Choroid plexus alterations in long COVID and their associations with Alzheimer's disease risks

37 min read Original article ↗

Abstract

INTRODUCTION

Choroid plexus (ChP) enlargement is a neuroimaging biomarker of neuroinflammation and neurodegeneration. However, evidence of ChP structural and perfusion alterations in long coronavirus disease (COVID) and their clinical relevance remains limited.

METHODS

This study included 86 long COVID, 67 recovered COVID, and 26 COVID‐negative healthy controls (HCs). ChP volume and cerebral blood flow (CBF) were quantified, and their associations with Alzheimer's disease (AD) symptoms and plasma biomarkers were examined.

RESULTS

Both patient groups showed higher ChP volume and lower CBF than HC. Relative to recovered COVID, long COVID patients had a larger ChP volume, but no significant difference in CBF. ChP volume correlated positively with glial fibrillary acidic protein (r = 0.35) and phosphorylated tau217 (p‐tau217; r = 0.54), while CBF correlated negatively with p‐tau217 (r = –0.56). Both ChP volume and CBF were associated with cognitive decline measured with Mini‐Mental State Examination and Clinical Dementia Rating.

DISCUSSION

These findings suggest that ChP differences in long COVID are associated with AD‐related cognitive decline and increased plasma biomarkers.

Highlights

  • Long coronavirus disease (COVID) patients show choroid plexus (ChP) enlargement and reduced cerebral blood flow.

  • ChP alterations are associated with Alzheimer's disease (AD)‐related symptoms and plasma biomarker changes.

  • ChP alterations on magnetic resonance imaging may serve as imaging markers for tracking neurological symptoms and AD‐related pathology in post‐COVID patients.

Keywords: Alzheimer's disease, blood flow, choroid plexus, long coronavirus disease, neurological symptoms, plasma biomarkers, volume

1. BACKGROUND

Coronavirus disease 2019 (COVID‐19), caused by the severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2), has had a profound impact on public health. Beyond the acute phase, infection can result in long‐term sequelae collectively referred to as long COVID. This is defined as symptoms persisting for at least 2 months and starting 3 months after a probable or confirmed SARS‐CoV‐2 infection. 1 Neurological manifestations are among the most prominent features of long COVID, including cognitive impairment, sleep disorders, olfactory dysfunction, psychiatric symptoms, and others. 2 In a cohort of 3305 patients, neurological symptoms were reported in 35.4% cases. 3 A recent longitudinal study further demonstrated that patients with long COVID may experience heightened neurological risks that remain for 2 years. 4 Notably, several of these neurological symptoms resemble those observed in patients with Alzheimer's disease (AD). The prevalence of dementia is higher among patients with COVID. 5 Besides, individuals with AD exhibit increased mortality after COVID‐19 infection, 6 suggesting potential overlaps in the pathogenic mechanisms of COVID‐19 and AD. Despite these observations, the mechanisms underlying AD‐related neurological symptoms in long COVID remain poorly understood.

Several hypotheses have been proposed for the pathogenesis of long COVID condition, including brain hypometabolism, 7 vascular injury from coagulopathy and endothelial dysfunction, 8 and neuroinflammation and immune dysregulation. 9 , 10 These processes may be driven by persistent SARS‐CoV‐2 reservoirs in tissues. 11 , 12 Additionally, long COVID has been associated with AD‐like signaling and pathology, including amyloidogenic processes such as accumulation of amyloid‐forming peptides, which may contribute to neuronal injury. 13 , 14 , 15 , 16 , 17 Although there is no consensus on a single pathological mechanism, neuroinflammation and immune dysfunction remain key hypotheses regarding the driving factors of the neurological sequelae of this post‐viral condition.

The choroid plexus (ChP) is a vasculature‐rich structure that serves as the primary site for cerebrospinal fluid (CSF) production, regulates immune responses, and maintains the blood–CSF barrier. Unlike the tightly regulated blood–brain barrier (BBB), the capillaries in the ChP are fenestrated and more permeable. These features make the ChP a potential entry route for peripheral inflammatory cells and blood‐borne pathogens to enter the central nervous system (CNS). 18 , 19 Experimental studies show that SARS‐CoV‐2 can infect and damage ChP epithelium, 19 , 20 , 21 leading to blood–CSF barrier (BCSFB) leakage. 21 , 22 Clinical observations further report abnormal CSF findings in long COVID patients with neurological symptoms. 23 , 24 , 25 , 26 , 27 Collectively, these findings underscore the ChP's pivotal role in neuroimmune regulation and its potential contribution to the neurological manifestations of long COVID.

Neuroimaging has emerged as a valuable tool for in vivo assessment of ChP. Recent imaging studies have reported increased ChP volume in long COVID subjects, 21 , 28 neuroinflammatory disorders, 29 normal aging, 30 and AD patients. 30 , 31 , 32 Despite the enlarged ChP volume in these conditions, the ChP functions, such as rate of CSF secretion 33 and glymphatic clearance 34 are reduced. Impaired blood perfusion in the ChP may further hinder CSF production, lead to waste buildup, and compromise the integrity of the BCSFB. Prior in vivo studies of normal aging have separately indicated structural and perfusion alterations in the ChP. 30 , 35 , 36 However, despite extensive research, no studies have explored the structural and perfusion alterations in long COVID and their associations with neurological symptoms and blood biomarkers, to date. Examining these alterations could provide new insights into BCSFB disruption and identify potential therapeutic targets for neurological symptoms in long COVID.

Therefore, this study included participants with long COVID, those who have recovered from COVID, and COVID‐negative healthy controls (HCs). Our objectives were 3‐fold: (1) to analyze structural and perfusion alterations in the ChP among individuals with long COVID compared to recovered subjects and HCs, (2) to explore the relationship between ChP alterations and neurological symptoms in long COVID subjects, and (3) to examine correlations between ChP imaging features and plasma biomarkers associated with neuroinflammation and AD‐related pathology.

RESEARCH IN CONTEXT

  1. Systematic review: The authors conducted a literature review using traditional databases (e.g., PubMed), meeting abstracts, and conference presentations with the search terms “COVID” and “choroid plexus (ChP).” While several studies have reported ChP enlargement in patients with long coronavirus disease (COVID), no prior work has investigated perfusion alterations and their clinical relevance. Relevant citations are appropriately included.

