MRI Markers of Perivascular Fluid Dynamics in Type 2 Diabetes and Their Associations with Glycemic Control

MRI Markers of Perivascular Fluid Dynamics in Type 2 Diabetes and Their Associations with Glycemic Control | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article MRI Markers of Perivascular Fluid Dynamics in Type 2 Diabetes and Their Associations with Glycemic Control Ying Cui, Yu Wang, Ying Luan, Tianyu Tang, Lijuan Zheng, Chunqiang Lu, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7859506/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Backgrounds: Perivascular fluid dynamics support brain-wide solute transport and have been described to be perturbed in type 2 diabetes mellitus (T2DM) models. In this study, we aimed to evaluate T2DM-related alterations in MRI perivascular markers and their associations with glycemic control and cognitive performances. Methods: Thirty-four T2DM patients and 34 age- and sex-matched controls underwent multimodal MRI to quantify perivascular metrics: the diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) index reflects perivenous diffusion anisotropy, white matter free water (WM-FW) indicates CSF-interstitial fluid exchange, and perivascular space volume fraction (PVSVF) indexes perivascular conduits burden. Cognitive function and metabolic indicators were evaluated. Group comparisons and associations were tested with multivariable linear regression adjusting for key demographic and intracranial covariates. Results: Compared with controls, T2DM showed a lower ALPS index (adjusted β = -0.091, 95% CI -0.155 to -0.027, p = 0.006; FDR-adjusted p = 0.025). The group differences were more prominent among male and older patients, although formal interactions were not significant. WM-FW and PVSVF showed no group difference after adjusting covariates. Within T2DM, lower ALPS was associated with higher postprandial glucose (partial r = -0.46) and HbA1c (partial r = -0.37) (all p ≤0.05, FDR-adjusted p = 0.11). Multivariable linear regression including all MRI markers showed that ALPS was the only imaging metric that remained independently associated with glycemic measures. Conclusions: T2DM is associated with altered perivascular diffusion anisotropy on MRI, indicating perivascular dysfunction. Associations with glycemic profile suggest these modifiable factors may influence perivascular health in T2DM and merit prospective evaluation. Type 2 diabetes insulin resistance cognitive impairment MRI Neuroimaging glymphatic system Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Background Type 2 diabetes mellitus (T2DM) is characterized by insulin resistance (IR) and abnormal glucose metabolism. In addition to its related metabolic consequences, T2MD is received increasing attention for its adverse impact on brain health [1]. Epidemiological evidence shows that cognitive decrements are more prevalent and progresses more rapidly among T2DM patients [2]. T2DM nearly doubles the risk of developing dementia compared to age-matched non-diabetic subjects, especially in elderly subjects [3]. Due to the increasing burden of T2DM worldwide, potential causative pathways in T2DM-associated cognitive dysfunction need to be identified in order to develop course-modifying therapies. The brain perivascular network is recently identified that is responsible for clearing metabolic waste products through cerebrospinal fluid (CSF) and interstitial fluid (ISF) exchange, a process facilitated by aquaporin-4 (AQP4) on astrocytic endfeet [4]. Disruption of this fluid dynamics has been increasingly implicated in neurodegenerative disease, particularly Alzheimer’s disease (AD), where impaired clearance of amyloid-β and tau is observed [5]. Importantly, the perivascular pathway is emerging as a therapeutic target, as pharmacologic and lifestyle interventions can reduce waste accumulation and improve cognition in preclinical models [6, 7]. Although the exact mechanism of T2DM-related cognitive decline is still under debate, perivascular dysfunction has been proposed as an important contributor. In diabetic animal models, CSF tracer clearance is slowed, accompanied by reduced AQP4 expression/polarization and greater amyloid burden that correlate with cognitive deficits [8, 9]. Despite these evidences, robust in-vivo human evidence of disturbed perivascular fluid dynamics in T2DM remains limited, underscoring the need for further investigation. Promising contrast‐free MRI approaches have been proposed to indirectly probe perivascular function [10, 11]. The diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) index is among the most used, which computes a ratio of diffusivities sampled in periventricular white matter (WM) near deep medullary veins at the lateral ventricle body [12]. Notably, ALPS has been recently challenged due to its sensitivity to ROI placement, local fiber geometry and acquisition parameters [13], and should therefore be regarded as an indirect proxy instead of a measure of waste clearance. White matter free water (WM-FW), quantifiable via diffusion MRI, suggests extracellular fluid accumulation