Choroid Plexus Aging: Structural and Vascular Insights from the HCP-Aging Dataset

doi:10.21203/rs.3.rs-5012373/v1

Download PDF

Research Article

Choroid Plexus Aging: Structural and Vascular Insights from the HCP-Aging Dataset

https://doi.org/10.21203/rs.3.rs-5012373/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background: The choroid plexus (ChP), a highly vascularized structure within the ventricles, is essential for cerebrospinal fluid (CSF) production and metabolic waste clearance, crucial for neurofluid homeostasis and cognitive function. ChP enlargement is seen in normal aging and neurodegenerative diseases like Alzheimer's disease (AD). Despite its key role of in the blood-CSF barrier (BCSFB), detailed studies on age-related changes in its perfusion and microstructure remain limited.

Methods: We analyzed data from 641 healthy individuals aged between 36 and 90, using the Human Connectome Project Aging (HCP-A) dataset. Volumetric, perfusion, and diffusion metrics of the ChP were derived from structural MRI, arterial spin labeling (ASL), and diffusion-weighted imaging (DWI), respectively. Partial correlations were used to explore age-related ChP changes, and independent t-tests to examine sex differences across age decades. One-way ANOVA was employed to compare perfusion characteristics among ChP, gray matter (GM), and white matter (WM). Relationships between volume, perfusion, and diffusion were investigated, adjusting for age and sex. Additionally, the distribution of cyst-like structures within the ChP and their diffusion/perfusion MRI characteristics were analyzed across different age groups.

Results: The ChP undergoes notable changes with age, including an increase in volume (r² = 0.2, P < 0.001), a decrease in blood flow (r² = 0.17, P < 0.001), and elevated mean diffusivity (MD) values (r² = 0.16, P < 0.001). Perfusion characteristics showed significant differences between the ChP, GM, and WM (P < 0.001). Both the ChP and GM exhibited age-related declines in CBF, with a more pronounced decline in the ChP. A negative correlation was observed between the age-related increase in ChP volume and the decrease in CBF, suggesting compensatory dystrophic hyperplasia in response to perfusion decline. Cyst-like structures in ChP, characterized by lower MD and reduced CBF, were found to be more prevalent in older individuals.

Conclusions: Our findings provide a detailed quantitative assessment of age-related changes in ChP perfusion and diffusion, which may affect CSF production and circulation, potentially leading to waste solute accumulation and cognitive impairment.

Choroid plexus

Aging

Human Connectome Project

Perfusion function

Vascular degeneration

The choroid plexus (ChP), a highly vascularized secretory structure in the brain’s ventricular system [1], plays key roles in immune function, maintaining the blood-cerebrospinal fluid (CSF) barrier (BCSFB), and clearing metabolic waste. The BCSFB, formed by specialized ChP epithelial cells connected by tight junctions [2], regulates the exchange of substances between the blood and the CSF, protecting the brain from harmful substances [3]. Blood plasma that flows through the fenestrated capillaries is filtered into the ChP interstitium or stromal tissue, primary due to hydrostatic pressure [4]. The ultrafiltrates including ions, nutrients, along with newly secreted CSF, are actively transported into the ventricles [2]. The ChP is the primary site for CSF production, crucial for cushioning the brain, regulating intracranial pressure, and driving CSF circulation. Histological studies have shown that the ChP undergoes age-related alterations, including shortened epithelial cells, impairment of tight junctions, stromal fibrosis, and calcium/iron deposition [5–7]. Changes in structure and blood perfusion of the ChP could potentially lead to impaired CSF production, accumulation of waste solutes, and increased permeability and impairment of the BCSFB [8, 9], which could further contribute to neurodegenerative processes in Alzheimer’s disease (AD) and other forms of dementia. Aging, a primary risk factor of dementia, leads to brain atrophy, loss of neurons and synapses, ventricular enlargement, and amyloid plaque accumulation. Given the role of ChP in AD, it is vital to investigate how normal aging affects the ChP in vivo and identify early diagnostic markers.

Prior MRI studies have predominantly focused on changes in ChP volume during normal aging and in neurodegenerative disorders such as AD, Parkinson's disease, and multiple sclerosis [10–13]. Recent advancements in arterial spin labeling (ASL) techniques show promise for quantifying ChP perfusion and blood-CSF water exchange [14–17]. While pseudo-continuous ASL (pCASL) MRI studies indicate reduced ChP perfusion with age [10, 18], most are limited to animals or small sample sizes of in vivo human studies. The Human Connectome Project in Aging (HCP-A) provides a comprehensive dataset for studying normal aging, including advanced sequences with high-quality perfusion and diffusion MRI data for over 700 healthy adults aged between 36 and 100 years. This resource allows for a detailed analysis of cerebral blood flow (CBF), arterial transit time (ATT) [19], and microstructural changes in the ChP associated with aging. Previous studies, constrained by small sample sizes and resolution, may not fully capture the complexity of age-related changes in the ChP, especially in the absence of vascular evaluation.

In this study, we aim to conduct a comprehensive analysis of age-related changes in ChP measurements, with a specific emphasis on perfusion metrics, using a large adult lifespan dataset. We believe this will enhance our understanding of vascular degeneration and its relationship with the volumetric and microstructural changes associated with normal aging.

