
Patients without PVS progression (N=66) |
p-value |
|---|---|---|---|---|
| Age at baseline MRI, years, mean (SD) | 72.6 (8.0) | 70.2 (7.8) | 73.4 (8.0) | 0.090 |
| Gender, male, n (%) | 53 (58.9) | 14 (58.3) | 39 (59.1) | 0.948 |
| Hypertension, n (%) | 63 (70.0) | 15 (62.5) | 48 (72.7) | 0.349 |
| Hypercholesterolemia, n (%) | 55 (61.1) | 10 (41.7) | 45 (68.2) | 0.023 |
| Antiplatelet use, n (%) | 5 (5.6) | 2 (8.3) | 3 (4.5) | 0.606 |
| Anticoagulant use, n (%) | 11 (12.2) | 4 (16.7) | 7 (10.6) | 0.680 |
| Baseline Fazekas score, median (IQR) | 2.5 (1 – 4) | 3 (2 – 3.8) | 2.0 (1 – 4) | 0.282 |
| Presence of lobar CMBs at baseline, n (%) | 68 (75.6) | 16 (66.7) | 52 (78.8) | 0.237 |
| Number of lobar CMBs at baseline, Median (IQR) | 1 (0.8 – 5) | 1 (0 – 10) | 1.5 (1 – 5) | 0.465 |
| Any CMB progression, n (%) | 48 (53.3) | 17 (70.8) | 31 (47.0) | 0.045 |
| Severe CMB progression (≥5 CMBs), n (%) | 36 (40.0) | 11 (45.8) | 25 (37.9) | 0.496 |
| Presence of cSS, n (%) | 20 (22.2) | 7 (29.2) | 13 (19.7) | 0.339 |
| Focal cSS, n (%) | 16 (17.8) | 5 (20.8) | 11 (16.7) | 0.884 |
| Disseminated cSS, n (%) | 4 (4.4) | 2 (8.3) | 2 (3.0) | 0.288 |
| cSS progression, n (%) | 41 (45.6) | 14 (58.3) | 27 (40.9) | 0.142 |
| Baseline severe CSO-PVS, n (%) | 46 (51.1) | 13 (54.2) | 33 (50.0) | 0.727 |
| Baseline severe BG-PVS, n (%) | 14 (15.6) | 5 (20.8) | 9 (13.6) | 0.614 |
| Baseline BG-PVS category, median (IQR) | 2 (1 – 2) | 2 (1 – 2) | 1.5 (1 – 2) | 0.586 |
| Baseline CSO-PVS category, median (IQR) | 3 (2 – 4) | 3 (1.3 – 3) | 2.5 (2 – 4) | 0.431 |
| Presence of ICH at baseline, n (%) | 53 (58.9) | 21 (87.5) | 32 (48.5) | 0.001 |
| Time from baseline to follow-up MRI, years, median (IQR) | 4.8 (3.6 – 6.6) | 6.2 (4.0 – 7.8) | 4.8 (3.5 – 6.2) | 0.034 |
Abbreviations: CAA= cerebral amyloid angiopathy; CMB = cerebral microbleeds; cSS = cortical superficial siderosis; ICH = intracerebral hemorrhage; IQR = interquartile range; SD = standard deviation; CSO = centrum semiovale; BG = basal ganglia; PVS = perivascular space; MRI = magnetic resonance imaging.
In multivariable analysis, any CMB progression (OR 4.12, 95% CI 1.31 – 12.95, p=0.015) and baseline presence of ICH (OR 8.61, 95% CI 2.09 – 35.52, p=0.003) were independent predictors of any MRI-detectable PVS progression after adjusting for age, hypertension, and time interval between baseline and follow-up MRI scan. We also explored the predictors of PVS progression in the BG and CSO. After adjusting for age at baseline MRI and time between baseline and follow-up MRI, presence of ICH at baseline (OR 8.78, 95% CI 1.74 – 44.35, p=0.009) and hypertension (OR 5.73, 95% CI 1.25 – 26.29, p=0.025) were independently associated with BG-PVS progression. However, only any CMB progression (OR 10.17, 95% CI 1.84 – 56.35, p=0.008) remained an independent predictor of CSO-PVS progression in the multivariable regression model. ICH at baseline was marginally associated with CSO-PVS progression (OR 4.57, 95% CI 0.85 – 24.49, p=0.076) (Table 2).