  2. Interpretation: Our findings demonstrate that patients with long COVID exhibit increased ChP volume and reduced cerebral blood flow. These imaging alterations were significantly correlated with Alzheimer's disease (AD)‐related plasma biomarkers and cognitive impairment. Collectively, these results highlight ChP assessment as a promising target for monitoring neurological dysfunction and AD‐related pathology after COVID infection.

  3. Future directions: Longitudinal studies across different states and stages of COVID, integrating comprehensive blood and cerebrospinal fluid biomarkers of neuroinflammation and AD‐related pathology, are needed to elucidate ChP alterations during post‐viral progression and to investigate the association between peripheral inflammation and central AD‐related processes.

2. METHODS

2.1. Participants

A total of 153 COVID‐19–positive patients and 26 HCs were recruited from the outpatient general medicine and neurology clinics at New York University (NYU) Langone Hospital between 2023 and 2025. The long COVID subgroup was defined by the following criteria: (1) a history of positive SARS‐Cov‐2 real‐time polymerase chain reaction or antigen test at least 3 months prior to study enrollment; and (2) neurological complaints temporally related to the SARS‐CoV‐2 infection. Exclusion criteria included: (1) presence or past medical history of structural brain lesions or CNS infection, (2) inability to complete neurological testing, (3) baseline modified Rankin score > 0 before the COVID‐19 pandemic, (4) time period < 6 weeks between acute COVID‐19 and first evaluation, and (5) severe white matter hyperintensities (WMHs; Fazekas grade > 2) on neuroimaging. The recovered COVID subgroup includes individuals who tested positive for SARS‐CoV‐2 nucleocapsid antibodies during the acute infection, whose symptoms resolved within 30 days of onset, and who were symptom free at enrollment. The HCs were defined as individuals with a negative SARS‐CoV‐2 nucleocapsid immunoglobulin G test result at the time of evaluation. This study was approved by the NYU Grossman School of Medicine Institutional Review Board. Written informed consent was obtained from all participants or their surrogates, as applicable.

2.2. Neurological and plasma blood tests

Clinical neurological symptoms, including sleep and olfactory disturbances, were assessed using the Epworth Sleepiness Scale (ESS) and the University of Pennsylvania Smell Identification Test (UPSIT), respectively. Cognitive impairment was evaluated using the Mini‐Mental State Examination (MMSE), Montreal Cognitive Assessment (MoCA), and Clinical Dementia Rating (CDR) scale.

Blood samples were collected into ethylenediaminetetraacetic acid (EDTA) vacutainers from a subsample of long COVID patients (neuroinflammation biomarkers, glial fibrillary acidic protein [GFAP], neurofilament light chain [NfL], and amyloid beta [Aβ]42/Aβ40: n = 64; phosphorylated tau217 [p‐tau217]: n = 41). Plasma fraction was separated and stored at –80°C until analysis. Inflammatory cytokines (human interleukin [hIL]‐6, hIL‐10, and tumor necrosis factor alpha [TNF‐α]) were measured with Corplex Cytokine 10‐Plex kits using Simoa SP‐X, and AD‐related biomarkers (NfL, GFAP, Aβ42, Aβ40, and p‐tau217) were measured with Neurology 4‐Plex Advantage kits and AlzPath pTau217, respectively, using Simoa HD‐X.

2.3. Magnetic resonance imaging acquisition

Magnetic resonance imaging (MRI) was performed on a 3T Siemens Prisma scanner equipped with a 64‐channel head coil. The imaging protocols included: (1) T1‐weighted structural imaging performed using T1 magnetization‐prepared rapid gradient echo (MPRAGE): repetition time (TR)/echo time (TE) = 2300/2.96 ms, field of view (FOV) = 256 × 256 mm2, voxel size = 1.0 × 1.0 × 1.0 mm3, slice number = 192; (2) 3D T2 fluid‐attenuated inversion recovery (FLAIR): inversion time (TI)/TR/TE = 1800/5000/394 ms, FOV = 250 × 230 mm2, voxel size = 0.5 × 0.5 × 1.0 mm3, slice number = 192; and (3) 3D pseudo‐continuous arterial spin labeling (pCASL) with gradient echo and spin echo (GRASE): TR/TE = 4120/12.56 ms, FOV = 220 × 220 mm2, voxel size = 3.4 × 3.4 × 4.0 mm3, slice number = 36, post labeling delay (PLD) = 2000 ms, four pairs of label and control images, labeling duration = 1800 ms, with background suppression. Proton density (PD)‐weighted equilibrium magnetization (M0) images were also acquired: TR/TE = 10,000/12.56 ms, FOV = 200 × 200 mm2, voxel size = 3.4 × 3.4 × 4.0 mm3, slice number = 36.

2.4. MRI processing

The pCASL data were processed using the ASL‐MRICloud platform. 37 The main processing steps included: (1) motion correction by aligning all control and label images to their respective first timepoint, (2) average of control and label images, (3) calculation of difference images by subtracting the averaged control‐label images, (4) estimation of cerebral blood flow (CBF) maps using a single‐compartment model with M0 images for quantitative CBF calculation (mL/100 g/min), and (5) co‐registration of CBF maps to individual T1‐weighted anatomical images with whole‐brain and regional CBF available for group comparison and correlations analyses. The parameters used for CBF quantification were as follows: blood T1 = 1650 ms, brain–blood partition coefficient = 0.9 mL/g, and labeling efficiency = 0.85.

2.5. ChP segmentation

For brain tissue segmentation preparation, 3D FLAIR images were registered to the T1‐weighted MPRAGE image using Advanced Normalization Tools (ANTs). Registration was performed with a rigid‐body transformation followed by affine alignment and non‐linear symmetric normalization. 

FIGURE 1.

FIGURE 1

Representative example of ChP segmentation and corresponding volume and CBF in each group. Long COVID patient demonstrates enlarged ChP volume and reduced CBF compared to recovered COVID and HC subjects. CBF, cerebral blood flow; ChP, choroid plexus; COVID, coronavirus disease; FLAIR, fluid‐attenuated inversion recovery; HC, healthy control; MPRAGE, magnetization‐prepared rapid gradient echo.

ChP segmentation was performed in three steps. First, anatomical T1‐weighted MPRAGE images were automatically processed using FreeSurfer software (version 7.0), and initial masks of the lateral ventricles and ChP were generated. We also calculated total intracranial volume (TIV), which was used as a nuisance in the following analysis. Second, a Gaussian mixture model (GMM) was applied to refine ChP segmentation within the lateral ventricles, enhancing the differentiation among CSF, ventricular wall, and ChP voxels, thereby improving segmentation accuracy. Finally, GMM‐derived masks were manually inspected and adjusted with reference to 3D FLAIR images to ensure precise delineation.