and reflects water-content changes that may arise from attenuated CSF-ISF exchange [14]. The visible perivascular space (PVS) burden is often interpreted as a drainage system, the enlargement of which has been hypothesized to be secondary to impaired interstitial drainage [15]. While none of these metrics alone constitutes a definitive readout of glymphatic activity, their combined measurements may provide a more comprehensive signature of perivascular fluid dynamics and potential waste-clearance deficits [11]. In T2DM, however, systematic evaluations of ALPS, WM-FW, and PVS, and their relationships to metabolic dysfunction and cognitive decline remain limited. This study evaluates MRI-derived metrics of perivascular fluid dynamics in T2DM, and tests their associations with metabolic dysregulation and cognitive performance. We hypothesized that T2DM would show alterations in these perivascular markers and that worse metabolic profiles would relate to more adverse imaging signatures and poorer cognition. Clarifying these links may enhance the understanding of diabetes-related brain disturbances and inform earlier risk stratification. Methods Participant Recruitment The inclusion criteria for all participants included: (1) aged between 50 to 75 years old; (2) right handedness; (3) a minimum education of 6 years; (4) no contraindications of MRI scanning, screened using MRI safety questionnaire; (5) previous routine CT/MRI examination of the head showed no severe brain abnormality or only mild leukoaraiosis, which will be reassured after the following MR scan. T2DM patients were recruited from the department of Endocrinology or the local community, and met the the clinical diagnostic criteria for T2DM recommended by the American Diabetes Association (ADA) [16]. Each patient had a documented disease duration of at least one year and was under regular self-monitoring of blood glucose. A group of healthy control subjects (HCs) was recruited through community advertisements, who were matched to the T2DM pateints with respect to age, sex, and educational level. Clinical Assessments and Biochemical Measurements On the study day, all participants arrived at the hospital at 7:00 a.m. following an overnight fast of at least 8 hours. Fasting venous blood samples were collected for the assessment of key metabolic indicators, including fasting plasma glucose (FPG), glycated hemoglobin (HbA1c), fasting insulin, and a comprehensive lipid profile (triglycerides, total cholesterol, low-density lipoprotein [LDL] cholesterol, and high-density lipoprotein [HDL] cholesterol). To assess glucose metabolism in the HC group, a standard oral glucose tolerance test (OGTT) was administered, consisting of 75 grams of dextrose monohydrate dissolved in 250 mL of water. At 9:00 a.m., two hours after OGTT administration in the HC group and two hours after a typical breakfast in the T2DM group, additional blood samples were collected to evaluate the 2-hour postprandial glucose (2-h PG). Insulin resistance was measured by Homeostatic Model Assessment for Insulin Resistance (HOMA-IR), calculated as (fasting insulin × fasting glucose)/22 [17]. Blood pressure was measured three times throughout the visit at standardized intervals, and the average of these readings was recorded. During the visit, a structured questionnaire was also administered to all participants to collect detailed information on medical history, current medication use, systemic illnesses, and family history. In addition, weight, height, waist and hip circumference were measured and recorded. After the clinical assessments and biochemical measurements, participants were excluded if they met any of the following criteria: (1) a documented history of neurological or psychiatric disorders unrelated to T2DM; (2) prior brain lesions (such as tumor, stroke or traumatic injury) or intracranial surgeries; (3) serious medical emergencies (such as heart, lung, liver and kidney failure); (4) current/past history of major physical diseases (such as thyroid/parathyroid dysfunction, malignant tumors, autoimmune diseases and other endocrine disorders); (5) HCs who met the criteria for prediabetes or T2DM; (6) subjects with severe WMH, defined as a Fazekas score >2 (i.e., grade 3) in either periventricular WMH (PVWMH) or deep WMH (DWMH). Neuropsychological tests Each participant underwent a comprehensive battery of neuropsychological tests to assess general cognitive status and multiple specific domains. Global cognitive functioning was evaluated using the Mini-Mental State Examination (MMSE). Visuospatial processing was assessed using the Clock-Drawing Test (CDT) and the copy trial of the Complex Figure Test (CFT-copy). Episodic memory was assessed using the Auditory Verbal Learning Test (AVLT), working memory with the Digit Span Test (DST), and spatial memory with the 20min delayed recall trial of the Complex Figure Test (CFT-delay). Processing speed and attention were evaluated using Part A of the Trail-Making Test (TMT-A), and executive function with Part B of the Trail-Making Test (TMT-B). All tests were