Table 2.
Multivariable regression analyses of factors associated with PVS progression in CAA patients.
| Variables | OR | 95% CI | p-value |
|---|---|---|---|
| Multivariable model for any PVS progression | |||
| Age, per year increase | 0.95 | 0.88 – 1.02 | 0.148 |
| Hypertension | 1.12 | 0.34 – 3.65 | 0.851 |
| Presence of ICH at baseline | 8.61 | 2.09 – 35.52 | 0.003 |
| Any CMB progression | 4.12 | 1.31 – 12.95 | 0.015 |
| Time from baseline to follow-up MRI, years | 1.09 | 0.89 – 1.33 | 0.397 |
| Multivariable model for CSO-PVS progression | |||
| Age, per year increase | 0.98 | 0.90 – 1.08 | 0.735 |
| Hypertension | 0.28 | 0.07 – 1.15 | 0.078 |
| Presence of ICH at baseline | 4.57 | 0.85 – 24.49 | 0.076 |
| Any CMB progression | 10.17 | 1.84 – 56.35 | 0.008 |
| Multivariable model for BG-PVS progression | |||
| Age, per year increase | 0.96 | 0.89 – 1.05 | 0.377 |
| Hypertension | 5.73 | 1.25 – 26.29 | 0.025 |
| Presence of ICH at baseline | 8.78 | 1.74 – 44.35 | 0.009 |
| Time from baseline to follow-up MRI, years | 1.15 | 0.93 – 1.43 | 0.194 |
Abbreviations: CAA = cerebral amyloid angiopathy; CMB = cerebral microbleeds; cSS = cortical superficial siderosis; ICH = intracerebral hemorrhage; IQR = interquartile range; SD = standard deviation; CSO = centrum semiovale; BG = basal ganglia; PVS = perivascular space; MRI = magnetic resonance imaging.
Discussion
In this observational study, we describe the prevalence and risk factors of MRI-detectable PVS progression in patients with CAA. We found that MRI-detectable PVS progression occurred in 26.7% of CAA patients when reimaged after a median of 4.8 years and was more likely to occur in those who presented with CAA-related ICH than those presenting without ICH. MRI-detectable PVS progression was also associated with appearance of new CMB over the same time interval.
PVS are readily detectable on axial T2-weighted brain MRI sequences used in clinical practice. The underlying mechanism for this dilation is still unclear.[8] Determinants and risk factors of PVS are poorly understood. Cross-sectional studies suggested that hypertension is associated with PVS in the BG.[9,22] No association has been demonstrated between diabetes and PVS.[10,23] In our study, we found that hypertension is an independent predictor of MRI-detectable BG-PVS progression, but not CSO-PVS progression. Advancing age is one of the most well-established risk factors for dilated PVS. A recent meta-analysis of 13 studies[24] and a pooled analysis of 10 population-based cohorts[25] showed that PVS visibility increases with age and this correlation is strongest in the BG. PVS is a readily detectable SVD marker on axial T2-weighted brain MRI sequences which possibly reflects impaired perivascular fluid drainage.[26–28] The dilation may also be driven by other region-specific determinants and risk factors. Dilated PVS has also been demonstrated to be strongly associated with various imaging markers of SVD.[3,9,11] Accumulating evidence suggests that enlarged BG-PVS is associated with hypertension, deep CMBs, greater WMH volume, and lacunes, whereas CSO-PVS is related to lobar CMBs and cSS, while the cross-sectional study design may limit the trajectory of SVD development.[9,10,22–24,29] Previous studies have hypothesized several distinct pathophysiological mechanisms for PVS enlargement. Arterial stiffness as a result of advancing age and hypertension could lead to perforating artery wall damage and reduced vasomotion that result in failure of fluid elimination.[21,30] In addition, concomitant severe PVS in the white matter and CAA pathology as observed in post-mortem studies indicated that aggregation of abnormal proteins such as amyloid β may block drainage pathway and contribute to extent of dilated PVS.[5,21,31]
Several studies have investigated the role of PVS in CAA patients.[4,29,32,33] However, the dynamic evolution of PVS has not been investigated in previous studies. The pathophysiological mechanisms for PVS progression demonstrated in the current analysis remain unclear. The observed association with CMB progression might suggest that MRI-detectable PVS progression is one feature of a broader underlying progression of CAA and CAA-related vasculopathy. The long interscan time interval required to detect substantial PVS progression suggests that either a more slowly progressive process or that subtle PVS progression is more difficult to detect. The low rate of progression over a long follow up indicates that PVS progression is unlikely to be a very useful biomarker in clinical trials.