2.6. Statistical analysis

Data were reported as median (interquartile range [IQR]), mean ± standard deviation, or counts with percentages (%), according to standard statistical conventions. Group differences in demographic and clinical variables were assessed using the Kruskal–Wallis H test for continuous variables and chi‐squared tests for categorical variables. ChP volume differences across groups were analyzed using the Kruskal–Wallis H test, while CBF was compared using one‐way analysis of covariance, with age, sex, and TIV as covariates. Pairwise group comparisons were conducted using Tukey post hoc correction. Partial correlation analyses were used to examine associations between ChP imaging measures and neurological symptoms or plasma biomarkers while controlling the age, sex, and TIV covariates. Multiple comparisons were corrected using the false discovery rate (FDR). All statistical analyses were conducted using IBM SPSS statistics (version 28.0) and Python (version 3.9), and statistical significance was set at two‐tailed p < 0.05.

3. RESULTS

3.1. Demographic and clinical characteristics

Among 153 participants, 86 had long COVID, 67 had recovered COVID, and 26 were HCs. Long COVID patients (median age, 61.5 years) were significantly younger than both the recovered COVID (median age, 72 years) and HC (median age, 72 years) participants (p < 0.001). Compared to the other groups, long COVID patients showed higher prevalence of alcohol consumption (62.79%), hypertension (33.72%), and hypercholesterolemia (34.88%), whereas those rates in the recovered COVID group were 26.47%, 16.18%, and 19.12%, respectively. In HCs, these numbers were 7.69% for each condition (p < 0.01). Mild cognitive impairment (MCI) was more common in the long COVID group (24.42%; p = 0.02). There were no significant group differences in ESS, UPSIT, MMSE, or MoCA scores (p > 0.05). Detailed demographic and clinical characteristics are summarized in Table 1.

TABLE 1.

Demographic and neuropsychological characteristics of study subjects.

Long COVID

(n = 86)

Recovered COVID

(n = 67)

HC

(n = 26)

 F/H/χ2 p Post hoc analysis
Pa Pb Pc
Demographic information
Age (years) 61.5 (47.88, 75.13) 72.00 (67.00, 77.00) 72.00 (66.8, 77.3) 30.21 <0.001 * <0.001 * <0.001 * 0.22
Sex (M/F) 41/45 21/46 8/18 5.35 0.07 – – –
BMI 26.96 (23.33, 30.59) 25.62 (22.15, 29.10) 26.10 (22.58, 29.62) 1.33 0.54 – – –
Smoke status (n, %) 22 (25.58%) 13 (19.12%) 3 (11.54%) 2.63 0.27 – – –
Alcohol consumption (n, %) 54 (62.79%) 18 (26.47%) 2 (7.69%) 34.71 <0.001 * <0.001 * <0.001 * 0.005 *
Hypertension (n, %) 29 (33.72%) 11 (16.18%) 2 (7.69%) 10.69 0.005 * 0.003 * 0.001 * 0.42
Hypercholesterolemia (n, %) 30 (34.88%) 13 (19.12%) 2 (7.69%) 9.89 0.007 * 0.024 0.004 * 0.61
Diabetes (n, %) 13 (15.12%) 4 (5.88%) 1 (3.85%) 4.88 0.09 – – –
Neuropsychological batteries
ESS 6.00 (2.75, 9.25) 5.00 (1.50, 8.50) 5.60 ± 2.51 1.99 0.37 – – –
UPSIT 34.00 (29.25, 38.75) 32.00 (26.50, 37.50) 30.80 ± 4.92 0.48 0.79 – – –
MMSE 29.00 (27.50, 30.5) 29.00 (28.00, 30.00) 28.60 ± 1.14 0.83 0.66 – – –
MoCA 27.00 (25.5, 28.5) 28.00 (26.50, 29.50) 25.60 ± 2.88 5.12 0.08 – – –
CDR‐0 (normal) 65 (75.58%) 56 (83.58%) 26 (100%) 8.81 0.01 * 0.77 0.03 * 0.20
CDR‐0.5 (MCI) 21 (24.42%) 11 (16.42%) 0 8.11 0.02 * 0.10 0.04 * 0.20
CDR‐1 (mild dementia) 1 (1.16%) 0 0 – – – – –
CDR‐SOB 0 (0, 0.5) 0 (0, 0) 0 3.03 0.08 – – –

3.2. ChP alterations

As shown in Figure 1, both long COVID and recovered COVID patients exhibited significantly increased ChP volume (long COVID: median 2025.72 mm3, IQR 1618.55–2322.36; recovered COVID: median 1844 mm3, IQR 1584.77–2188) and reduced CBF (long COVID: 31.16 ± 10.75 mL/100 g/min; recovered COVID: 33.42 ± 11.37 mL/100 g/min) compared to HCl (mean volume: 1468.67 ± 363.28 mm3; CBF: 41.96 ± 12.38 mL/100 g/min; p < 0.05). Within the COVID groups, long COVID patients showed significantly larger ChP volume than recovered COVID patients (p = 0.02); however, CBF did not differ significantly between them (p = 0.22). Statistical results are shown in Figure 2 and Table 2. ChP volume was significantly negatively correlated with ChP CBF (r = –0.48, p < 0.001) after adjusting for age, sex, and TIV. Additionally, ChP CBF showed a positive correlation with gray matter (GM) CBF (r = 0.24, p = 0.04). ChP volume also showed a negative correlation with GM CBF (r = –0.39, p = 0.001) after adjusting for TIV. However, this association lost significance after adjusting for age.

FIGURE 2.

FIGURE 2

Group‐wise comparisons of ChP volume (A) and blood flow (B) between long COVID, recovered COVID, and HC groups. Boxes represent the median and interquartile range, with crosses indicating group means. *Denotes statistically significant difference between compared groups. Specifically, long COVID patients showed significantly larger ChP volume compared to recovered COVID and HCs, and significantly reduced CBF compared to HCs. CBF, cerebral blood flow; ChP, choroid plexus; COVID, coronavirus disease; HC, healthy control.

TABLE 2.

Comparisons of choroid plexus imaging characteristics among long COVID, recovered COVID, and HC subjects.

