Prognostic value of magnetic resonance imaging-visible perivascular spaces in patients with idiopathic normal pressure hydrocephalus

Introduction

Idiopathic normal pressure hydrocephalus (iNPH) is a reversible neurological syndrome of uncertain cause, typically presenting with a triad of gait disturbance, cognitive decline, and urinary dysfunction (1,2). Shunting surgery can lead to symptomatic improvement in approximately 50–80% of affected individuals (2-4). Although the cerebrospinal fluid (CSF) tap test remains the main diagnostic method for determining the prognosis after iNPH surgery, it is invasive, may result in complications, and has a low negative predictive value (2,5). Therefore, there is a pressing need for reliable, noninvasive markers to assist in determining which patients with iNPH are suitable candidates for surgery.

The glymphatic system is a recently clearance pathway specific to the central nervous system (6-8). It allows the dynamic exchange of interstitial and CSFs (6-8). Several studies have shown that glymphatic system impairment plays a substantial role in the development of certain neuropathological abnormalities, including iNPH (7-9). Magnetic resonance imaging (MRI)-visible perivascular spaces (PVS) are reportedly being involved in impaired drainage of interstitial fluid in the white matter (WM) and reflect glymphatic system impairment (7-9)]. Previous studies on MRI-visible PVS mainly evaluated the number of PVS using visual rating scales (10-14). The semi-quantitative visual rating methods based on the number of MRI-visible PVS have been developed and applied to different diseases (10-14). The score setting of the existing PVS scales in different brain areas are inconsistent, which inconsistency arises from variations in the scales’ definitions (10-14). However, the ceiling-and-floor effect occurred when PVS visual rating methods were used in previous studies (15,16). Therefore, recent efforts have been made to investigate PVS quantitatively by segmenting and calculating its volume to avoid the ceiling-and-floor effect (15,17).

Although many previous studies have shown associations between MRI-visible PVS in different brain regions and neurodegenerative diseases such as cerebral amyloid angiopathy, Alzheimer’s disease (AD), and iNPH (8,9,18), the association between iNPH and PVS in different brain regions may not address the key clinical challenges, such as prediction of shunt outcomes in patients with iNPH. In fact, the value of MRI-visible PVS in predicting shunt outcomes in patients with iNPH remains unclear.

Therefore, the aim of this study was to determine whether the MRI-visible PVS can serve as predictors of outcome after surgery, potentially informing future patient selection protocols. We used the semi-quantitative visual rating methods and quantitative volume of MRI-visible PVS to minimize selective bias due to inappropriate selection of PVS measures. We present this article in accordance with the TRIPOD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1098/rc).

Methods

Study cohort

Consecutive patients, aged ≥60 years, and referred to the hydrocephalus center of Shenzhen Second People’s Hospital between September 2018 and December 2021 for suspected iNPH were prospectively studied. Figure 1 depicts the flowchart of inclusion of patients with iNPH from initial screening to final analysis. This study initially included 169 patients who visited the hospital for ≥1 progressive gait disorder, cognitive impairment, and urinary incontinence. These patients were evaluated by senior neurologists and underwent relevant head MRI examinations within a duration of 1–2 weeks before shunt surgery.

Figure 1 Study flow chart from the initial screening to the final analysis. CSF, cerebrospinal fluid; iNPH, idiopathic normal pressure hydrocephalus; MRI, magnetic resonance imaging.

According to previous studies and iNPH guidelines (1-4,19,20), the inclusion criteria were as follows: (I) meeting the diagnostic criteria of the iNPH guidelines (1,2), age ≥60 years, presence of one or more symptoms of the triad, which were measurable on iNPH grading scale; (II) ventricular enlargement, Evans’ index score of ≥0.3 on MRI; (III) no disorders causing ventriculomegaly; normal CSF content (protein ≤50 mg/dL and cell count ≤3 cells/mL) and pressure (≤20 cmH₂O); (IV) available brain MRI within 1 month before diagnostic lumbar CSF puncture; (V) and a positive response on the CSF tap tests. The exclusion criteria were as follows: (I) the presence of a musculoskeletal, renal, or mental disorder that would make it difficult to evaluate changes in symptoms; (II) less than 1 year of follow-up; (III) obstructive hydrocephalus; (IV) and not meet the MRI protocol in this study.