The occurrence of PVS progression in the BG is unexpected, as CAA is generally not associated with BG-PVS in cross-sectional analyses. The prevailing view is that CAA rarely affects the basal ganglia arterioles, future studies with larger sample size and pathological evidence are needed to confirm whether CAA’s effects on promoting progressive PVS dilation may extend beyond the CSO to the BG compartment as well. In the current analysis, ICH events and hypertension were independently associated with BG-PVS progression. It is speculated that apart from the possibly detrimental effect of lobar hematoma to upstream fluid drainage pathway, concurrent arteriolosclerosis attributed to hypertension and aging could also contribute to BG-PVS progression.
The strengths of our study include the relatively large sample of CAA patients with baseline and follow-up MRIs >2 years apart, blinded MRI assessments, and high interrater reliability for the primary outcome marker of PVS progression. Our study has several limitations. First, our study may have substantial selection bias due to the requirement for long time interval between baseline and follow-up MRI scans in all included patients. Given the relatively small sample size and limited number of outcome events, the findings may yield unstable coefficients and wide confidence intervals and should be interpreted as exploratory and hypothesis-generating. All MRIs were performed for clinical indications, and were not always acquired on the same scanner or with identical imaging parameters in the clinical practice during the long study period, raising the possibility of confounding by indication. These differences may have introduced variability in image contrast and signal intensity, which could potentially influence the visual assessment of PVS burden. However, the high inter-rater agreement suggests that our PVS progression could be reliably detected using clinical MRIs. Second, we also note that the study included individuals with possible as well as probable CAA, raising the possibility that some may not have had CAA as the actual ICH cause. Future studies including non-CAA control cohorts would also be valuable for confirming the PVS progression frequency and patterns observed in CAA. Third, PVS was scored qualitative instead of quantitative. A continuous measure of PVS volume would be more likely to reveal change over time but is of course more technically demanding and may not be possible with heterogeneous MRI acquisition sequences. Fourth, systematic biomarkers or pathological evidence of concomitant Alzheimer’s disease (AD) were not available. Considering that there is potential overlap of CAA and AD, it is of potential interest to assess the role of AD pathologies in the PVS progression. We note finally that PVS progression is currently not known to predict any type of clinical event and should currently be considered a possible marker of progressive underlying vascular changes rather than a clinically meaningful finding.
In summary, we have demonstrated that PVS progression occurs in a substantial subset of CAA patients during long-term follow-up and is associated with ICH at presentation as well as CMB progression. If replicated by future studies, this marker can be considered as a potentially useful measure of ongoing CAA-related vascular changes.
Supplementary Material
Funding Sources
This study was supported by the following NIH grants: R01AG26484, R01AG047975, and R01NS070834. Qi Li was supported by NIH StrokeNet fellowship and National Natural Science Foundation of China (No. 82071337).
Footnotes
Statement of Ethics
This study protocol was reviewed and approved by the Mass General Brigham Institutional Review Board under protocol #2006P000570’s amendment 116. Written informed consent was waived for this study due to the retrospective design.
Conflict of Interest Statement
All the authors have nothing to disclose.
Data Availability Statement
The data that support the findings of this study are not publicly available due to information that could compromise the privacy of research participants, but are available from QL at qili_md@126.com upon request.