Characteristics Long COVID Recovered COVID HC F/H p Post hoc comparisons
Pa Pb Pc
Volume (mm3) 2025.72 (1618.55, 2322.36) 1844 (1584.77, 2188) 1468.67 ± 363.28 31.74 <0.001 * 0.02 * <0.001 * 0.004 *
CBF (mL/100 g/min) 31.16 ± 10.75 33.42 ± 11.37 41.96 ± 12.38 6.57 0.002 * 0.22 0.001 * 0.03 *

To evaluate whether ChP perfusion alterations were associated with global perfusion, we additionally compared GM CBF and the ChP‐to‐GM CBF ratio (ChP/GM CBF) across groups. GM CBF did not differ significantly among the three groups (p > 0.05). In contrast, the ChP/GM CBF ratio was significantly lower in long COVID compared to HCs (p = 0.02), but did not differ from the recovered COVID group (p > 0.05).

3.3. Correlations between ChP imaging measures and plasma biomarkers

In the long COVID group, we examined two classes of plasma biomarkers: inflammatory and AD related. As shown in Figure 3, ChP volume was positively correlated with IL‐6 (r = 0.28, p = 0.02) and TNF‐α (r = 0.33, p = 0.007), while ChP CBF was negatively correlated with TNF‐α (r = –0.39, p = 0.003). These associations lost significance after adjusting for age, sex, and TIV. Among AD‐related biomarkers, both ChP volume and CBF showed strong correlations with p‐tau217 (volume: r = 0.54, p < 0.001; CBF: r = –0.56, p = 0.001). ChP volume was also positively correlated with GFAP (r = 0.35, p = 0.005). These AD biomarker‐related associations remained significant after adjustment for age, sex, and TIV covariates. See Table 3 for complete partial correlation results.

FIGURE 3.

FIGURE 3

Significant correlation analyses between ChP imaging characteristics and peripheral plasma biomarkers. A, Correlations among ChP volume, ChP CBF, and inflammatory biomarkers; (B) correlations among ChP volume, ChP CBF, and Alzheimer's disease (AD)‐related biomarkers. r uncorr/p uncorr: Correlation coefficients and p values without variables correction; r corr/p corr: Partial correlation coefficients and p values adjusted for age, sex, and intracranial volume. *Denotes statistically significant correlation in each model. CBF, cerebral blood flow; ChP, choroid plexus; GFAP, glial fibrillary acidic protein; IL, interleukin; p‐tau217, phosphorylated tau217; TNF‐α, tumor necrosis factor alpha.

TABLE 3.

Correlation analyses of ChP imaging characteristics with plasma biomarkers in long COVID patients.

ChP volume ChP CBF
(r uncorr/P uncorr) (r corr/Pcorr ) (r uncorr/P uncorr) (r corr/Pcorr )
Inflammatory biomarker
IL‐6 (pg/mL) 0.28/0.02 * 0.26/0.048 −0.18/0.20 −0.10/0.50
IL‐10 (pg/mL) 0.27/0.03 0.26/0.044 −0.08/0.57 −0.02/0.88
TNF‐α (pg/mL) 0.33/0.007 * 0.20/0.11 −0.39/0.003 * −0.25/0.07
AD‐related biomarker
NfL (pg/mL) 0.20/0.12 −0.07/0.58 −0.11/0.41 0.17/0.23
GFAP (pg/mL) 0.46/0.001 * 0.35/0.005 * −0.41/0.002 * −0.26/0.06
p‐tau217 (pg/mL) 0.48/0.002 * 0.54/ < 0.001 * −0.49/0.004 * −0.56/0.001 *
Aβ42/Aβ40 −0.16/0.20 −0.05/0.73 0.21/0.13 −0.04/0.78

3.4. Correlations between ChP imaging measures and neurological symptoms

In the long COVID group, partial correlation analyses revealed that ChP volume was positively correlated with sleepiness (ESS; r = 0.37, p = 0.004) and negatively with CBF of ChP (r = –0.47, p < 0.001), after adjusting for age, sex, and TIV. Additionally, ChP volume showed a negative correlation with MMSE scores (r = –0.29, p = 0.02), and ChP CBF was negatively correlated with CDR Sum of Boxes scores (r = –0.33, p = 0.01; Figure 4). Detailed statistics are presented in Table 4.

FIGURE 4.

FIGURE 4

Significant partial correlation analyses between ChP imaging characteristics and neurological assessments. A, Associations of ChP volume with ESS (left) and ChP CBF with ESS (right). B, Associations of ChP volume with MMSE (left) and ChP, CBF with CDR‐SOB (right). r corr/p corr: Partial correlation coefficients and p values adjusted for age, sex, and intracranial volume. *Denotes statistically significant correlation in each model. CBF, cerebral blood flow; CDR‐SOB, Clinical Dementia Rating Scale Sum of Boxes; ChP, choroid plexus; ESS, Epworth Sleepiness Scale; MMSE, Mini‐Mental State Examination.

TABLE 4.

Correlations between ChP imaging characteristics and neuropsychological batteries in long COVID patients.

ChP volume ChP CBF
Neuropsychological batteries (r uncorr/P uncorr) (r corr/Pcorr ) (r uncorr/P uncorr) (r corr/Pcorr )
ESS 0.34/0.007 * 0.37/0.004 * −0.41/0.002 * −0.47/ < 0.001 *
UPSIT −0.29/0.03 −0.17/0.23 0.33/0.02 * 0.14/0.35
MMSE −0.41/ < 0.001 * −0.29/0.02 * 0.32/0.01 * 0.15/0.28
MoCA −0.23/0.06 −0.11/0.40 0.33/0.01 * 0.18/0.17
CDR‐SOB 0.19/0.11 0.19/0.11 −0.43/0.001 * −0.33/0.01 *

4. DISCUSSION

The present study investigated structural and perfusion alterations in the ChP and their associations with neurological symptoms and plasma biomarkers in individuals with long COVID. Overall, these patients exhibited enlarged ChP and reduced blood perfusion. These alterations were associated with sleep disorders and cognitive symptoms and correlated with AD‐related plasma biomarkers and neurodegeneration. Our findings extend prior research results that focus mainly on ChP volume and introduce novel evidence of blood perfusion alterations as well as their correlations with cognitive symptoms and plasma biomarkers. These results suggest that ChP dysfunction in long COVID patients may underlie deficits in CSF synthesis and BCSFB integrity, possibly playing a role in the development of AD‐like pathology and cognitive symptoms.