administered in a fixed sequence, and the entire battery required approximately 60 minutes to complete. Participants scoring below 24 on the MMSE were considered to meet the criteria for dementia and were excluded from the study. MRI data acquisition Subjects were scanned using a 3 Tesla MR instrument (MAGENETOM Trio, Siemens Medical Solutions, Erlangen, Germany) with a birdcage head coil. Foam padding was used to minimize head motion and earplugs were used to reduce the scanner noise. All subjects were instructed to rest with their eyes closed, not think of anything in particular, not fall asleep and avoid any head motion during fMRI scanning. High-resolution three-dimensional T1-weighted structural images were acquired using a magnetization-prepared rapid acquisition gradient-echo (MPRAGE) sequence (repetition time [TR]/echo time [TE] = 1900ms/2.48 ms, flip angle = 9°, acquisition matrix size = 256 × 256, field of view [FOV] = 250 mm × 250 mm, slice thickness = 1 mm, 176 slices in the sagittal orientation). Diffusion tensor images (DTI) were obtained using a spin echo-based echo planar imaging sequence (TR/TE = 10000 ms/95 ms, FOV = 256 mm × 256 mm, matrix size = 128 × 128, section thickness = 2 mm). Images were obtained with both 30-direction encoding (b-value = 1000 s/mm 2 for each direction) and no diffusion encoding (b = 0 s/mm 2 ), consisting of 70 contiguous axial slices covering the whole brain. T2-weighred images based on spin echo sequence were obtained: TR/TE = 6000 ms/95 ms, field of view = 212mm × 229 mm, slice thickness = 5 mm. Finally, T2 fluid-attenuated inversion recovery (T2-FLAIR) images were acquired to evaluate white matter lesions: TR/TE = 8500 ms/94 ms, flip angle = 150°, matrix size = 256 × 256, FOV = 230 mm × 207 mm, slice thickness = 5 mm, 20 slices. MR data analyses Data preprocessing All T1w images were preprocessed using FreeSurfer v7.4.1 (https://surfer.nmr.mgh.harvard.edu), including subject motion correction, denoising, nonuniform intensity normalization, and skull stripping. The preprocessed images were used for the segmentation of cerebral gray matter (GM) and white matter (WM), and for the calculation of total intracranial volume (TIV). Diffusion data were preprocessed using the FSL toolbox, including conversion to 4D NIFTI, visual inspection for image quality, and AC-PC alignment via rigid-body FLIRT registration. The transformation matrix was retained for gradient correction. Skull stripping was performed using the Brain Extraction Tool (BET) on the b0 image to obtain a brain mask. Further preprocessing included PCA-based denoising, Gibbs ringing removal (MRtrix), eddy current correction (FSL’s eddy), and N4 bias field correction (ANTs). Subsequently, diffusion tensors were fitted using FSL’s dtifit to yield fractional anisotropy (FA) and three diffusivity maps (Dxx, Dyy, and Dzz, corresponding to the x-, y-, and z-axes in image space, respectively). Individual FA map was warped to standard JHU-ICBM-FA template. The transformation matrix was obtained to transform three diffusivity maps into standard space. ALPS Index Calculation The DTI-ALPS index was calculated following the pipeline previously proposed by Taoka et al [12]. Four 3-mm spherical masks were placed at the lateral ventricle body level using the fsleyes viewer, covering both the bilateral projection fibers (i.e., superior and posterior corona radiata) and association fibers (i.e., superior longitudinal fasciculus). The position of the regions of interests (ROIs) was verified for each subject by two Radiologists with over 10-year experience (Y.C. and Y.W.) and any misplacement was corrected by manual adjustment. These ROI masks were then superimposed on individual diffusivity maps to extract the diffusivity values along the three axes (i.e., Dxx, Dyy, Dzz). The ALPS index was then calculated using the following formula: where Dxx_proj and Dxx_assoc represent diffusivity along the x-axis within projection and association fiber ROIs, respectively, and Dyy_proj and Dzz_assoc represent diffusivity perpendicular to the respective fibers. The ALPS index was independently calculated for the left and right hemispheres, and their average (mean ALPS) was used in subsequent statistical analyses. PVS and WMH Segmentation PVS segmentation was performed on T1-weighted images using an automated and reliable quantification method described previously [18]. The Frangi filter was applied to preprocessed T1w images to estimate the vesselness measure at each voxel [19], derived from the eigenvectors of the image Hessian matrix, using the Quantitative Imaging Toolkit (QIT, https://cabeen.io/qitwiki/). The default parameters (α = 0.5, β = 0.5) of the Frangi filter were used, while parameter c was set to half the value of the maximum Hessian norm according to a previous study. Vesselness measures were calculated across multiple scales ranging from 0.1 to 5 voxels, providing optimal sensitivity for capturing various sizes of PVS structures. Following this procedure, the volumes of PVS within the WM and basal ganglion (BG) regions, using the masks derived from T1 