The cohort and MRI protocols were approved by the Bioethics Committee of Shenzhen Second People’s Hospital (approval No. KS20190114001). The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Informed consent was obtained from all the study participants or the legal guardians of these participants prior to the commencement of the study.

Clinical evaluation

According to previous studies and iNPH guidelines (2-4,19,20), the modified Rankin Scale (mRS) was used to assess the general disability level and overall situation of the patients to evaluate the outcomes of shunt surgery; the iNPH grading scale (iNPHGS) was used to evaluate individual symptoms related to the triad of gait, cognition, and urination, and the total score represented the overall severity of clinical symptoms (2-4,19,20); the improvement was determined as improvement of 1 level on the mRS or ≥1 point on the iNPHGS (2-4,19,20). The detailed assessment methods of mRS and iNPHGS are in Appendix 1. The primary endpoint was an improvement in the mRS score of ≥1 point (favorable outcome) one year after surgery. The secondary endpoint was an improvement in the iNPHGS score of ≥1 point 1 year after surgery. All patients underwent comprehensive clinical examinations by trained research team of neurologists and neurosurgeons preoperatively and 1 year postoperatively. All trained research coordinators were blinded to the baseline information of the imaging markers.

MRI acquisition

All preoperative MRI scans were performed using a 3.0T MRI scanner (Prisma, Siemens, Erlangen, Germany). The MRI protocol in this study included three-dimensional T1-weighted imaging (3D-T1WI), T2-weighted imaging (T2WI), and fluid-attenuated inversion recovery (FLAIR) in all patients with iNPH. 3D-T1WI: repetition time (TR)/echo time (TE) =2,300/3.55 ms; field of view =240 mm × 240 mm, flip angle =8°, voxel size =0.4 mm × 0.4 mm × 0.9 mm, slice thickness =0.9 mm, and scan time =5 min 20 s. T2WI: TR/TE =4,000/117 ms; field of view =230 mm × 230 mm, flip angle =8°, voxel size =0.7 mm × 0.7 mm × 4.0 mm, slice thickness =4 mm, and slices = 30. FLAIR: TR/TE = 9,000/81 ms; field of view =256 mm × 248 mm, flip angle =8°, voxel size =0.7 mm × 0.7 mm × 4.0 mm, slice thickness =4 mm, and slices =30.

Measurements of PVS visual rating methods

Standard axial images were paralleled to the anterior commissure-posterior commissure plane and acquired with the multiplanar reformer function using Mango Medical 3D software in 3D-T1-weighted images (Figure 2A,2B). When PVS were difficult to categorize, T2 and FLAIR images were used to check that their signal was identical to that of CSF (Figure 2C-2G).

Figure 2 Illustration of typical MRI-visible PVS, lacunes, and the volume of MRI-visible PVS. (A) Reconstructed baseline of axial images (red line indicates the anterior commissure-posterior commissure line) in 3D-T1WI. (B-E) Schematic representation of typical MRI-visible PVS as they appear. The spaces appear as thin black lines that extend from the lateral ventricle towards the cortex. Some PVS run through the CSO and BG (yellow arrows). Since they follow the course of penetrating vessels, they appear linear when imaged parallel to the course of the vessel, and round or ovoid, with a diameter generally <3 mm, when imaged perpendicular to the course of the vessel. PVS have a signal intensity similar to that of CSF. (E) MRI-visible PVS can show focal enlargement, and can be especially enlarged (up to 5–15 mm, even with mass effect) in the inferior basal ganglia (yellow arrow). In contrast with the lacunes, PVS do not have a T2-hyperintense rim around the fluid-filled space on T2WI or fluid-attenuated inversion recovery images. (F,G) Lacunes were round or ovoid, subcortical, fluid-filled (similar signal as CSF) cavities on T1WI or T2WI, and measured between 3–15 mm in diameter (yellow arrows). Lacunes generally have a central CSF-like hypointensity with a surrounding rim of hyperintensity in FLAIR images. (H-J) Illustration of MRI-visible PVS volume measurement (red marks) in reconstructed axial 3D-T1WI, reconstructed sagittal 3D-T1WI, and 3D volume reconstruction, respectively. 3D, three-dimensional; BG, basal ganglia; CSF, cerebrospinal fluid; CSO, centrum semiovale; FLAIR, fluid-attenuated inversion recovery; MRI, magnetic resonance imaging; PVS, perivascular spaces; T1WI, T1-weighted imaging; T2WI, T2-weighted imaging.