2.1 Participants from HCP-Aging dataset

The participants included in the HCP-A dataset are healthy volunteers, representing normal conditions for their age group. They have no identified pathological causes of cognitive decline, such as stroke, tumor, or clinical diagnosed dementia [20]. All participants provided informed, written consent for this Institutional Review Board (IRB) approved cross-sectional study. Data from participants older than 90 years were excluded due to the absence of precise age. More details regarding the cognitive assessments and additional inclusion and exclusion criteria for the HCP-A dataset have been described in the previous study [21]. In this study, 641 subjects (age: 36–90 years; mean ± SD = 60 ± 16 years; females/males = 359/282) who underwent the complete imaging protocol, including structural, perfusion, and diffusion MRI.

2.2 Imaging acquisition

MR images of HCP-A were acquired with 3T using 32-channel head coil [20]. The imaging protocols included: 1) T1-weighted structural imaging was performed using multi-echo MPRAGE: TR/TI = 2500/1000 ms, TE = 1.8/3.6/5.4/7.2 ms, flip angle (FA) = 8°, voxel size = 0.8 mm isotropic, number of echoes = 4; 2) T2-weighted structural imaging was performed using a variable-flip-angle turbo-spin-echo (TSE) sequence: TR/TE = 3200/564 ms, turbo factor = 314, voxel size = 0.8 mm isotropic; 3) pCASL data were acquired using a single-shot 2D multi-band echo-planar imaging (MB-EPI) readout with the following parameters [22]: TR/TE = 3580/19 ms, partial Fourier = 6/8, labeling duration = 1500 ms, five post-labeling delays (PLDs) = 200 ms (control/label pairs = 6), 700 ms (control/label pairs = 6), 1200 ms (control/label pairs = 6), 1700 ms (control/label pairs = 10), and 2200 ms (control/label pairs = 15). A simultaneous multi-slice acquisition (SMS) with a multiband factor of 6 was used to achieve a 2.5mm isotropic resolution (2.27mm slice thickness plus 10% gap). The total acquisition time was approximately 5.5 minutes. Two 2.5 mm isotropic proton density (PD)-weighted equilibrium magnetization (M0) images were acquired at the end of scan for calibration. A pair of phase-encoding reversed spin-echo EPI scans that matched ASL data were acquired for distortion correction. More details of this advanced ASL acquisition were also described previously [23, 24]. 4) Diffusion imaging was performed using the multiband 2D spin-echo EPI with following parameters: TE/TR = 89.20/3230 ms, 1.5mm isotropic, MB factor = 4, 28 non-diffusion-weighted (b0) images and 2 b shells (b = 1500/3000 s/mm²) with 92–93 directions per shell, acquisition time = 21 min. More detailed imaging parameters were described [20] and available on: https://www.humanconnectome.org/study/hcp-lifespan-aging/project-protocol/imaging-protocols-hcp-aging.

2.3 Improvement of the choroid plexus segmentation

The initial processing of T1-weighted anatomical images was conducted automatically using the FreeSurfer-based HCP minimal processing pipelines (http://surfer.nmr.mgh.harvard.edu/) [25, 26]. However, due to the ChP having a small volumetric size and being immersed within the CSF, the segmentation obtained from FreeSurfer lacks sufficient accuracy. To overcome this limitation, a Bayesian Gaussian mixture model (GMM) was implemented in a stepwise manner (https://github.com/EhsanTadayon/choroid-plexus-segmentation). This method improved the differentiation of signal intensity between the CSF, the ventricular wall, and ChP voxels within the lateral ventricles, which in turn enhanced the accuracy of the segmentation [27].

2.4 Image processing

The open-source HCP-ASL pipeline (https://github.com/physimals/hcp-asl) was utilized for the ASL-MRI image processing. The initial step involved correcting the label and control ASL images for a linear intensity variation, known as the banding effect, within each 10-slice band, caused by magnetization transfer (MT) effects. The data was also corrected for a slice-wise SMS-related saturation-recovery intensity variation by fitting the saturation recovery curve [28]. Motion correction was applied to improve saturation recovery fitting, and alignment of the de-banded ASL data [28]. The phase encoding-reversed spin-echo images were used for susceptibility distortion correction using TOPUP [29]. Following alignment with the pre-processed T1w image, perfusion analysis was conducted on ASL data, while the original 2.5 mm isotropic ASL voxel grid was preserved. After pre-processing, a general linear model (GLM) was used to perform the label-control subtraction. This helped rescale the perfusion-weighted signal and separate it from variations of through-slice motion [30]. The computation of CBF and ATT was carried out based on a Bayesian Inference for Arterial Spin Labeling (BASIL) toolkit implemented in FSL (version 6.0.5, Oxford, UK), with partial volume correction [31–33]. It allows fitting a two-compartment model to estimate the tissue and macrovascular contribution to the perfusion signal separately generating CBF and ATT maps [34, 35].

Diffusion Tensor Imaging (DTI) was preprocessed using MRtrix3 [36], including denoise, Gibbs' ringing artifacts removal, distortion correction, and eddy current correction. After preprocessing, diffusion tensor was calculated and metrics including mean diffusivity (MD), and anisotropy (FA) were extracted afterwards.