Our findings are consistent with existing evidence that implicates the ChP abnormalities in the systemic COVID‐19 infection. 21 , 22 , 28 Studies in brain organoids have demonstrated a clear tropism of SARS‐CoV‐2 for ChP epithelium, even with minimal or no infection of glial cells or neurons. 22 ChP is a highly vascularized structure that forms the BCSFB, composed of fenestrated capillaries and a tight epithelial layer. Beyond CSF production, ChP is an active site for immune surveillance, and the ChP epithelial cells produce cytokines, chemokines, and adhesion molecules that regulate recruitment and activation of immune cells. Neurological complications of COVID‐19 may result from aberrant inflammation initiated in the peripheral circulatory system, potentially entering the CNS and mediated by the ChP. Analysis of CSF from COVID patients further supports this hypothesis of a “cytokine storm” syndrome, exhibiting an overactive immune response characterized by excessive inflammation with rare viral presence in the CSF, but frequent abnormalities such as elevated protein levels, increased inflammatory factors, and pleocytosis. 24 , 38 Therefore, the ChP serves as a mediator between the CNS and peripheral immune system, and may act as a portal for virus‐induced pathogen entry into the brain, as evidenced by prior animal studies. 39

Our results demonstrated that ChP enlargement was positively correlated with plasma IL‐6, TNF‐α, and GFAP. However, associations between ChP volume and inflammatory factors (IL‐6, TNF‐α), but not GFAP, were attenuated after adjusting for age, indicating the strong influence of age on inflammation and age‐related ChP degeneration. This may also indicate that inflammatory activity is diminished in these patients during the chronic, rather than the acute, phase of infection. GFAP, an astrocytic biomarker, elevated during neuronal injury, glial activation, and scarring, 40 is recently considered a potential biomarker of AD. 41 Increasing evidence suggests that blood GFAP levels can be used to detect early‐stage AD. 42 Elevated GFAP levels have also been reported in both acute SARS‐CoV‐2 infection and in long COVID, 43 indicating immune‐mediated astroglia injury and hypoxia‐driven inflammation may contribute to post‐infection neurological sequelae.

However, ChP enlargement has long been reported in normal aging 30 , 36 , 44 in addition to several neurodegenerative and neuroinflammatory disorders, including AD, 31 multiple sclerosis, 45 and amyotrophic lateral sclerosis. 46 Post mortem analyses of human ChP provide insights into these in vivo observations, revealing increased stromal fibrosis with thickening of vessel walls and capillary membranes, 47 stromal calcification, 48 and tortuous or occluded ChP arteries. 49 The volumetric findings in our study may therefore reflect these underlying histopathological changes that are often seen in neurodegenerative and neuroinflammatory conditions. Furthermore, one prior study suggests that chronic inflammation may upregulate ChP‐derived growth factors, resulting in epithelial cell proliferation and further contribute to ChP volume increases. 35 The present study identified that ChP enlargement is positively correlated with increased sleepiness and negatively correlated with lower MMSE scores, which are both pertinent indicators of neurodegenerative or AD‐related clinical alterations. Previous studies also reported ChP enlargement in patients with excessive daytime sleepiness and cognitive impairment. 21 , 50 Sleep disorders are common in long COVID, with reported rates ranging from 6% to > 70%, and encompass insomnia, sleep‐disordered breathing, or hypersomnolence. 51 Larger ChP volume has also been associated with lower CSF secretion rates, resulting in diminished CSF turnover and impaired glymphatic function, 52 which is considered a potential pathogenic mechanism of AD. Cognitive symptoms are also among the most common neurological manifestations of long COVID. 53

In addition to structural alterations, we observed decreased blood flow in the ChP. The ChP is supplied by the anterior and posterior choroidal arteries, regulated by molecules such as arginine vasopressin, angiotensin II, dopamine, and serotonin. SARS‐CoV‐2 can damage capillary endothelium via the angiotensin‐converting enzyme‐2 (ACE2), leading to endotheliitis, vascular constriction, and vessel wall thickening. 54 Dysregulation of neurotransmitter and hormonal signaling may further contribute to vascular alterations in long COVID. In addition, emerging evidence suggests that COVID‐19 may accelerate vascular degeneration, characterized by increased arterial stiffness and endothelial dysfunction. 55 Given the highly vascularized nature of the ChP, such systemic vascular alterations could contribute to the ChP hypoperfusion observed in long COVID. However, reduced ChP CBF may also potentially reflect a more global cerebral hypoperfusion. Our previous work demonstrated age‐related declines in both GM and ChP perfusion, with ChP perfusion showing a steeper rate of decline. 30 To distinguish regional from global perfusion effects, we compared GM CBF and calculated the ChP‐to‐GM (ChP/GM) CBF ratio. While GM CBF did not differ significantly among groups, both ChP CBF and the ChP/GM CBF ratio were significantly reduced in the long COVID group. These findings indicate a regionally selective vulnerability of the ChP in long COVID, which is consistent with prior evidence that the BCSFB of the ChP is susceptible to SARS‐CoV‐2 infection 56 and that the ChP is particularly responsive to inflammatory and related stimuli in the ChP microenvironment. 57 Functionally, reduced ChP perfusion has been linked to diminished CSF secretion and impaired fluid circulation. 33 , 58 Thus, the combination of increased volume and decreased perfusion may restrict blood–epithelium exchange, disrupting CSF regulation. Because CSF is essential for immune response and CSF production as well as metabolic waste clearance, such dysfunction may underlie the neurological symptoms observed in long COVID. In our study, reduced ChP blood flow was correlated with cognitive impairment and daytime sleepiness. Daytime sleepiness, recognized as a risk factor for AD, has been linked to increased arterial stiffness and poorer cognitive performance, which may in turn contribute to reduced ChP perfusion and impaired glymphatic clearance, ultimately promoting AD occurrence in long COVID. 59 , 60 , 61

We also found that both ChP volume and CBF were significantly correlated with blood p‐tau217 levels. Plasma p‐tTau217 is a highly sensitive biomarker for identifying AD, reflecting both amyloid and tau burden. 62 SARS‐CoV‐2 infection has also been associated with Aβ pathology and is thought to increase the risk of future AD. 63 Accumulating evidence suggests a potential association between COVID‐19 infection and AD pathogenesis. For example, tau aggregation may be promoted through activation of the NLR family pyrin domain containing 3 (NLRP3), which is triggered during SARS‐CoV‐2 infection. 64 In addition, the apolipoprotein E ε4 allele, a well‐established genetic risk factor for AD, has been proposed as a biomarker for COVID‐19. 65 Because ChP also plays an essential role in neuroprotection by synthesizing and secreting proteins such as transthyretin and the receptor megalin, which facilitates Aβ clearance, 66 , 67 the observed ChP enlargement and reduced perfusion, together with elevated p‐tau217, may reflect a greater burden of AD‐related pathology in long COVID. However, further studies are warranted to confirm these associations and to clarify their implications for AD pathogenesis.