Previous studies focused on basal ganglia (BG), centrum semiovale (CSO), hippocampus, and midbrain for PVS measurement (8,9,21). According to previous studies (10,13,21), the scores of PVS in BG, CSO, and hippocampus were coded as follows: 0 = no PVS, 1 = 1–10 PVS, 2 = 11–20 PVS, 3 = 21–40 PVS, and 4 = >40 PVS; PVS in midbrain were rated 0 (none visible) or 1 (visible) (12); the numbers refer to PVS on one side of the brain; a higher score was assigned if there was asymmetry between the sides, and PVS were counted in the slice with the highest number. Meanwhile, we counted the number of PVS in the slice with the highest number of BG, CSO, hippocampus and midbrain. Furthermore, based on the PVS scales used in previous studies, we removed similar PVS scales and screened four typical PVS scales that were used to assess patients with iNPH (10-13). Table 1 summarizes the four PVS scales. Two experienced neuroradiologists, blinded to all clinical information, independently assessed the preoperative images. Any discrepancies between their evaluations were resolved through joint review to achieve consensus.

Volumetric analysis of PVS

We used ITK-SNAP (Version 3.6.0) with the Paintbrush Mode (brush size =1, brush style = circle) to manually measure the PVS volume of the whole brain, BG, and WM and the total intracranial volume on 3D-T1WI (Figure 2H-2J). Furthermore, we calculated the relative total PVS volume (rPVS), relative BG-PVS (rBG-PVS), and relative WM-PVS (rWM-PVS), i.e., the total PVS volume, BG-PVS volume, and WM-PVS volume, divided by the total intracranial volume, respectively. The intracranial volume was measured using the Segment Editor module in 3D Slicer version 4.11 (https://www.slicer.org/). Intracranial volume was defined as the sum of the brain parenchyma volume and the CSF volume both within and outside the ventricular system, all confined within the cranial vault. Volumetric analysis was performed by a senior neuroradiologist with 8 years of experience and confirmed by another neuroradiologist with 22 years of experience. In cases of inconsistent results, the image was re-evaluated until a consensus was reached.

Reliability of visual rating methods and quantitative volume of PVS

Another neuroradiologist crosschecked 20 random patients with iNPH to assess inter-rater reliability.

Statistical analysis

Statistical analyses were conducted with SPSS (version 24.0, IBM) and MedCalc software. Inter-observer reproducibility for assessing preoperative visual rating methods and quantitative volume of MRI-visible PVS were evaluated using Cohen’s kappa and interclass correlation coefficients. Normality of the data distribution was examined using the Shapiro-Wilk test. Variables following a normal distribution were presented as mean ± standard deviation, while skewed data were described by median and interquartile range (25th and 75th percentile). Categorical variables were expressed as number and proportion. Spearman’s correlation coefficient was used to determine the correlation among preoperative visual rating methods and quantitative volume of MRI-visible PVS and clinical parameters. Differences in all MRI-visible PVS parameters and clinical variables between the improvement and non-improvement groups at the 1-year follow-up after shunt surgery were analyzed using independent t-tests, Chi-squared tests, or Mann-Whitney U tests, as appropriate. Receiver operating characteristic (ROC) curve analysis was used to evaluate the predictive effectiveness of the visual rating methods and quantitative volume of MRI-visible PVS. Predictors of postoperative improvement were analyzed using appropriate univariate analyses, which was followed by multivariate logistic regression analysis after adjusting for sex, age, duration, smoking, alcohol consumption, diabetes, and hypertension. Odds ratios (ORs) with 95% confidence intervals (CIs) were used as independent predictors. Statistical significance was set at P<0.05 (two-tailed).