References
- 1.Charidimou A, Gang Q, Werring DJ. Sporadic cerebral amyloid angiopathy revisited: recent insights into pathophysiology and clinical spectrum. J Neurol Neurosurg Psychiatry. 2012. Feb;83(2):124–37. [DOI] [PubMed] [Google Scholar]
- 2.Viswanathan A, Greenberg SM. Cerebral amyloid angiopathy in the elderly. Annals of Neurology. 2011. Dec;70(6):871–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Koo H-W, Jo K-I, Yeon J-Y, Kim J-S, Hong S-C. Clinical features of high-degree centrum semiovale-perivascular spaces in cerebral amyloid angiopathy. Journal of the Neurological Sciences. 2016. Aug;367:89–94. [DOI] [PubMed] [Google Scholar]
- 4.Raposo N, Planton M, Payoux P, Péran P, Albucher JF, Calviere L, et al. Enlarged perivascular spaces and florbetapir uptake in patients with intracerebral hemorrhage. Eur J Nucl Med Mol Imaging. 2019. Oct;46(11):2339–47. [DOI] [PubMed] [Google Scholar]
- 5.Charidimou A, Jaunmuktane Z, Baron J-C, Burnell M, Varlet P, Peeters A, et al. White matter perivascular spaces: An MRI marker in pathology-proven cerebral amyloid angiopathy? Neurology. 2014. Jan;82(1):57–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Cannistraro RJ, Badi M, Eidelman BH, Dickson DW, Middlebrooks EH, Meschia JF. CNS small vessel disease: A clinical review. Neurology. 2019. June;92(24):1146–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Wardlaw JM, Smith EE, Biessels GJ, Cordonnier C, Fazekas F, Frayne R, et al. Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration. Lancet Neurol. 2013. Aug;12(8):822–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Wardlaw JM, Benveniste H, Nedergaard M, Zlokovic BV, Mestre H, Lee H, et al. Perivascular spaces in the brain: anatomy, physiology and pathology. Nat Rev Neurol. 2020. Mar;16(3):137–53. [DOI] [PubMed] [Google Scholar]
- 9.Charidimou A, Meegahage R, Fox Z, Peeters A, Vandermeeren Y, Laloux P, et al. Enlarged perivascular spaces as a marker of underlying arteriopathy in intracerebral haemorrhage: a multicentre MRI cohort study. Journal of Neurology, Neurosurgery & Psychiatry. 2013. June;84(6):624–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Martinez-Ramirez S, Pontes-Neto OM, Dumas AP, Auriel E, Halpin A, Quimby M, et al. Topography of dilated perivascular spaces in subjects from a memory clinic cohort. Neurology. 2013. Apr;80(17):1551–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Charidimou A, Boulouis G, Pasi M, Auriel E, Van Etten ES, Haley K, et al. MRI-visible perivascular spaces in cerebral amyloid angiopathy and hypertensive arteriopathy. Neurology. 2017. Mar;88(12):1157–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Banerjee G, Kim HJ, Fox Z, Jäger HR, Wilson D, Charidimou A, et al. MRI-visible perivascular space location is associated with Alzheimer’s disease independently of amyloid burden. Brain. 2017. Apr;140(4):1107–16. [DOI] [PubMed] [Google Scholar]
- 13.Zhu Y-C, Dufouil C, Soumaré A, Mazoyer B, Chabriat H, Tzourio C. High Degree of Dilated Virchow-Robin Spaces on MRI is Associated with Increased Risk of Dementia. JAD. 2010;22(2):663–72. [DOI] [PubMed] [Google Scholar]
- 14.Linn J, Halpin A, Demaerel P, Ruhland J, Giese AD, Dichgans M, et al. Prevalence of superficial siderosis in patients with cerebral amyloid angiopathy. Neurology. 2010;74(17):1346–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Li Q, Zanon Zotin MC, Warren AD, Ma Y, Gurol E, Goldstein JN, et al. CT-Visible Convexity Subarachnoid Hemorrhage is Associated With Cortical Superficial Siderosis and Predicts Recurrent ICH. Neurology. 2021. Feb;96(7):e986–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Pongpitakmetha T, Fotiadis P, Pasi M, Boulouis G, Xiong L, Warren AD, et al. Cortical superficial siderosis progression in cerebral amyloid angiopathy: Prospective MRI study. Neurology. 2020. Apr;94(17):e1853–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Greenberg SM, Vernooij MW, Cordonnier C, Viswanathan A, Al-Shahi Salman R, Warach S, et al. Cerebral microbleeds: a guide to detection and interpretation. Lancet Neurol. 2009. Feb;8(2):165–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Gregoire SM, Chaudhary UJ, Brown MM, Yousry TA, Kallis C, Jäger HR, et al. The Microbleed Anatomical Rating Scale (MARS): Reliability of a tool to map brain microbleeds. Neurology. 2009. Nov;73(21):1759–66. [DOI] [PubMed] [Google Scholar]