images, were computed. In the meantime, WM hyperintensity (WMH) mask was constructed from FLAIR images using the Lesion Segmentation Toolbox implemented in Statistical Parametric Mapping 12 (https://www.fil.ion.ucl.ac.uk/spm/software/spm12/), which was then registered to the T1w image space by applying the FLAIR-to-T1w image transformation matrix. The WMH mask were then subtracted from the PVS mask to eliminate PVS mis-segmentation caused by WMH. The PVS volume was measured in the WM, BG and the sum of these structures. The PVS volume fraction (PVSVF; PVSVF = PVS volume/intracranial volume) was then obtained to eliminate interindividual variability in brain size. FW Calculation FW maps were generated from DTI data using the single-shell FW elimination model implemented in the Diffusion Imaging in Python (DIPY, version 1.4.0) software package. For each subject, the mean cerebral FW value was extracted by applying a WM mask derived from the individual’s T1 image. To ensure accurate alignment, the T1w-based WM mask was linearly registered to the diffusion data using affine transformation, resulting in a WM mask in diffusion space. To minimize confounding from PVS, regions identified as PVS on T2WI images were subtracted from the WM mask by applying the T2-to-T1w transformation matrices. In addition, WMH regions segmented on T2-FLAIR images were also excluded from the final WM mask using the same transformation procedure (FLAIR-to-T1w). This process ensured that the calculated FW values reflected normal-appearing WM while minimizing contamination from PVS and WMH lesions. Statistical Analysis All statistical analyses were conducted using R software (version 4.4.1). Group differences in demographic and biochemical variables were assessed using independent two-sample t-tests for normally distributed continuous variables and Mann-Whitney U tests for non-normally distributed data. Categorical variables were compared using the chi-squared test. Between-group differences in cognitive measures and MRI-derived measures of perivascular function (DTI-ALPS, WM-FW, PVSVF) were estimated with general linear models (ANCOVA), with group as the fixed factor and age, sex, education (years), periventricular and deep WMH Fazekas grades, and GM volume fraction (GMVF = GM/TIV) as covariates. To assess potential multi-collinearity among covariates, we computed variance inflation factors (VIF). All adjusted VIF values were <1.5, below the commonly used threshold of 5, indicating no concerning collinearity. Because perivascular metrics may be influenced by age and sex, we tested group × age and group × sex interactions in the covariate-adjusted GLMs. To address residual imbalance in age and sex, we applied propensity score (PS) methods using age and sex to model group assignment, through both 1:1 nearest-neighbor PS matching and stabilized inverse probability of treatment weighting (IPTW). The same covariate-adjusted models were re-fit in the matched and IPTW samples. Finally, subgroup analyses stratified by median age (younger vs older) and by sex (male vs female) were further conducted to explore potential effect heterogeneity in the imaging metrics. Primary partial correlations (controlling for the covariates used in group comparisons) were restricted to MRI parameters and clinical/cognitive outcomes that showed significant between-group differences, to limit multiple-comparison burden. To assess the independent contribution of each MRI parameter, we also fit multivariable linear regression models including all four MRI metrics simultaneously, with the same covariates entered. In addition, exploratory correlations were examined for MRI metrics without group differences and clinical variables that have been previously indicated in modulating perivascular dynamics. Statistical significance was set at two-tailed p < 0.05. Multiple testing in group contrasts (four MRI metrics) and in correlation analyses was controlled using the Benjamini–Hochberg false discovery rate (FDR). For correlation analyses, FDR correction was applied separately within metabolic outcomes and within cognitive outcomes. Results Demographic and Clinical Characteristics of the Study Population Initially a total of 40 T2DM patients and 39 HCs were recruited. After excluding participants with excessive head motion (two T2DM and one HC), Fazekas scale > 3 (three T2DM and two HC), unqualified FLAIR images (one T2DM) and prediabetes (two HC), this study finally included 34 healthy controls (12 men and 22 women; mean age, 57.3 ± 6.67 years), and 34 patients with T2DM (18 men and 16 women; mean age, 59.9 ± 7.23 years) (Table 1). Age, sex, education years, BMI, SBP and DBP, blood lipid parameters, the presence of hypertension and hyperlipidemia, GMV ratio and Fazekas scales did not vary significantly between the controls and T2DM patients. Compared with HCs, T2DM patients showed significantly higher HbA1C, fasting plasma glucose (FPG), 2-hour postprandial glucose (2-h PG) and HOMA-IR levels (obtained in all HCs and 25 patients who are not dependent on insulin treatment nor insulin ) (all P values < 