Results

The demographic characteristics and preoperative neuroimaging markers of the 87 patients with iNPH are summarized in Table 2. Of the participants, 51 (59%) were men. The mean age of the study population at the time of shunt surgery was 69±7 years. The median (interquartile range) duration of symptoms before shunting was 24 (IQR, 10–40) months. The typical preoperative symptoms were distributed as follows: all 87 patients had gait disturbance, 76 had cognitive impairment, 40 had urinary symptoms, 1 had gait disturbance and urinary symptoms, 36 had gait disturbance and cognitive impairment, and 40 had the classic triad.

For continuous variables, the inter-rater reliability ranged between 0.85–0.98 (intraclass correlation coefficient), and for ordinal variables, between 0.72–0.86 (κ), as detailed in Table S1. We did not find any correlation between preoperatively visual rating methods and quantitative volume of MRI-visible PVS burden and preoperative mRS score and iNPHGS score (all correlations are detailed in the Table S2). We found significant positive correlations between PVS visual rating scores and PVS volume indices (Figures S1-S3).

According to the primary and secondary outcomes, patients with iNPH were divided into improvement and non-improvement groups at 1 year of follow-up, respectively. Fifty-three (61%) patients showed improvement in the primary outcome; while 34 showed no improvement or condition worsened. Fifty-nine (68%) patients showed improvement in the secondary outcomes; while 28 showed no improvement or condition worsened. According to the primary and secondary outcomes, there were no significant differences in age, sex, mRS score, iNPHGS score, symptom duration, smoking, alcohol consumption, diabetes, or hypertension between the two groups (all P>0.05), as detailed in Table 3.

Based on the primary outcome, rWM-PVS was significantly lower in the improvement group than in the non-improvement group (P=0.041). However, based on the primary or secondary outcomes, there were no significant differences between the two groups in terms of number and scores of MRI-visible PVS in all tested brain regions or the four PVS scales (all P>0.05). Moreover, no significant differences were observed between patients with iNPH who did and did not show improvements in the volume of MRI-visible PVS (all P>0.05), as detailed in Table 3.

Similarly, based on the primary or secondary outcome, the ROC analysis of visual rating methods and quantitative volume of MRI-visible PVS showed no significant differences between the improvement and non-improvement groups (P>0.05) (Figure 3).

Figure 3 ROCs and AUCs of the neuroimaging markers. (A-C) The outcomes of mRS for the non-improvement and improvement groups at 1 year postoperatively. (D-F) The outcomes of iNPHGS for the non-improvement and improvement groups at 1 year postoperatively. AUC, area under the ROC curve; BG, basal ganglia; CI, confidence interval; CSO, centrum semiovale; iNPH, idiopathic normal pressure hydrocephalus; iNPHGS, iNPH grading scale; mRS, modified Rankin Scale; PVS, perivascular spaces; rBG-PVS, relative BG-PVS; ROC, receiver operating characteristic; rPVS, relative PVS; rWM-PVS, relative WM-PVS.

According to the primary and secondary outcomes, multivariable logistic regression analysis showed that no visual rating methods and quantitative volume of MRI-visible PVS were independent predictors for achieving improvement at 1-year follow-up after adjusting for sex, age, duration, smoking, alcohol consumption, diabetes, and hypertension, as detailed in Figure 4.