- 19.Charidimou A, Linn J, Vernooij MW, Opherk C, Akoudad S, Baron J-C, et al. Cortical superficial siderosis: detection and clinical significance in cerebral amyloid angiopathy and related conditions. Brain. 2015. Aug;138(8):2126–39. [DOI] [PubMed] [Google Scholar]
- 20.Fazekas F, Chawluk J, Alavi A, Hurtig H, Zimmerman R. MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. Am J Roentgenol. 1987;149(2):351–6. [DOI] [PubMed] [Google Scholar]
- 21.Bown CW, Carare RO, Schrag MS, Jefferson AL. Physiology and Clinical Relevance of Enlarged Perivascular Spaces in the Aging Brain. Neurology. 2022. Jan;98(3):107–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Doubal FN, MacLullich AMJ, Ferguson KJ, Dennis MS, Wardlaw JM. Enlarged Perivascular Spaces on MRI Are a Feature of Cerebral Small Vessel Disease. Stroke. 2010. Mar;41(3):450–4. [DOI] [PubMed] [Google Scholar]
- 23.Potter GM, Doubal FN, Jackson CA, Chappell FM, Sudlow CL, Dennis MS, et al. Enlarged Perivascular Spaces and Cerebral Small Vessel Disease. International Journal of Stroke. 2015. Apr;10(3):376–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Francis F, Ballerini L, Wardlaw JM. Perivascular spaces and their associations with risk factors, clinical disorders and neuroimaging features: A systematic review and meta-analysis. International Journal of Stroke. 2019. June;14(4):359–71. [DOI] [PubMed] [Google Scholar]
- 25.Evans TE, Knol MJ, Schwingenschuh P, Wittfeld K, Hilal S, Ikram MA, et al. Determinants of Perivascular Spaces in the General Population: A Pooled Cohort Analysis of Individual Participant Data. Neurology. 2023. Jan;100(2):e107–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Jessen NA, Munk ASF, Lundgaard I, Nedergaard M. The Glymphatic System: A Beginner’s Guide. Neurochem Res. 2015. Dec;40(12):2583–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Rasmussen MK, Mestre H, Nedergaard M. The glymphatic pathway in neurological disorders. Lancet Neurol. 2018. Nov;17(11):1016–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Bakker ENTP, Bacskai BJ, Arbel-Ornath M, Aldea R, Bedussi B, Morris AWJ, et al. Lymphatic Clearance of the Brain: Perivascular, Paravascular and Significance for Neurodegenerative Diseases. Cell Mol Neurobiol. 2016. Mar;36(2):181–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Charidimou A, Jäger RH, Peeters A, Vandermeeren Y, Laloux P, Baron J-C, et al. White Matter Perivascular Spaces Are Related to Cortical Superficial Siderosis in Cerebral Amyloid Angiopathy. Stroke. 2014. Oct;45(10):2930–5. [DOI] [PubMed] [Google Scholar]
- 30.Riba-Llena I, Jiménez-Balado J, Castañé X, Girona A, López-Rueda A, Mundet X, et al. Arterial Stiffness Is Associated With Basal Ganglia Enlarged Perivascular Spaces and Cerebral Small Vessel Disease Load. Stroke. 2018. May;49(5):1279–81. [DOI] [PubMed] [Google Scholar]
- 31.Van Veluw SJ, Biessels GJ, Bouvy WH, Spliet WG, Zwanenburg JJ, Luijten PR, et al. Cerebral amyloid angiopathy severity is linked to dilation of juxtacortical perivascular spaces. J Cereb Blood Flow Metab. 2016. Mar;36(3):576–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Charidimou A, Hong YT, Jäger HR, Fox Z, Aigbirhio FI, Fryer TD, et al. White Matter Perivascular Spaces on Magnetic Resonance Imaging: Marker of Cerebrovascular Amyloid Burden? Stroke. 2015. June;46(6):1707–9. [DOI] [PubMed] [Google Scholar]
- 33.Boulouis G, Charidimou A, Pasi M, Roongpiboonsopit D, Xiong L, Auriel E, et al. Hemorrhage recurrence risk factors in cerebral amyloid angiopathy: Comparative analysis of the overall small vessel disease severity score versus individual neuroimaging markers. Journal of the Neurological Sciences. 2017. Sept;380:64–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are not publicly available due to information that could compromise the privacy of research participants, but are available from QL at qili_md@126.com upon request.