0.05, Table 1). In terms of cognitive performance, patients exhibited significantly longer completion time in TMT-B, and lower scores in CFT-delay and DST-forward, reflecting impairments in executive functioning, long-term visuospatial memory and working memory, respectively. In the meantime, patients also performed worse in the other cognitive tests, but the results did not reach statistical significance ( Table 1 ). Group Differences in MRI Measurements In unadjusted analyses, T2DM group showed a lower ALPS index (p < 0.001). This difference remained after covariate adjustment (general linear model controlling for age, sex, education years, Fazekas grades, and GMVF) (β = -0.091, 95% CI -0.155 to -0.027, p = 0.006, FDR-adjusted p = 0.025) ( Figure 1A ). WM-FW was higher in T2DM unadjusted (p = 0.007), but the between-group difference was not significant after adjustment (β = 0.003, p = 0.53) (Figure 1B). For PVSVF in BG and WM, as well as their sums, group means were higher in T2DM group, but none reached significance before or after adjustment (all p values > 0.05) (Figure 1C and D). Subgroup Analyses To examine whether group differences were modified by age or sex, we next tested the age × group and sex × group interaction, controlled for all covariates. None of the interaction effects reached statistical significance (all p values > 0.1). Sensitivity analyses using age- and sex-balanced samples, via PS matching and stabilized IPTW, yielded similar, non-significant between-group differences, consistent with the primary analyses. These results indicated that the influence of age or sex on MRI-derived perivascular metrics was consistent across groups, and the observed effects were not driven by imbalances in age or sex. Participants were then stratified by median age of 59 (determined based on all participants, younger vs. older) and by sex (male vs. female). Within each stratum, imaging parameters were compared between groups using the same covariate-adjusted model. In the younger subgroup, no significant between-group differences were observed for any MRI parameter (all p values > 0.05). In contrast, among older participants, the ALPS index was significantly reduced in T2DM compared with HC (β = -0.117, p = 0.013, FDR-corrected p = 0.05), suggesting that perivascular dysfunction was more prominent in older diabetic individuals. No significant differences were detected for other parameters within the older subgroup. When stratified by sex, male patients showed a significantly lower ALPS index than male controls (β = -0.125, p = 0.016, FDR-corrected p = 0.06), whereas females showed no significant group difference. Subgroup comparisons of ALPS are presented in Figure 2. Correlational Analyses We first examined interrelationships among MRI measures (ALPS index, PVSVF, and WM-FW). In the overall cohort, WM-FW showed a significant positive correlation with PVSVF in the BG regions ( r = 0.521, p < 0.001), and was negatively correlated with the mean ALPS index ( r = -0.493, p < 0.001). These correlations remained significant in either patient or control group, indicating robust associations independent of group status (Figure 3). In the T2DM group, primary correlation analyses (covariate-adjusted) were performed between DTI-ALPS and four metabolic measures/three cognitive tests that showed group differences. Specifically, average ALPS index was negatively correlated with 2-h PG (partial r = -0.46, p = 0.048) and HbA1c (partial r = -0.37, p = 0.052). But the correlations did not survive FDR correction (corrected p = 0.11). Exploratory analyses were next performed for the other MR metrics and interested outcomes including disease duration, BMI and BP control. Among the MR metrics, PVSVF in BG were both positively correlated with SBP (partial r = 0.40, p = 0.053) and DBP (partial r = 0.49, p = 0.015). None of these p value remained significant after FDR correction. In contrast, the HC group exhibited no significant associations between the above imaging parameters and clinical variables (Figure 4). Multivariable regression analyses including all four MRI metrics (ALPS, WM-FW, BG PVSVF, and WM PVS) were further performed for metabolic outcome with covariate adjustment. In the T2DM group, the ALPS index was the only metric that remained negatively associated with HbA1c (partial r = -0.43; p = 0.029) and a trend for 2-h PG (partial r = -0.50; p = 0.050). Associations for WM-FW and PVS metrics with glycemic outcomes were not significant (all p > 0.05). After FDR correction (4 MRI metrics × 4 outcomes), no association remained significant. These adjusted associations of MRI-derived perivascular metrics with glycemic outcomes in both T2DM and HC groups are summarized in forest plots (Figure 5), where each line represents the standardized β coefficient and 95% confidence interval. Discussion In the current study, we assessed perivascular fluid dynamics in T2DM patients using multimodal MRI-derived markers, including the DTI-ALPS index, WM-FW, and PVS burden, and highlighted their potential associations with glycemic control and blood pressure. Among