Figure 4 According to the outcomes of mRS and iNPHGS for the non-improvement and improvement groups at 1 year postoperatively, forest plots with sex-, age-, duration-, smoking-, alcohol-, diabetes-, and hypertension-adjusted odds ratios for visual rating methods and quantitative volume of MRI-visible PVS, respectively. BG, basal ganglia; CI, confidence interval; CSO, centrum semiovale; iNPH, idiopathic normal pressure hydrocephalus; iNPHGS, iNPH grading scale; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; PVS, perivascular spaces; rBG-PVS, relative BG-PVS; rPVS, relative PVS; rWM-PVS, relative WM-PVS.

Discussion

In this study, we used visual rating methods and quantitative volume of MRI-visible PVS to explore the value of MRI-visible PVS in predicting the prognosis of shunt surgery for iNPH. We found that neither PVS visual rating methods nor PVS volume parameters were associated with prognosis at 1 year after surgery in patients with iNPH.

MRI-visible PVS are often difficult to distinguish from lacunes which resulted in moderate reproducibility of the PVS visual rating scales in previous studies (22-24). The recently published STRIVE-2 further normalized the differential diagnosis of PVS and lacunes based on STRIVE-1 and recommended at least 3D-T1WI as the standard scanning sequence for PVS measurement (21). In addition, due to the very small size of MRI-visible PVS, some PVS are only displayed at one or two axial slices, which might lead to differences in the counts of PVS in the axial slice based on different scanning baselines. Therefore, we distinguished PVS from lacunes in the 3D-T1WI sequence using the axial image reconstructed from a unified standard scanning baseline in this study. Reportedly, the reliability of the existing automatic measurement of MRI-visible PVS volume is suboptimal (25,26) and can be strongly influenced by several factors (27), such as WM hyperintensities and ventricular enlargement, which are common in iNPH. In this study, volumetric analyses of PVS were performed manually to avoid above phenomenon. All visual rating methods and quantitative volume of MRI-visible PVS showed good repeatability in our study. Specifically, PVS volume measurements exhibited high inter-rater agreement (see the last three rows of Table S1), supporting the robustness and reproducibility of our quantitative evaluation. These results strengthen the reliability and generalizability of our conclusions.

The current visual rating methods of MRI-visible PVS are essentially the measurement of number of PVS in previous studies (8,9). Although semi-quantitative PVS visual rating methods are currently the common methods of evaluating MR-visible PVS and have been widely used in previous studies (8,9,21), no universally recognized visual scoring method exists. In addition, previous studies used only one PVS assessment method, which may be the reason for the controversial results (23,28-30). We analyzed the methods of existing PVS scales and screened four representative scales that included the characteristics of the existing MRI-visible PVS to avoid methodological bias (10-13). We found that none of the preoperative baseline number, scores, nor scales of MRI-visible PVS were significantly and independently associated with shunt outcomes. These results are similar to those of Gertje et al. and Hilal et al. (23,30), who found PVS had a limited role in clinical application.

In addition, we found that almost all patients with iNPH had PVS rating score of 1 in the hippocampus and midbrain, especially in the midbrain, which showed floor effects. The ceiling effects were seen in the PVS scales in this study. This may be attributed to the fact that the setting of PVS rating scores is inconsistent with the actual number of PVS in these brain regions, which results in their scores failing to reflect the difference in PVS in this region. This has also been by previous studies (15,16). Although the number of small blood vessels is huge and varies from individual to individual, the number of small blood vessels might be relatively fixed. The disparity in the number of PVS may be caused by individual differences, rather than the disease itself. With the improvement of MRI resolution, the number of MRI-visible PVS in individuals might be gradually increase. It may be questionable whether the current visual rating methods of MRI-visible PVS based on the number of MRI-visible PVS are still applicable. In addition, no matter the number, scores, or scales of MRI-visible PVS, most of the MRI-visible PVS visual rating methods above are quantitative or semi-quantitative measurements based on the slice with the highest number (9-13,21), and cannot objectively reflect the actual number of MRI-visible PVS. This represents the MRI-visible PVS burden in that slice rather than the MRI-visible PVS burden of the brain regions. In contrast, the volume of MRI-visible PVS could be a useful reflection of the overall PVS burden in brain regions.