The detailed image postprocessing steps are illustrated in Fig. 1. For each participant, the b0 image was registered to the T2w image, which exhibit similar image contrast, using 6-parameter linear rigid registration. The calculated transformation matrix was then applied to the corresponding MD map and the mean MD value within the ROI of ChP was calculated. CBF and ATT values was calculated in T1w space within the GM, WM, and ChP.

2.5 Statistical analysis

We first employed the extra sum-of-squares F test to determine whether a linear or nonlinear model provided a better fit for the data. The test outcomes indicated a preference for a second order polynomial model for the volume and MD changes with aging, while a linear model was optimal for the correlation between age and CBF. The impact of age on the volume, CBF, and MD of the ChP was evaluated, accounting for sex as a covariate. The participants were roughly equally divided into three age categories: 36–49 years (33%), 50–69 years (37%), and 70–90 years (30%). Independent t-tests were conducted within each age bracket to discern the disparities between males and females in these measurements. One-way ANOVA analyses were employed to underscore the variations in perfusion attributes among the ChP, GM, and WM tissues. These analyses compared CBF and ATT values of different tissue types within each age group with Tukey’s multiple comparison correction. Partial correlations were computed between the CBF and ChP volume, and between the MD and ChP volume, after adjusting for age and sex. To reveal the difference between the ChP with and without the presence of cyst-like structures, independent t-tests were employed to compare the MD and CBF across different age groups. The Bonferroni correction was implemented to adjust for multiple comparisons, setting the significance level at an adjusted P < 0.016.

3.1 Demographic characteristics

We exploited 641 healthy participants aged from 36 to 90 years old (60±16 years; 282 males) from HCP-Aging dataset. Detailed demographic data and MRI measurements from different age quartiles can be find in Table 1.

Table 1. Detailed demographic data and MRI measurements.

	All subjects 36-90 years (n=641)	Q1 36-49 years (n=211)	Q2 50-69 years (n=243)	Q3 70-90 years (n=187)
Age (years)	60±15	43±4	60±6	79±5
Sex (F/M)	359/282	123/88	136/107	100/87
Vol_ChP/ICV (_*10^-3)	1.93±0.59	1.56±0.45	1.86±0.52	2.17±0.65
MD (mm²/s*10^-3)	2.08±0.73	1.86±0.57	1.96±0.64	2.50±0.83
CBF (ml/100g/min)
ChP	58.85±18.05	65.6±17.6	58.2±15.8	49.8±15.3
GM	48.29±11.42	51.9±10.9	48.0±9.7	43.1±11.7
WM	21.64±5.36	22.3±5.1	21.7±4.9	20.8±6.1
ATT (seconds)
ChP	1.39±0.11	1.33±0.1	1.40±0.1	1.46±0.1
GM	1.44±0.12	1.36±0.1	1.44±0.1	1.51±0.1
WM	1.48±0.06	1.45±0.05	1.48±0.05	1.50±0.06

3.2 Age-related alterations in the choroid plexus

As shown in Figure 2, T1w-MRI images show an enlargement of the ChP in elderly individuals. On T2w-MRI, the ChP appears less densely packed, with fewer hypo-intensive vessels and cyst formation. By combining T1w and T2w data and applying T2w²/T1w ratio, the ChP contrast is improved due to different T2 relaxation times between ChP blood and CSF. This results in a more defined outline of cyst-like structures. The outline of these cyst-like structures is improved on MD maps as well, which exhibit lower MD than CSF, making them more distinguishable.

3.3 Effects of sex on the measurements of the choroid plexus
As shown in Figure 4, after adjusting the intracranial volume, there was no significant difference between males and females in terms of the ChP volume for all age groups (P30-50 = 0.05, P50-70 = 0.14, P70-90 = 0.96), even though male showed a faster ChP expansion. The CBF of the ChP was higher in the females compared to the males for all 3 age groups (P30-50 < 0.0001, P50-70 < 0.0001, P70-90 < 0.0001). No significant differences were observed between males and females regarding the MD values (P30-50 = 0.49, P50-70 = 0.09, P70-90 = 0.23).

3.4 Perfusion characteristics of the ChP

One-way ANOVA followed by Tukey’s multiple comparison correction showed significant differences between the perfusion characteristics of ChP, GM, and WM. Detailed values are provided in Table 1. For all age groups, ChP has the highest CBF and the shortest ATT among the three tissue types, with the order of CBF: ChP > GM > WM, and ATT: ChP < GM < WM (P < 0.0001)_.

Both CBF in the ChP (CBF_ChP) and GM (CBF_GM) decreased with age (r = -0.40, P < 0.001; r = -0.37, P < 0.001, respectively). However, the rate of decline in CBF was faster in the ChP compared to the GM (𝛽_ChP = -0.54, 𝛽_GM= -0.34) (Figure 5A). After adjusting for age and sex as covariates, a negative correlation was observed between ChP volume and CBF (r = -0.40, P <0.001) (Figure 5B). However, no significant correlation was found between CBF and MD (P = 0.1).