these markers, T2DM patients exhibited significantly reduced ALPS, especially in older and male individuals, even after adjusting for key demographic and intracranial covariates. Notably, the mean ALPS index was associated with glycemic control, as measured by HbA1c and postprandial glucose. While no group-level difference was observed in PVSVF, its burden in BG was positively correlated with DBP and SBP in diabetic cohorts. Consistent with prior work, we observed a lower DTI-ALPS index in T2DM compared with controls. Recent clinical studies similarly report reduced ALPS in T2DM and link this reduction to cognitive performance and diffuse WM injury, supporting a diabetes-related perturbation of perivascular dynamics [20-22]. Similarly, animal models supported such findings, which observed retention of imaging contrast and CSF-ISF tracer in important regions such as hippocampus and hypothalamus [9], leading to cognitive worsening. Biologically, chronic hyperglycemia and insulin resistance may comprise astrocytic AQP4 polarization, stiffen small vessels, and promote low-grade inflammation, each of which may hinder CSF-ISF exchange and lower ALPS [23, 24]. Although neither the age/sex×group interaction effects nor the sensitivity analyses using age- and sex-balanced samples reached statistical significance, the stratified analyses revealed that older and male T2DM participants tended to exhibit lower ALPS indices compared with their matched controls. Such apparent discrepancies likely reflect limited statistical power caused by the moderate sample size. Nevertheless, such observations align with prior evidence that ALPS declines with advancing age in healthy adults [25], reflecting age-related reductions in glymphatic efficiency. Similarly, sex-related differences in ALPS have been variably reported, with several studies showing higher ALPS values in females than in males [25, 26]. Male T2DM patients often experience greater vascular stiffness, metabolic burden, and brain atrophy, which could jointly impair perivascular fluid transport and contribute to lower ALPS values [27]. Therefore, while the interaction effects were not statistically significant, the trends observed across subgroups may indicate potential additive or parallel effects of diabetes with aging and sex on perivascular function. These exploratory findings underscore the need for replication in larger cohorts to clarify such effects. Our findings that lower ALPS relates to poorer glycemic control are consistent with population-based evidence showing that greater cumulative blood-glucose exposure is associated with lower ALPS values [28], suggesting a potential vulnerability of perivascular fluid transport to chronic hyperglycemia. A previous T2DM study have likewise reported inverse correlations between ALPS and HbA1c, supporting a possible dose-response relationship between long-term glycemic burden and perivascular dysfunction [29]. Mechanistically, sustained or postprandial hyperglycemia can promote microvascular stiffness, low-grade inflammation, and disrupt astrocytic AQP4 polarization [30], which would be expected to impede CSF-ISF exchange and lower ALPS. Beyond perivascular function, higher glucose levels have also been linked to greater dementia risk and cerebral alterations, underscoring the broader brain vulnerability to dysglycemia [31, 32]. Taken together, both chronic (HbA1c) and fluctuating (postprandial) hyperglycemia may impose stress on perivascular clearance pathways, contributing to the observed ALPS reductions. As DTI-ALPS remains an indirect metric of perivascualr fluid dynamics, further mechanistic validation is warranted. FW represents extracellular water molecules and serves as a diffusion-based marker of interstitial fluid content [33]. Elevated WM-FW generally indicates increased extracellular fluid accumulation and has been associated with WM injury and cognitive decline, particularly in aging and in individuals with MCI and AD [34, 35]. In our study, the loss of statistical significance after covariate adjustment makes it difficult to isolate the independent effect of diabetes within a moderate sample size. Nevertheless, the observed associations between WM-FW and other MRI measures, including the ALPS index and PVS burden, are consistent with recent studies [36, 37]. These findings suggest that altered perivascular dynamics may underlie the concurrent patterns of lower ALPS and higher WM-FW, where impaired interstitial drainage along perivascular pathways could contribute to extracellular water accumulation within WM. Although we observed no between-group difference in PVS burden, within the T2DM cohort BG-PVS volume correlated positively with both systolic and diastolic BP. This pattern is consistent with evidence that higher BP load and vascular stiffness relate to greater BG-PVS burden. For example, higher ambulatory SBP was independently associated with enlarged BG-PVS [38]. A large-sample study suggested cumulative BP exposure as an independent risk factor for enlarged PVS, with BG-PVS partially mediating BP-cognition relationships [39]. Individuals with both T2DM and hypertension show larger PVS volumes, suggesting that metabolic and hemodynamic stressors jointly amplify PVS changes [40]. Taken together, our T2DM-restricted BP-PVS association likely reflects a diabetes-related vulnerability, where perivascular structures are more sensitive to hemodynamic stress. Several limitations need to be acknowledged. First, this single-center study had a modest sample, limiting generalizability and statistical power. Longitudinal and multi-center samples should clarify the temporal sequence between metabolic disturbances, perivascular dynamics changes and cognitive impairment in T2DM. Second, the adopted MRI indices are indirect proxies of fluid regulation, their use as markers still warrants validation and interpretation should be cautious. Third, complementary biomarkers (e.g., Aβ/tau, inflammatory markers) should be included to probe the underlying mechanisms of T2DM-related perivascular dysfunction. Conclusion In summary, patients with T2DM showed lower DTI-ALPS indices than controls even after adjustment for demographic and intracranial covariates, and lower ALPS was related to poorer glycemic profiles. Although overall PVS burden did not differ by group, basal-ganglia PVS volume correlated positively with blood pressure within the T2DM cohort. Together, these findings suggested altered perivascular dynamics in T2DM and highlight glycemic and blood-pressure control as clinically relevant correlates. Because ALPS and PVS are indirect MRI metrics, their causal links with these metabolic profiles still require confirmation in longitudinal and interventional studies. Declarations Ethics approval and consent to participate The current study was approved by the local Medical Research Ethics Committee of Zhongda Hospital (2024ZDSYLL021-P01). Written informed consent was obtained from all participants prior to their inclusion in the study. All procedures were conducted in accordance with the Declaration of Helsinki and relevant institutional guidelines, ensuring that participation was fully voluntary and that ethical standards were strictly upheld. Consent for publication Not applicable. Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Competing interests All the listed authors have reviewed and approved the final manuscript. There are no conflicts of interest to declare. Funding This research was supported by National Natural Science Foundation of China (82471968 to Y.C., 82427803 to S.J., 82202131 to Y.L. and 82472072 to T.T.), Natural Science Foundation of Jiangsu Province (BK20241688 to Y.C.). Author Contributions Y.C. conceived of the idea and designed the study. Y.C., Y.W. and and Y.L. drafted and revised the manuscript. Y.C., T.T., L.Z and C.L. analyzed the MR data and performed the statistical analyses. Y.C.W. and S.J. provided critical revisions to the manuscript. S.J. supervised the data collection and analyses. S.J. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. References Srikanth, V., et al., Type 2 diabetes and cognitive dysfunction-towards effective management of both comorbidities. Lancet Diabetes Endocrinol, 2020. 8 (6): p. 535-545. Biessels, G.J. and F. Despa, Cognitive decline and dementia in diabetes mellitus: mechanisms and clinical implications. Nat Rev Endocrinol, 2018. 14 (10): p. 591-604. Dove, A., et al., The impact of diabetes on cognitive impairment and its progression to dementia. Alzheimers Dement, 2021. 17 (11): p. 1769-1778. Hablitz, L.M. and M. Nedergaard, The Glymphatic System: A Novel Component of Fundamental Neurobiology. J Neurosci, 2021. 41 (37): p. 7698-7711. Mestre, H., Y. Mori, and M. Nedergaard, The Brain's Glymphatic System: Current Controversies. Trends Neurosci, 2020. 43 (7): p. 458-466. Lian, X., et al., Targeting the glymphatic system to promote alpha-synuclein clearance: a novel therapeutic strategy for Parkinson's disease. Neural Regen Res, 2026. 21 (1): p. 233-247. van Hattem, T., et al., Targeting Sleep Physiology to Modulate Glymphatic Brain Clearance. Physiology (Bethesda), 2025. 40 (3): p. 0. Lee, K.S., et al., Hyperglycemia enhances brain susceptibility to lipopolysaccharide-induced neuroinflammation via astrocyte reprogramming. J Neuroinflammation, 2024. 21 (1): p. 137. Jiang, Q., et al., Impairment of the glymphatic system after diabetes. J Cereb Blood Flow Metab, 2017. 37 (4): p. 1326-1337. Kamagata, K., et al., Association of MRI Indices of Glymphatic System With Amyloid Deposition and Cognition in Mild Cognitive Impairment and Alzheimer Disease. Neurology, 2022. 99 (24): p. e2648-e2660. Li, H., et al., Perivascular Spaces, Diffusivity Along Perivascular Spaces, and Free Water in Cerebral Small Vessel Disease. Neurology, 2024. 102 (9): p. e209306. Taoka, T., et al., Evaluation of glymphatic system activity with the diffusion MR technique: diffusion tensor image analysis along the perivascular space (DTI-ALPS) in Alzheimer's disease cases. Jpn J Radiol, 2017. 35 (4): p. 172-178. Li, S., et al., Microstructural Bias in the Assessment of Periventricular Flow as Revealed in Postmortem Brains. Radiology, 2025. 