Recently, the volumetric evaluations of MRI-visible PVS were added to reduce the potential ceiling-and-floor effects of semi-quantitative methods in previous studies and to assess the overall PVS burden in the whole brain (15,31). We found no significant difference in any of the MRI-visible PVS volume was detected between the improvement group and the non-improvement group. More and more studies demonstrated that MRI-visible PVS burden reflect glymphatic system impairment. The combination of PVS visual rating methods and quantitative PVS volume is more likely to represent comprehensively and accurately to MRI-visible PVS burden. These facts and the aforementioned findings suggest that, regardless of the severe impairment of MRI-visible PVS burden in patients with iNPH, it is possible to benefit from shunt surgery.

There are some limitations in this study. First, the study was conducted at a single center. The results should be confirmed in other centers. Second, due to the strict differential diagnoses and robust inclusion criteria, the cohort was small. Third, during the postoperative follow-up, some patients lacked the data of specific postoperative gait and cognitive rating scales, such as the Timed Up and Go test, which may have resulted in insufficient evidence regarding the relationship between neuroimaging markers and iNPH-specific clinical symptoms. The iNPHGS and mRS scores can efficiently and easily assess patient conditions and effectively reflect the degree of improvement in patients with iNPH (2,20). Fourth, manual measurements of MRI-visible PVS volume were time-consuming and not suitable for large-sample studies. In addition, due to limitations of manual measurements, our study did not analyze the overall number, length, shape, and other morphological characteristics of PVS. Furthermore, because this study was cross-sectional, it did not evaluate the relationship between pre- versus post-shunt changes in PVS and surgical outcomes. Further prospective longitudinal studies are needed to systematically quantify multidimensional changes in PVS (e.g., volume, length, morphology) before and after surgery and to determine their association with postoperative prognosis.

Conclusions

This study demonstrated that the preoperative baseline visual rating methods and quantitative volume of MRI-visible PVS were not associated with improvement 1 year after surgery in patients with iNPH. The preoperative baseline MRI-visible PVS burden should not be considered as biomarkers to exclude patients with iNPH from shunt surgery.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1098/rc

Data Sharing Statement: Available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1098/dss