3.5 Observation of cyst-like structures in the ChP

Upon evaluating various modalities across all participants, we observed that cyst-like structures within the ChP are more common in the older population. These structures are distinguished by a sparse vasculature on the cyst wall and restricted diffusion or low MD values. As shown in Figure 2, an 81-year-old subject with an enlarged ChP exhibits a cyst-like structure with hypo-to-iso-intense T1 and hyperintense T2 signals. These signals are similar with CSF intensity, but with lower MD than CSF. Figure 6A presents illustrative images from age-matched subjects with and without ChP cysts, demonstrating that cysts have lower MD and CBF values compared to the non-cyst ChP. The frequency of cyst occurrence, calculated for each decade, increases with age (Figure 6B). There is a significantly decrease in CBF in ChP with cysts compared to non-cyst ChP in the 70-90 years age group (P = 0.005). Conversely, in the 30-50 years age group, CBF was slightly higher in ChP with cysts than in those without cysts (P = 0.01) (Figure 6C).

In this research, we assessed age-related changes of ChP using perfusion and microstructural imaging in a large, well-characterized healthy cohort from the HCP-A datasets, aged between 36 and 90. In line with previous studies, we noticed an enlargement of the ChP associated with age, likely due to stromal fibrosis, dystrophic degeneration, calcification, and lipofuscin deposition [5]. Despite the increase in ChP volume with age, we noted a decrease in ChP blood flow. This seemingly contradictory relationship between blood flow and ChP size could indicate a hypertrophic response to chronic hypoperfused status. Unlike the ChP hyperplasia found in pediatrics, which leads to CSF overproduction [37], the enlargement of ChP in adults might be due to the accumulation of damaged cells and extracellular matrix components, or stromal fibrosis. Diffusion MRI revealed an increase in MD values with age, suggesting increased water mobility due to microstructural damage [38, 39]. The ChP exhibited the highest CBF, and shortest ATT compared to GM and WM, reflecting its highly vascularized nature essential for CSF filtration and secretion. The ChP showed a faster decline in CBF with aging compared to the GM. Despite the ChP's abundant anastomoses, it might be more susceptible to age-related capillary loss. This could lead to inadequate CSF production rather than ischemic pathology, given its specific functions.

The glymphatic system, a recently identified waste removal system, facilitates a continuous exchange between CSF and interstitial fluid (ISF) via convective influx of CSF along the periarterial space [8]. Although the role of ChP in the glymphatic system remains largely unclear, the production of CSF and its subsequent entry along the perivascular space is considered crucial for ISF-CSF exchange and clearance function. The glymphatic influx serves as an essential route for the distribution of electrolytes, macromolecules, and other larger compounds that predominantly enter the brain through the BCSFB at the ChP, as hypothesized [40]. In an average adult, the ChP produces approximately 500 ml CSF per day, which circulates about 150ml at any given time within the brain and spinal cord. We considered that the ChP acts as the driving force for the CSF-ISF circulation within the brain [40]. The age-related decrease in CSF production due to a reduction in CBF, in conjunction with microstructural modifications to the ChP and the BCSFB, may lead to dysfunction in waste removal. This could potentially contribute to the accumulation of misfolded and hyperphosphorylated proteins, thereby increasing the brain’s vulnerability to the onset of neurodegenerative pathologies or accelerating the progression of cognitive dysfunction [40].

The observed decrease in ChP perfusion with age is consistent with findings from Alisch et al. [10], who reported a decrease in ChP perfusion with aging using single post-labeling delay pCASL MRI. Vascular aging can lead to increased vessel tortuosity due to changes in the mechanical properties of the vessel walls and the pulsatile hemodynamic forces acting on them. This is particularly evident in structures with rich vascular network, such as ChP. Over time, these vessels undergo structural and functional alterations, including the weakening of the vessel wall and a decline in elasticity. This process, often referred to as arterial stiffening, results from the accumulation of collagen, fragmentation of elastin fibers, and the presence of calcification [41]. As a result of these changes, the vessel wall’s capacity to withstand mechanical stress decreases, making the vessel more prone to deformation due to the pulsatile blood flow, and higher pressure on the tortuous vessel wall during each cardiac cycle over decades. Animal studies using immunochemical staining have shown an increase in ChP vessel tortuosity with age [42]. Our previous studies have also identified age-related vascular tortuosity in both large [43] and small [44] brain arteries in humans. We believe that vascular tortuosity and stiffness combined with decreased vascular density [45] could be a key factor in the reduced ChP perfusion observed with aging.

Eisma et al. [18] utilized PC-MRI and observed a decrease in net CSF flow through the aqueduct of Sylvius (in the cranial to caudal direction), which directly correlates with ChP perfusion. Studies conducted on both humans and in rodents have revealed a reduction in CSF production and turnover in the aging brain [46]. Meanwhile, an increase in the concentrations of Aβ₁₋₄₀ and Aβ₁₋₄₂ in the cortex and hippocampus of aging mice has been reported [47]. This suggests that the decrease in CSF turnover and the ongoing accumulation of Aβ in the brain contribute to AD pathology, potentially highlighting the link between CSF dynamics and waste solute clearance. The decline in CBF, in conjunction with ChP enlargement, may reflect a compensatory mechanism involving dystrophic hyperplasia triggered by reduced perfusion [48]. If glymphatic CSF to interstitial flow is compromised, the ChP could potentially increase CSF production to compensate and facilitate the glymphatic clearance of waste products from the interstitial space [48]. In the aged population, the epithelial cells of the ChP shorten and the amount of Aquaporin-1 (AQP1) protein decreases, which diminishes the capacity to produce additional CSF. While the exact mechanism remains to be established, we propose that the ChP might instead undergo non-functional stromal expansion involving stimulation from immunoreactivity [7, 49], though the exact mechanism remains unclear.