Several limitations should be acknowledged. First, ChP segmentation may not fully capture its complex structure due to partial‐volume effect. To improve accuracy, we applied an automated GMM approach followed by manual inspection and adjustment on 3D FLAIR images, which revealed better ChP contrast than T1‐weighted imaging. Second, the overall sample size was limited, particularly for the HC group, resulting in an imbalanced age distribution across groups. Although age was included as a covariate in the analysis, validation in larger cohorts with age‐matched and age‐stratified analysis is needed to better account for age effects. Third, the standard single‐delay pCASL sequence used in this study may not accurately quantify ChP perfusion, underscoring the need for high‐resolution multiple PLD‐based pCASL with improved CBF measurement. Fourth, incomplete neurological and biomarker profiles may have biased correlation analyses, which highlights the importance of comprehensive data collection in future studies. Finally, the cross‐sectional design limits causal inference. It remains unclear whether ChP alterations represent a cause of neurodegeneration or a secondary response to parenchymal changes. Longitudinal follow‐up is necessary to clarify the temporal sequence and the progression of ChP alterations across the disease course.

In conclusion, we identified ChP alterations in long COVID, characterized by enlarged volume and reduced blood flow. These alterations were associated with sleep disturbances, cognitive impairment, and plasma biomarkers of AD, suggesting a potential link to future dementia risk. Extending prior volumetric studies, our findings highlight perfusion abnormalities, closely associated with ChP enlargement, as an additional feature of ChP pathology in long COVID. The ChP abnormalities thus represent a promising marker for predicting AD risk.

AUTHOR CONTRIBUTIONS

Conceptualization: Yulin Ge and Huize Pang. Methodology: Huize Pang and Zhe Sun. Software: Huize Pang, Chenyang Li, Zhe Sun, and Li Jiang. Investigation: Huize Pang, Jennifer Frontera, Chenyang Li, Alok Vedvyas, Arjun V. Masurkar, Allal Boutajangout, Ludovic Debure, Mobeena Ghuman, Li Jiang, and Thomas Wisniewski. Resources: Yulin Ge, Jennifer Frontera, and Thomas Wisniewski. Original draft: Huize Pang. Writing—review and editing: Yulin Ge, Jennifer Frontera, Thomas Wisniewski. Supervision: Yulin Ge, Jennifer Frontera, and Thomas Wisniewski.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflicts of interest. Author disclosures are available in the supporting information.

Supporting information

Supporting information

ACKNOWLEDGMENTS

This research was supported by grants from the National Institutes of Health/National Institute on Aging R01AG077422, U24 NS135568, and P30AG066512. NIH R01AG077422, U24 NS135568, and P30AG066512.

Pang H, Frontera J, Jiang L, et al. Choroid plexus alterations in long COVID and their associations with Alzheimer's disease risks. Alzheimer's Dement. 2026;22:e71020. 10.1002/alz.71020

DATA AVAILABILITY STATEMENT

All human subjects provided informed consent. The data used and analyzed in this study are available from the corresponding author upon reasonable request.