Funding: This work was funded by the National Natural Science Foundation of China (No. 82171913) and the Project of Shenzhen Basic Development Project (No. JCYJ 20220818101816036).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-2025-1098/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was approved by the Bioethics Committee of Shenzhen second People’s Hospital (approval no. KS20190114001) and was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. Informed consent was obtained from all the study participants or the legal guardians of these participants prior to the commencement of the study.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Relkin N, Marmarou A, Klinge P, Bergsneider M, Black PM. Diagnosing idiopathic normal-pressure hydrocephalus. Neurosurgery 2005;57:S4-16; discussion ii-v. [Crossref] [PubMed]
2. Nakajima M, Yamada S, Miyajima M, Ishii K, Kuriyama N, Kazui H, et al. Guidelines for Management of Idiopathic Normal Pressure Hydrocephalus (Third Edition): Endorsed by the Japanese Society of Normal Pressure Hydrocephalus. Neurol Med Chir (Tokyo) 2021;61:63-97.
3. Shinoda N, Hirai O, Hori S, Mikami K, Bando T, Shimo D, Kuroyama T, Kuramoto Y, Matsumoto M, Ueno Y. Utility of MRI-based disproportionately enlarged subarachnoid space hydrocephalus scoring for predicting prognosis after surgery for idiopathic normal pressure hydrocephalus: clinical research. J Neurosurg 2017;127:1436-42. [Crossref] [PubMed]
4. Chen J, He W, Zhang X, Lv M, Zhou X, Yang X, Wei H, Ma H, Li H, Xia J. Value of MRI-based semi-quantitative structural neuroimaging in predicting the prognosis of patients with idiopathic normal pressure hydrocephalus after shunt surgery. Eur Radiol 2022;32:7800-10. [Crossref] [PubMed]
5. Mihalj M, Dolić K, Kolić K, Ledenko V. CSF tap test - Obsolete or appropriate test for predicting shunt responsiveness? A systemic review. J Neurol Sci 2016;362:78-84. [Crossref] [PubMed]
6. Nedergaard M, Goldman SA. Glymphatic failure as a final common pathway to dementia. Science 2020;370:50-6. [Crossref] [PubMed]
7. Klostranec JM, Vucevic D, Bhatia KD, Kortman HGJ, Krings T, Murphy KP, terBrugge KG, Mikulis DJ. Current Concepts in Intracranial Interstitial Fluid Transport and the Glymphatic System: Part I-Anatomy and Physiology. Radiology 2021;301:502-14. [Crossref] [PubMed]
8. Klostranec JM, Vucevic D, Bhatia KD, Kortman HGJ, Krings T, Murphy KP, terBrugge KG, Mikulis DJ. Current Concepts in Intracranial Interstitial Fluid Transport and the Glymphatic System: Part II-Imaging Techniques and Clinical Applications. Radiology 2021;301:516-32. [Crossref] [PubMed]
9. 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;16:137-53. [Crossref] [PubMed]
10. Doubal FN, MacLullich AM, Ferguson KJ, Dennis MS, Wardlaw JM. Enlarged perivascular spaces on MRI are a feature of cerebral small vessel disease. Stroke 2010;41:450-4. [Crossref] [PubMed]
11. Zhu YC, Tzourio C, Soumaré A, Mazoyer B, Dufouil C, Chabriat H. 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;41:2483-90. [Crossref] [PubMed]
12. Potter GM, Chappell FM, Morris Z, Wardlaw JM. Cerebral perivascular spaces visible on magnetic resonance imaging: development of a qualitative rating scale and its observer reliability. Cerebrovasc Dis 2015;39:224-31. [Crossref] [PubMed]
13. Aribisala BS, Wiseman S, Morris Z, Valdés-Hernández MC, Royle NA, Maniega SM, Gow AJ, Corley J, Bastin ME, Starr J, Deary IJ, Wardlaw JM. Circulating inflammatory markers are associated with magnetic resonance imaging-visible perivascular spaces but not directly with white matter hyperintensities. Stroke 2014;45:605-7. [Crossref] [PubMed]
14. Gutierrez J, Rundek T, Ekind MS, Sacco RL, Wright CB. Perivascular spaces are associated with atherosclerosis: an insight from the Northern Manhattan Study. AJNR Am J Neuroradiol 2013;34:1711-6. [Crossref] [PubMed]
15. Kim JY, Nam Y, Kim S, Shin NY, Kim HG. MRI-visible Perivascular Spaces in the Neonatal Brain. Radiology 2023;307:e221314. [Crossref] [PubMed]