The observed increase in MD with aging could potentially signify a loosening of the ChP structure and a decrease in the microstructural integrity of the BCSFB. MD measures the overall diffusion and motion of water molecules, with higher MD values indicating greater water mobility. This finding is consistent with a previous study [10], that reported decreased FA, and increased T1 and T2 values. However, in our study, we did not find a significant correlation between FA values and age, nor any sex-based differences in MD values. This discrepancy could be due to variations in segmentation and registration methodologies. Furthermore, the ChP is not a highly anisotropic structure, and FA is less sensitive to its microstructural changes compared to MD.

It is worth noting that the CBF in the ChP is significantly higher in females than in males across all age groups. This pattern aligns with previous studies that indicate greater whole-brain and GM perfusion in females compared to males [50, 51]. This phenomenon could potentially be attributed to the effects of estrogens [52]. Furthermore, the rate of CBF reduction in the ChP with age was faster in females. While no significant sex differences were observed in the ICV-adjusted ChP volume and MD values, the rates of age-related changes for ChP enlargement and MD values were higher in females. This suggests a greater susceptibility in females to age-related perfusion and microstructural changes in the ChP. It has been suggested that males and females with AD exhibit different cognitive and psychiatric symptoms, and females demonstrate faster cognitive decline after diagnosis of AD dementia [53]. In addition, a significantly weaker glymphatic function, as indicated by neuronal activity-CSF coupling strength, has been observed in females than in males [54]. Taken together, these sex effects within the ChP may contribute to the sex-specific clinical and pathological findings of age-related neurodegenerative diseases, considering the interactions between ChP, the glymphatic system, and AD.

Despite the age-associated increase in MD, a higher prevalence of cyst-like structures, characterized by restricted water movement and thus lower MD, was noted in older individuals. Surgical findings have confirmed that ChP cysts contain a serous fluid akin to CSF, with the protein concentration in the cyst fluid being slightly elevated compared to that of CSF [55]. Histologically, the thickened walls of the cysts were composed of abundant connective tissue with blood vessels and foci of calcification [56]. The contents of these connective tissue-lined cysts was gelatinous and tested positive for fibrinogen and albumin immunohistochemical staining, indicating a highly proteinaceous composition [56]. An MRI study employing oscillating gradient spin-echo (OGSE) DWI with shortened diffusion time suggested that the lower ADC values of the ChP cysts, compared to free CSF [57], imply the existence of spatially restricted diffusion and increased viscosity of the cysts. Due to their various lipid and blood components, these structures can present heterogeneous contrast on MR images. Without histological validation, differentiation age-related cystic changes with ChP xanthogranulomas (XGs) with psammoma bodies can be challenging, as both exhibit similar characteristics on DWI. XGs are benign lesions with an autopsy incidence of 1.6–7%, which is lower than the rate of cysts observed in our study. It has been validated that the cyst wall consists of rich connective tissue and foci of calcification. The restricted water movement in aged ChP could be due to calcium deposits, cholesterol crystal accumulation, or gelatinous fluid. Although the formation of these cysts is generally benign, their continued growth could result in the gradual destruction of choroidal villi and increased dystrophic hyperplasia [58]. Regardless of the cyst type, we propose that these structures could negatively impact the vascular component of the ChP and potentially disrupt its homeostatic function in the brain.

From a technical standpoint, the HCP-A dataset provides high quality pCASL MRI with multiple PLDs, allowing for simultaneous measurements of ATT and CBF. While segmented 3D PCASL is recommended for ASL in clinical applications, it is susceptible to head movement, especially in older individuals. The ASL sequence in the HCP-A protocol incorporates cutting-edge developments in data acquisition like using multi-band echo planar imaging (MB-EPI) multi-slice (SMS) imaging to reduce acquisition time and minimize motion artifacts in whole brain perfusion imaging. However, the application of SMS may result in signal voids at imaging band edges. According to Li et al. [23], the geometry factor (g-factor) and the total leakage factor (TLF)-induced confounding effects, have a minimal impact on the CBF estimation.