REFERENCES

  • 1. Soriano JB, Murthy S, Marshall JC, Relan P, Diaz JV. A clinical case definition of post‐COVID‐19 condition by a Delphi consensus. Lancet Infect Dis. 2022;22(4):e102–e107. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Premraj L, Kannapadi NV, Briggs J, et al. Mid and long‐term neurological and neuropsychiatric manifestations of post‐COVID‐19 syndrome: a meta‐analysis. J Neurol Sci. 2022;434:120162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Pinzon RT, Wijaya VO, Jody AA, Nunsio PN, Buana RB. Persistent neurological manifestations in long COVID‐19 syndrome: a systematic review and meta‐analysis. J Infect Public Health. 2022;15(8):856–869. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Taquet M, Sillett R, Zhu L, et al. Neurological and psychiatric risk trajectories after SARS‐CoV‐2 infection: an analysis of 2‐year retrospective cohort studies including 1 284 437 patients. Lancet Psychiatry. 2022;9(10):815–827. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Atkins JL, Masoli JAH, Delgado J, et al. Preexisting comorbidities predicting COVID‐19 and mortality in the UK Biobank Community Cohort. J Gerontol A Biol Sci Med Sci. 2020;75(11):2224–2230. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Liu N, Sun J, Wang X, Zhao M, Huang Q, Li H. The impact of dementia on the clinical outcome of COVID‐19: a systematic review and meta‐analysis. J Alzheimers Dis. 2020;78(4):1775–1782. [DOI] [PubMed] [Google Scholar]
  • 7. Guedj E, Campion JY, Dudouet P, et al. 18F‐FDG brain PET hypometabolism in patients with long COVID. Eur J Nucl Med Mol Imaging. 2021;48(9):2823–2833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Turner S, Khan MA, Putrino D, Woodcock A, Kell DB, Pretorius E. Long COVID: pathophysiological factors and abnormalities of coagulation. Trends Endocrinol Metab. 2023;34(6):321–344. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Greene C, Connolly R, Brennan D, et al. Blood–brain barrier disruption and sustained systemic inflammation in individuals with long COVID‐associated cognitive impairment. Nat Neurosci. 2024;27(3):421‐432. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Gupta G, Buonsenso D, Wood J, Mohandas S, Warburton D. Mechanistic insights into long covid: viral persistence, immune dysregulation, and multi‐organ dysfunction. Compr Physiol. 2025;15(3):e70019. [DOI] [PubMed] [Google Scholar]
  • 11. Spudich S, Nath A. Nervous system consequences of COVID‐19. Science. 2022;375(6578):267–269. [DOI] [PubMed] [Google Scholar]
  • 12. Zuo W, He D, Liang C, et al. The persistence of SARS‐CoV‐2 in tissues and its association with long COVID symptoms: a cross‐sectional cohort study in China. Lancet Infect Dis. 2024;24(8):845–855. [DOI] [PubMed] [Google Scholar]
  • 13. Charnley M, Islam S, Bindra GK, et al. Neurotoxic amyloidogenic peptides in the proteome of SARS‐COV2: potential implications for neurological symptoms in COVID‐19. Nat Commun. 2022;13(1):3387. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Reas ET, Alderson‐Myers A, Solders SK, et al. Long COVID‐related blood‐brain barrier breakdown and microstructure in older adults are modified by sex and Alzheimer's disease genetic risk. Imaging Neurosci (Camb). 2025;3:IMAGa23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Priemer DS, Rhodes CH, Karlovich E, Perl DP, Goldman JE. Aβ deposits in the neocortex of adult and infant hypoxic brains, including in cases of COVID‐19. J Neuropathol Exp Neurol. 2022;81(12):988–995. [DOI] [PubMed] [Google Scholar]
  • 16. Wang L, Davis PB, Volkow ND, Berger NA, Kaelber DC, Xu R. Association of COVID‐19 with new‐onset Alzheimer's disease. J Alzheimers Dis. 2022;89(2):411–414. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17. Frontera JA, Boutajangout A, Masurkar AV, et al. Comparison of serum neurodegenerative biomarkers among hospitalized COVID‐19 patients versus non‐COVID subjects with normal cognition, mild cognitive impairment, or Alzheimer's dementia. Alzheimer's Dement. 2022;18(5):899–910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Engelhardt B, Sorokin L. The blood‐brain and the blood‐cerebrospinal fluid barriers: function and dysfunction. Semin Immunopathol. 2009;31(4):497–511. [DOI] [PubMed] [Google Scholar]
  • 19. Xu H, Lotfy P, Gelb S, et al. The choroid plexus synergizes with immune cells during neuroinflammation. Cell. 2024;187(18):4946–4963. e17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Burks SM, Rosas‐Hernandez H, Alejandro Ramirez‐Lee M, Cuevas E, Talpos JC. Can SARS‐CoV‐2 infect the central nervous system via the olfactory bulb or the blood‐brain barrier? Brain Behav Immun. 2021;95:7–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Diez‐Cirarda M, Yus‐Fuertes M, Delgado‐Alonso C, et al. Choroid plexus volume is enlarged in long COVID and associated with cognitive and brain changes. Mol Psychiatry. 2025;30(7):2821–2830. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Pellegrini L, Albecka A, Mallery DL, et al. SARS‐CoV‐2 infects the brain choroid plexus and disrupts the blood‐CSF barrier in human brain organoids. Cell Stem Cell. 2020;27(6):951–961. e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Apple AC, Oddi A, Peluso MJ, et al. Risk factors and abnormal cerebrospinal fluid associate with cognitive symptoms after mild COVID‐19. Ann Clin Transl Neurol. 2022;9(2):221–226. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Tandon M, Kataria S, Patel J, et al. A comprehensive systematic review of CSF analysis that defines neurological manifestations of COVID‐19. Int J Infect Dis. 2021;104:390–397. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Pilotto A, Masciocchi S, Volonghi I, et al. Severe acute respiratory syndrome coronavirus 2 (SARS‐CoV‐2) encephalitis is a cytokine release syndrome: evidences from cerebrospinal fluid analyses. Clin Infect Dis. 2021;73(9):e3019–e3026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Virhammar J, Nääs A, Fällmar D, et al. Biomarkers for central nervous system injury in cerebrospinal fluid are elevated in COVID‐19 and associated with neurological symptoms and disease severity. Eur J Neurol. 2021;28(10):3324–3331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Sriwastava S, Tandon M, Podury S, et al. COVID‐19 and neuroinflammation: a literature review of relevant neuroimaging and CSF markers in central nervous system inflammatory disorders from SARS‐COV2. J Neurol. 2021;268(12):4448–4478. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Rau A, Gonzalez‐Escamilla G, Schroeter N, et al. Inflammation‐triggered enlargement of choroid plexus in subacute COVID‐19 patients with neurological symptoms. Ann Neurol. 2024;96(4):715–725. [DOI] [PubMed] [Google Scholar]
  • 29. Althubaity N, Schubert J, Martins D, et al. Choroid plexus enlargement is associated with neuroinflammation and reduction of blood brain barrier permeability in depression. NeuroImage Clin. 2022;33:102926. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Sun Z, Li C, Zhang J, Wisniewski T, Ge Y. Choroid plexus aging: structural and vascular insights from the HCP‐aging dataset. Fluids Barriers CNS. 2024;21(1):98. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Xia H, Feng Y, Zhu H, et al. Association of choroid plexus volume with white matter microstructure, glymphatic function, and peripheral systemic inflammation in Alzheimer's disease. Transl Psychiatry. 2025;15(1):238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Leitner DF, Kanshin E, Faustin A, et al. Localized proteomic differences in the choroid plexus of Alzheimer's disease and epilepsy patients. Front Neurol. 2023;14:1221775. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Eisma JJ, McKnight CD, Hett K, et al. Choroid plexus perfusion and bulk cerebrospinal fluid flow across the adult lifespan. J Cereb Blood Flow Metab. 2023;43(2):269–280. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Li Y, Zhou Y, Zhong W, et al. Choroid plexus enlargement exacerbates white matter hyperintensity growth through glymphatic impairment. Ann Neurol. 2023;94(1):182–195. [DOI] [PubMed] [Google Scholar]