16. Kim HG, Shin NY, Nam Y, Yun E, Yoon U, Lee HS, Ahn KJ. MRI-visible Dilated Perivascular Space in the Brain by Age: The Human Connectome Project. Radiology 2023;306:e213254. [Crossref] [PubMed]
17. Hicks AJ, Sinclair B, Shultz SR, Pham W, Silbert LC, Schwartz DL, Rowe CC, Ponsford JL, Law M, Spitz G. Associations of Enlarged Perivascular Spaces With Brain Lesions, Brain Age, and Clinical Outcomes in Chronic Traumatic Brain Injury. Neurology 2023;101:e63-73. [Crossref] [PubMed]
18. Ramirez J, Berezuk C, McNeely AA, Scott CJ, Gao F, Black SE. Visible Virchow-Robin spaces on magnetic resonance imaging of Alzheimer's disease patients and normal elderly from the Sunnybrook Dementia Study. J Alzheimers Dis 2015;43:415-24. [Crossref] [PubMed]
19. He WJ, Zhang XJ, Xu QZ, Bai RT, Chen JK, Zhou X, Xia J. Are preoperative phase-contrast CSF flow parameters ideal for predicting the outcome of shunt surgery in patients with idiopathic normal pressure hydrocephalus? Front Neurol 2022;13:959450. [Crossref] [PubMed]
20. Kazui H, Miyajima M, Mori E, Ishikawa M. SINPHONI-2 Investigators. Lumboperitoneal shunt surgery for idiopathic normal pressure hydrocephalus (SINPHONI-2): an open-label randomised trial. Lancet Neurol 2015;14:585-94. [Crossref] [PubMed]
21. Duering M, Biessels GJ, Brodtmann A, Chen C, Cordonnier C, de Leeuw FE, et al. Neuroimaging standards for research into small vessel disease-advances since 2013. Lancet Neurol 2023;22:602-18. [Crossref] [PubMed]
22. Lau KK, Li L, Lovelock CE, Zamboni G, Chan TT, Chiang MF, Lo KT, Küker W, Mak HK, Rothwell PM. Clinical Correlates, Ethnic Differences, and Prognostic Implications of Perivascular Spaces in Transient Ischemic Attack and Ischemic Stroke. Stroke 2017;48:1470-7. [Crossref] [PubMed]
23. Gertje EC, van Westen D, Panizo C, Mattsson-Carlgren N, Hansson O. Association of Enlarged Perivascular Spaces and Measures of Small Vessel and Alzheimer Disease. Neurology 2021;96:e193-202. [Crossref] [PubMed]
24. Benjamin P, Trippier S, Lawrence AJ, Lambert C, Zeestraten E, Williams OA, Patel B, Morris RG, Barrick TR, MacKinnon AD, Markus HS. Lacunar Infarcts, but Not Perivascular Spaces, Are Predictors of Cognitive Decline in Cerebral Small-Vessel Disease. Stroke 2018;49:586-93. [Crossref] [PubMed]
25. Dubost F, Yilmaz P, Adams H, Bortsova G, Ikram MA, Niessen W, Vernooij M, de Bruijne M. Enlarged perivascular spaces in brain MRI: Automated quantification in four regions. Neuroimage 2019;185:534-44. [Crossref] [PubMed]
26. Moses J, Sinclair B, Law M, O'Brien TJ, Vivash L. Automated Methods for Detecting and Quantitation of Enlarged Perivascular spaces on MRI. J Magn Reson Imaging 2023;57:11-24. [Crossref] [PubMed]
27. De Guio F, Duering M, Fazekas F, De Leeuw FE, Greenberg SM, Pantoni L, Aghetti A, Smith EE, Wardlaw J, Jouvent E. Brain atrophy in cerebral small vessel diseases: Extent, consequences, technical limitations and perspectives: The HARNESS initiative. J Cereb Blood Flow Metab 2020;40:231-45. [Crossref] [PubMed]
28. Zou Q, Wang M, Wei X, Li W. Prevalence and Risk Factors for Enlarged Perivascular Spaces in Young Adults from a Neurology Clinic-Based Cohort. Brain Sci 2022;12:1164. [Crossref] [PubMed]
29. Passiak BS, Liu D, Kresge HA, Cambronero FE, Pechman KR, Osborn KE, Gifford KA, Hohman TJ, Schrag MS, Davis LT, Jefferson AL. Perivascular spaces contribute to cognition beyond other small vessel disease markers. Neurology 2019;92:e1309-21. [Crossref] [PubMed]
30. Hilal S, Tan CS, Adams HHH, Habes M, Mok V, Venketasubramanian N, Hofer E, Ikram MK, Abrigo J, Vernooij MW, Chen C, Hosten N, Volzke H, Grabe HJ, Schmidt R, Ikram MA. Enlarged perivascular spaces and cognition: A meta-analysis of 5 population-based studies. Neurology 2018;91:e832-42. [Crossref] [PubMed]
31. Bown CW, Carare RO, Schrag MS, Jefferson AL. Physiology and Clinical Relevance of Enlarged Perivascular Spaces in the Aging Brain. Neurology 2022;98:107-17. [Crossref] [PubMed]