Certain limitations of the study should be acknowledged. Firstly, the ChP is immersed within the CSF, which necessitates consideration of partial volume effects (PVEs). However, we consider that the large dataset and high-quality images used in this study could help counterbalance these PVEs. Furthermore, the application of the GMM method to improve segmentation could also aid in bias reduction. A combination of T1w and T2w images (T2w²/T1w) was also used to augment the tissue contrast between CSF and ChP, sharpening their boundary. The enhanced vascular contrast on this ratio image is primarily due to the hypointense signal of the vascular glomus on T2, caused by flow voids, and slightly hyperintense T1 signal relative to CSF, which enhances the tissue contrast. The T2 relaxation time of the fluid within cyst-like structure might differ from that of the ChP blood and CSF. Therefore, the use of the T2w²/T1w ratio could amplify the contrast between these compartments. This improves visualization of the cyst-like structures and enables more accurate neuromorphometric analyses [59]. Our prior study using high-resolution contrast-enhanced MRI on 7T revealed detailed ChP anatomy, showing age-related decrease in vascular density [45] and corroborating the current study’s findings of reduced CBF with age. Secondly, this study does not include clinical data such as medication usage and vascular risk factors (i.e., blood pressure, glucose level, and cholesterol level). These factors could potentially influence vascular health and cognitive function [21]. However, it is crucial to highlight that, based on prior research, aging is the most significant risk factor for numerous neurodegenerative disorders. For example, we did not observe significant effects of these factors on the ChP in our prior study [45]. Future research should include these parameters for a more comprehensive evaluation.

The large-scale HCP-Aging dataset, covering a wide age range, provides high-quality structural, diffusion, and perfusion MR images. Our findings show that the ChP undergoes age-related changes, including enlargement, CBF decline, and elevated MD. These alterations may result from stromal expansion, reduced vascular density, and compromised BCSFB integrity, potentially leading to insufficiency in CSF production, waste accumulation, and impaired exchange between CSF and ISF in the glymphatic system, ultimately affecting cognitive function.

Grant Support

This work was supported in part by the NIH U01AG052564, P30AG066512, P01AG060882, RF1 NS110041, R01 NS108491, U24 NS135568.

6. Acknowledgements

The HCP-Aging 2.0 Release data used in this report came from DOI: 10.15154/1520707 and was supported by the National Institute of Mental Health of the National Institutes of Health under Award Number U01MH109589 and by funds provided by the McDonnell Center for Systems Neuroscience at Washington University in St. Louis.

7. Funding

This work was supported in part by the NIH U01AG052564, P30AG066512, P01AG060882, RF1 NS110041, R01 NS108491, U24 NS135568.

8. Ethics approval and consent to participate

Not applicable

9. Consent for publication

All authors have approved the manuscript submission. The content of this manuscript has not been published or submitted for publication elsewhere.

10. Availability of data and materials

Data in this study may be shared with researchers upon reasonable requests to the corresponding author.

11. Competing interests

Not applicable

12. Authors' contributions - provide individual author contribution

Conceptualization, Y.G., and Z.S.; Methodology, Y.G., J.Z., Z.S., and C.L.; Writing – Original Draft, Z.S.; Writing –Review & Editing, Z.S., C.L., T.W., and Y.G.; Funding Acquisition, Y.G.; Supervision, Y.G.

13. Authors' information

Department of Radiology, NYU Grossman School of Medicine, New York, NY, USA

Zhe Sun, Chenyang Li, Jiangyang Zhang, Yulin Ge

Vilcek Institute of Graduate Medical Sciences, NYU Grossman School of Medicine, New York, NY, USA

Zhe Sun, Chenyang Li

Departments of Neurology, Pathology and Psychiatry, NYU Grossman School of Medicine, New York, NY, USA

Thomas Wisniewski

Center for Cognitive Neurology, NYU Grossman School of Medicine, New York, NY, USA