  • 35. Alisch JSR, Kiely M, Triebswetter C, 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;13:734992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Li C, Sun Z, Zhang J, Ge Y. Relaxation‐selective intravoxel incoherent motion imaging of microvascular perfusion and fluid compartments in the human choroid plexus. NMR Biomed. 2025;38(10):e70144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Li Y, Liu P, Li Y, et al. ASL‐MRICloud: an online tool for the processing of ASL MRI data. NMR Biomed. 2019;32(2):e4051. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Destras G, Bal A, Escuret V, Morfin F, Lina B, Josset L. Systematic SARS‐CoV‐2 screening in cerebrospinal fluid during the COVID‐19 pandemic. Lancet Microbe. 2020;1(4):e149. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Marques F, Sousa JC, Correia‐Neves M, Oliveira P, Sousa N, Palha JA. The choroid plexus response to peripheral inflammatory stimulus. Neuroscience. 2007;144(2):424–430. [DOI] [PubMed] [Google Scholar]
  • 40. Kanberg N, Ashton NJ, Andersson LM, et al. Neurochemical evidence of astrocytic and neuronal injury commonly found in COVID‐19. Neurology. 2020;95(12):e1754–e1759. [DOI] [PubMed] [Google Scholar]
  • 41. Kim KY, Shin KY, Chang KA. GFAP as a potential biomarker for Alzheimer's disease: a systematic review and meta‐analysis. Cells. 2023;12(9):1309. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Bouhrara M, Walker KA, Alisch JSR, et al. Association of plasma markers of Alzheimer's disease, neurodegeneration, and neuroinflammation with the choroid plexus integrity in aging. Aging Dis. 2024;15(5):2230–2240. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Spanos M, Shachar S, Sweeney T, et al. Elevation of neural injury markers in patients with neurologic sequelae after hospitalization for SARS‐CoV‐2 infection. iScience. 2022;25(8):104833. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Sun Z, Li C, Leitner D, et al. Age‐related vascular alterations in the choroid plexus: novel insights from pathophysiology and imaging studies. Aging Dis. 2025;e0735. 10.14336/AD.2025.0735 [DOI] [PubMed] [Google Scholar]
  • 45. Müller J, Sinnecker T, Wendebourg MJ, et al. Choroid plexus volume in multiple sclerosis vs neuromyelitis Optica spectrum disorder. Neurol Neuroimmunol Neuroinflamm. 2022;9(3):e1147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Dai T, Lou J, Kong D, et al. Choroid plexus enlargement in amyotrophic lateral sclerosis patients and its correlation with clinical disability and blood‐CSF barrier permeability. Fluids Barriers CNS. 2024;21(1):36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Serot JM, Béné MC, Foliguet B, Faure GC. Morphological alterations of the choroid plexus in late‐onset Alzheimer's disease. Acta Neuropathol. 2000;99(2):105–108. [DOI] [PubMed] [Google Scholar]
  • 48. Alcolado JC, Moore IE, Weller RO. Calcification in the human choroid plexus, meningiomas and pineal gland. Neuropathol Appl Neurobiol. 1986;12(3):235–250. [DOI] [PubMed] [Google Scholar]
  • 49. Shuangshoti S, Netsky MG. Human choroid plexus: morphologic and histochemical alterations with age. Am J Anat. 1970;128(1):73–95. [DOI] [PubMed] [Google Scholar]
  • 50. Hu P, Yuan Y, Zou Y, et al. Alterations in the DTI‐ALPS index and choroid plexus volume are associated with clinical symptoms in participants with narcolepsy type 1. Sleep Med. 2024;124:471–478. [DOI] [PubMed] [Google Scholar]
  • 51. Tedjasukmana R, Budikayanti A, Islamiyah WR, Witjaksono A, Hakim M. Sleep disturbance in post COVID‐19 conditions: prevalence and quality of life. Front Neurol. 2022;13:1095606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52. Damkier HH, Brown PD, Praetorius J. Cerebrospinal fluid secretion by the choroid plexus. Physiol Rev. 2013;93(4):1847–1892. [DOI] [PubMed] [Google Scholar]
  • 53. Thaweethai T, Jolley SE, Karlson EW, et al. Development of a definition of post‐acute sequelae of SARS‐CoV‐2 infection. JAMA. 2023;329(22):1934–1946. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Mehboob R, von Kries JP, Ehsan K, Almansouri M, Bamaga AK. Role of endothelial cells and angiotensin converting enzyme‐II in COVID‐19 and brain damages post‐infection. Front Neurol. 2023;14:1210194. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Walters H. COVID‐19 accelerates vascular aging. Nature Aging. 2025;5(9):1653. [DOI] [PubMed] [Google Scholar]
  • 56. Stüdle C, Nishihara H, Wischnewski S, et al. SARS‐CoV‐2 infects epithelial cells of the blood‐cerebrospinal fluid barrier rather than endothelial cells or pericytes of the blood‐brain barrier. Fluids Barriers CNS. 2023;20(1):76. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57. Kaur C, Rathnasamy G, Ling EA. The choroid plexus in healthy and diseased brain. J Neuropathol Exp Neurol. 2016;75(3):198–213. [DOI] [PubMed] [Google Scholar]
  • 58. Schalk KA, Faraci FM, Heistad DD. Effect of endothelin on production of cerebrospinal fluid in rabbits. Stroke. 1992;23(4):560–563. [DOI] [PubMed] [Google Scholar]
  • 59. Tang YL, Chen HB, Liu P, Liao YH, Xie A. Glymphatic function alterations in sleep disorder patients post‐COVID‐19: a Longitudinal DTI‐ALPS Study. Nat Sci Sleep. 2025;17:1377–1390. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60. Kim M, Lee KH, Ko JS, et al. Neurologic symptoms as a hallmark of glymphatic alteration in recovered patients with COVID‐19. BMC Neurol. 2025;25(1):187. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61. Butts B, Hu WT, Verble DD, et al. Relationships between excessive daytime sleepiness and arterial stiffness with cognitive function in a racially diverse high‐risk cohort. Alzheimer's Dement. 2023;19(S18):e077150. [Google Scholar]
  • 62. Therriault J, Brum WS, Trudel L, et al. Blood phosphorylated tau for the diagnosis of Alzheimer's disease: a systematic review and meta‐analysis. Lancet Neurol. 2025;24(9):740–752. [DOI] [PubMed] [Google Scholar]
  • 63. Duff EP, Zetterberg H, Heslegrave A, et al. Plasma proteomic evidence for increased β‐amyloid pathology after SARS‐CoV‐2 infection. Nat Med. 2025;31(3):797–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64. Reiken S, Sittenfeld L, Dridi H, Liu Y, Liu X, Marks AR. Alzheimer's‐like signaling in brains of COVID‐19 patients. Alzheimer's Dement. 2022;18(5):955–965. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65. Kuo CL, Pilling LC, Atkins JL, et al. APOE ε4 genotype predicts severe COVID‐19 in the UK Biobank Community Cohort. J Gerontol A Biol Sci Med Sci. 2020;75(11):2231–2232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66. Carro E, Spuch C, Trejo JL, Antequera D, Torres‐Aleman I. Choroid plexus megalin is involved in neuroprotection by serum insulin‐like growth factor I. J Neurosci. 2005;25(47):10884–10893. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67. Sousa JC, Cardoso I, Marques F, Saraiva MJ, Palha JA. Transthyretin and Alzheimer's disease: where in the brain? Neurobiol Aging. 2007;28(5):713–718. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supporting information

Data Availability Statement

All human subjects provided informed consent. The data used and analyzed in this study are available from the corresponding author upon reasonable request.