Thomas Wisniewski

Hutton, D., et al., Choroid plexus and CSF: an updated review. Br J Neurosurg, 2022. 36(3): p. 307-315.
Lun, M.P., E.S. Monuki, and M.K. Lehtinen, Development and functions of the choroid plexus-cerebrospinal fluid system. Nat Rev Neurosci, 2015. 16(8): p. 445-57.
Liddelow, S.A., et al., Development of the lateral ventricular choroid plexus in a marsupial, Monodelphis domestica. Cerebrospinal fluid research, 2010. 7: p. 1-10.
Oreskovic, D., M. Rados, and M. Klarica, Role of choroid plexus in cerebrospinal fluid hydrodynamics. Neuroscience, 2017. 354: p. 69-87.
Serot, J.M., M.C. Bene, and G.C. Faure, Choroid plexus, aging of the brain, and Alzheimer's disease. Front Biosci, 2003. 8: p. s515-21.
Serot, J.M., et al., Choroid plexus and ageing in rats: a morphometric and ultrastructural study. Eur J Neurosci, 2001. 14(5): p. 794-8.
Prineas, J.W., J.D. Parratt, and P.D. Kirwan, Fibrosis of the Choroid Plexus Filtration Membrane. J Neuropathol Exp Neurol, 2016. 75(9): p. 855-67.
Iliff, J.J., et al., A paravascular pathway facilitates CSF flow through the brain parenchyma and the clearance of interstitial solutes, including amyloid beta. Sci Transl Med, 2012. 4(147): p. 147ra111.
Chen, R.L., et al., Age-related changes in choroid plexus and blood-cerebrospinal fluid barrier function in the sheep. Exp Gerontol, 2009. 44(4): p. 289-96.
Alisch, J.S.R., et al., Characterization of Age-Related Differences in the Human Choroid Plexus Volume, Microstructural Integrity, and Blood Perfusion Using Multiparameter Magnetic Resonance Imaging. Front Aging Neurosci, 2021. 13: p. 734992.
Choi, J.D., et al., Choroid Plexus Volume and Permeability at Brain MRI within the Alzheimer Disease Clinical Spectrum. Radiology, 2022. 304(3): p. 635-645.
Jeong, S.H., et al., Association of choroid plexus volume with motor symptoms and dopaminergic degeneration in Parkinson's disease. J Neurol Neurosurg Psychiatry, 2023. 94(12): p. 1047-1055.
Akaishi, T., J. Fujimori, and I. Nakashima, Enlarged choroid plexus in multiple sclerosis is associated with increased lesion load and atrophy in white matter but not gray matter atrophy. Mult Scler Relat Disord, 2024. 82: p. 105424.
Zhao, L., et al., Non-invasive measurement of choroid plexus apparent blood flow with arterial spin labeling. Fluids Barriers CNS, 2020. 17(1): p. 58.
Evans, P.G., et al., Non-Invasive MRI of Blood-Cerebrospinal Fluid Barrier Function. Nat Commun, 2020. 11(1): p. 2081.
Petitclerc, L., et al., Ultra-long-TE arterial spin labeling reveals rapid and brain-wide blood-to-CSF water transport in humans. Neuroimage, 2021. 245: p. 118755.
Perera, C., et al., Pharmacological MRI with Simultaneous Measurement of Cerebral Perfusion and Blood-Cerebrospinal Fluid Barrier Function using Interleaved Echo-Time Arterial Spin Labelling. NeuroImage, 2021. 238: p. 118270.
Eisma, J.J., et al., Choroid plexus perfusion and bulk cerebrospinal fluid flow across the adult lifespan. J Cereb Blood Flow Metab, 2023. 43(2): p. 269-280.
Alexander, A.L., et al., Diffusion tensor imaging of the brain. Neurotherapeutics, 2007. 4(3): p. 316-329.
Harms, M.P., et al., Extending the Human Connectome Project across ages: Imaging protocols for the Lifespan Development and Aging projects. Neuroimage, 2018. 183: p. 972-984.
Bookheimer, S.Y., et al., The Lifespan Human Connectome Project in Aging: An overview. Neuroimage, 2019. 185: p. 335-348.
Li, X., et al., Pushing the limits of asl imaging for the lifespan human connectome projects. International Society for Magnetic Resonance Imaging.[Google Scholar], 2018.
Li, X., et al., Theoretical and experimental evaluation of multi-band EPI for high-resolution whole brain pCASL Imaging. Neuroimage, 2015. 106: p. 170-81.
Juttukonda, M.R., et al., Characterizing cerebral hemodynamics across the adult lifespan with arterial spin labeling MRI data from the Human Connectome Project-Aging. Neuroimage, 2021. 230: p. 117807.
Fischl, B., M.I. Sereno, and A.M. Dale, Cortical surface-based analysis. II: Inflation, flattening, and a surface-based coordinate system. Neuroimage, 1999. 9(2): p. 195-207.
Glasser, M.F., et al., The minimal preprocessing pipelines for the Human Connectome Project. Neuroimage, 2013. 80: p. 105-24.
Tadayon, E., et al., Improving Choroid Plexus Segmentation in the Healthy and Diseased Brain: Relevance for Tau-PET Imaging in Dementia. J Alzheimers Dis, 2020. 74(4): p. 1057-1068.
McConnell, F.K., et al., The minimal processing pipeline for arterial spin labeling data from the Human Connectome Project Lifespan studies of Aging and Development. Proc. Intl. Soc. Mag. Reson. Med. 30, 2022. 0905.
Andersson, J.L., S. Skare, and J. Ashburner, How to correct susceptibility distortions in spin-echo echo-planar images: application to diffusion tensor imaging. Neuroimage, 2003. 20(2): p. 870-88.
Suzuki, Y., et al., A framework for motion correction of background suppressed arterial spin labeling perfusion images acquired with simultaneous multi-slice EPI. Magnetic Resonance in Medicine, 2019. 81(3): p. 1553-1565.
Kirk, T.F., et al., Toblerone: Surface-Based Partial Volume Estimation. IEEE Trans Med Imaging, 2020. 39(5): p. 1501-1510.
Chappell, M.A., et al., Variational Bayesian Inference for a Nonlinear Forward Model. IEEE Transactions on Signal Processing, 2009. 57(1): p. 223-236.
Chappell, M.A., et al., Partial volume correction of multiple inversion time arterial spin labeling MRI data. Magn Reson Med, 2011. 65(4): p. 1173-83.
Chappell, M.A., et al., Separation of macrovascular signal in multi-inversion time arterial spin labelling MRI. Magn Reson Med, 2010. 63(5): p. 1357-65.
Groves, A.R., M.A. Chappell, and M.W. Woolrich, Combined spatial and non-spatial prior for inference on MRI time-series. Neuroimage, 2009. 45(3): p. 795-809.
Tournier, J.D., et al., MRtrix3: A fast, flexible and open software framework for medical image processing and visualisation. Neuroimage, 2019. 202: p. 116137.
Kim, J.W., et al., Choroid Plexus Hyperplasia : Report of Two Cases with Unique Radiologic Findings. J Korean Neurosurg Soc, 2024. 67(3): p. 376-381.
Alicioglu, B., et al., Diffusion-weighted magnetic resonance imaging in the assessment of choroid plexus aging. Neuroradiol J, 2017. 30(5): p. 490-495.
Solár, P., et al., Choroid plexus and the blood–cerebrospinal fluid barrier in disease. Fluids and Barriers of the CNS, 2020. 17(1): p. 35.
Jessen, N.A., et al., The Glymphatic System: A Beginner's Guide. Neurochem Res, 2015. 40(12): p. 2583-99.
Heinz, A., Elastic fibers during aging and disease. Ageing Research Reviews, 2021. 66: p. 101255.
Dani, N., et al., A cellular and spatial map of the choroid plexus across brain ventricles and ages. Cell, 2021. 184(11): p. 3056-3074 e21.
Sun, Z., et al., Age-Related Tortuosity of Carotid and Vertebral Arteries: Quantitative Evaluation With MR Angiography. Front Neurol, 2022. 13: p. 858805.
Sun, Z., et al., In Vivo Detection of Age-Related Tortuous Cerebral Small Vessels using Ferumoxytol-enhanced 7T MRI. Aging Dis, 2024. 15(4): p. 1913-1926.
Sun, Z., et al., Vascular Aging in the Choroid Plexus: A 7T Ultrasmall Superparamagnetic Iron Oxide (USPIO)-MRI Study. J Magn Reson Imaging, 2024.
Simon, M.J. and J.J. Iliff, Regulation of cerebrospinal fluid (CSF) flow in neurodegenerative, neurovascular and neuroinflammatory disease. Biochim Biophys Acta, 2016. 1862(3): p. 442-51.
Kawarabayashi, T., et al., Age-dependent changes in brain, CSF, and plasma amyloid (beta) protein in the Tg2576 transgenic mouse model of Alzheimer's disease. J Neurosci, 2001. 21(2): p. 372-81.
Johnson, S.E., et al., Choroid plexus perfusion and intracranial cerebrospinal fluid changes after angiogenesis. J Cereb Blood Flow Metab, 2020. 40(8): p. 1658-1671.
Solar, P., et al., Choroid plexus and the blood-cerebrospinal fluid barrier in disease. Fluids Barriers CNS, 2020. 17(1): p. 35.
Parkes, L.M., et al., Normal cerebral perfusion measurements using arterial spin labeling: Reproducibility, stability, and age and gender effects. Magnetic Resonance in Medicine, 2004. 51(4): p. 736-743.
Bertsch, K., et al., Resting cerebral blood flow, attention, and aging. Brain Research, 2009. 1267: p. 77-88.
Kastrup, A., et al., Gender-related effects of indomethacin on cerebrovascular CO2 reactivity. Journal of the neurological sciences, 1999. 162(2): p. 127-132.
Ferretti, M.T., et al., Sex differences in Alzheimer disease — the gateway to precision medicine. Nature Reviews Neurology, 2018. 14(8): p. 457-469.
Han, F., et al., Reduced coupling between cerebrospinal fluid flow and global brain activity is linked to Alzheimer disease-related pathology. PLoS Biol, 2021. 19(6): p. e3001233.
Odake, G., H. Tenjin, and N. Murakami, Cyst of the choroid plexus in the lateral ventricle: case report and review of the literature. Neurosurgery, 1990. 27(3): p. 470-476.
Kinoshita, T., et al., Clinically silent choroid plexus cyst: evaluation by diffusion-weighted MRI. Neuroradiology, 2005. 47(4): p. 251-255.
Maekawa, T., et al., Choroid plexus cysts analyzed using diffusion-weighted imaging with short diffusion-time. Magn Reson Imaging, 2019. 57: p. 323-327.
Jovanovic, I., et al., Psammoma bodies - friends or foes of the aging choroid plexus. Med Hypotheses, 2010. 74(6): p. 1017-20.
Misaki, M., et al., Contrast enhancement by combining T1- and T2-weighted structural brain MR Images. Magnetic Resonance in Medicine, 2015. 74(6): p. 1609-1620.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
28 Oct, 2024
Reviews received at journal
13 Oct, 2024
Reviewers agreed at journal
05 Oct, 2024
Reviews received at journal
02 Oct, 2024
Reviewers agreed at journal
27 Sep, 2024
Reviewers agreed at journal
22 Sep, 2024
Reviewers invited by journal
10 Sep, 2024
Editor assigned by journal
06 Sep, 2024
Submission checks completed at journal
06 Sep, 2024
First submitted to journal
01 Sep, 2024

You are reading this latest preprint version

Choroid Plexus Aging: Structural and Vascular Insights from the HCP-Aging Dataset

Status:

Version 1

Abstract

Figures

1. Background

2. Methods

2.1 Participants from HCP-Aging dataset

2.2 Imaging acquisition

2.3 Improvement of the choroid plexus segmentation

2.4 Image processing

2.5 Statistical analysis

3. Results

4. Discussion

5. Conclusion

Declarations

References

Additional Declarations

Status:

Version 1