316 (3): p. e250753. Pasternak, O., et al., Free water elimination and mapping from diffusion MRI. Magn Reson Med, 2009. 62 (3): p. 717-30. Wardlaw, J.M., et al., Perivascular spaces in the brain: anatomy, physiology and pathology. Nat Rev Neurol, 2020. 16 (3): p. 137-153. American Diabetes Association Professional Practice, C., 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes-2024. Diabetes Care, 2024. 47 (Suppl 1): p. S20-S42. Wallace, T.M., J.C. Levy, and D.R. Matthews, Use and abuse of HOMA modeling. Diabetes Care, 2004. 27 (6): p. 1487-95. Sepehrband, F., et al., Image processing approaches to enhance perivascular space visibility and quantification using MRI. Sci Rep, 2019. 9 (1): p. 12351. Frangi, A.F., et al. Multiscale vessel enhancement filtering . 1998. Berlin, Heidelberg: Springer Berlin Heidelberg. Tuerxun, R., et al., Assessing interstitial fluid dynamics in type 2 diabetes mellitus and prediabetes cases through diffusion tensor imaging analysis along the perivascular space. Front Aging Neurosci, 2024. 16 : p. 1362457. Yu, S., et al., DTI-ALPS index decreased in patients with Type 2 Diabetes Mellitus. Front Neurosci, 2024. 18 : p. 1383780. Hu, P., et al., Association of diffusion tensor imaging along the perivascular space index with cognitive impairment in type 2 diabetes mellitus. Quant Imaging Med Surg, 2025. 15 (2): p. 1491-1504. Kim, Y.K., K.I. Nam, and J. Song, The Glymphatic System in Diabetes-Induced Dementia. Front Neurol, 2018. 9 : p. 867. Boyd, E.D., et al., The Glymphatic Response to the Development of Type 2 Diabetes. Biomedicines, 2024. 12 (2). Zhang, Y., et al., The Influence of Demographics and Vascular Risk Factors on Glymphatic Function Measured by Diffusion Along Perivascular Space. Front Aging Neurosci, 2021. 13 : p. 693787. Ozsahin, I., et al., Diffusion Tensor Imaging Along Perivascular Spaces (DTI-ALPS) to Assess Effects of Age, Sex, and Head Size on Interstitial Fluid Dynamics in Healthy Subjects. J Alzheimers Dis Rep, 2024. 8 (1): p. 355-361. Antal, B., et al., Type 2 diabetes mellitus accelerates brain aging and cognitive decline: Complementary findings from UK Biobank and meta-analyses. Elife, 2022. 11 . Zhang, Y., et al., Impaired glymphatic function in relation to cumulative blood glucose exposure: A population-based cohort study. IBRO Neurosci Rep, 2025. 19 : p. 437-444. Wang, N., et al., Evaluation of the glymphatic system using the DTI-ALPS index in type 2 diabetes mellitus-induced cognitive impairment. J Clin Imaging Sci, 2025. 15 : p. 31. Lashkari, A., The effects of diabetes on the glymphatic system: recent advances and mechanistic insights. Cardiovasc Diabetol Endocrinol Rep, 2025. 11 (1): p. 14. Shin, J., et al., Prediabetic HbA1c and Cortical Atrophy: Underlying Neurobiology. Diabetes Care, 2023. 46 (12): p. 2267-2272. Crane, P.K., R. Walker, and E.B. Larson, Glucose levels and risk of dementia. N Engl J Med, 2013. 369 (19): p. 1863-4. Maillard, P., et al., Cerebral white matter free water: A sensitive biomarker of cognition and function. Neurology, 2019. 92 (19): p. e2221-e2231. Berger, M., et al., Free water diffusion MRI and executive function with a speed component in healthy aging. Neuroimage, 2022. 257 : p. 119303. Zhu, Z., et al., White Matter Free Water Outperforms Cerebral Small Vessel Disease Total Score in Predicting Cognitive Decline in Persons with Mild Cognitive Impairment. J Alzheimers Dis, 2022. 86 (2): p. 741-751. Liu, X., et al., MRI free water mediates the association between diffusion tensor image analysis along the perivascular space and executive function in four independent middle to aged cohorts. Alzheimers Dement, 2025. 21 (2): p. e14453. Toro, A.R., et al., Association of MRI visible Perivascular Spaces with Early White Matter Injury. AJNR Am J Neuroradiol, 2025. Yang, S., et al., Higher ambulatory systolic blood pressure independently associated with enlarged perivascular spaces in basal ganglia. Neurol Res, 2017. 39 (9): p. 787-794. Shi, H., et al., Enlarged Perivascular Spaces in Relation to Cumulative Blood Pressure Exposure and Cognitive Impairment. Hypertension, 2023. 80 (10): p. 2088-2098. Zebarth, J., et al., Perivascular spaces mediate a relationship between diabetes and other cerebral small vessel disease markers in cerebrovascular and neurodegenerative diseases. J Stroke Cerebrovasc Dis, 2023. 32 (9): p. 107273. Table 1 Table 1. Demographics and clinical characteristics for T2DM and control groups Measures HC (n=34) T2DM (n=34) p Value Age 57.32± 6.67 59.87 ± 7.27 0.11 a Sex (male/female) 11/23 18/16 0.09 b Years of education 10.16 ± 2.38 9.76 ± 3.69 0.58 a Diabetes duration (years) / 8.95 ± 4.63 / Insulin treatment (n) / 10 / BMI (kg/m2) 24.10 ± 2.79 24.66 ± 3.06 0.40 c FPG (mmol/L) 5.53 ± 0.43 8.03 ± 2.12 <0.01 c 2h-PPG (mmol/L) 6.18 ± 1.46 14.76 ± 4.33 <0.01 a HbA1c (%, mmol/mol) 5.61 ± 0.36 7.93 ± 1.35 <0.01 a HOMA-IR 2.59 ± 1.52 3.37 ± 2.03 0.03 a Triglyceride (mmol/L) 1.27± 0.72 1.34 ± 0.63 0.65 a Cholesterol (mmol/L) 5.19 ± 0.95 5.36 ± 1.12 0.46 c LDL (mmol/L) 3.06 ± 0.65 3.25 ± 0.78 0.25 c HDL (mmol/L) 1.34 ± 0.28 1.36 ± 0.30 0.76 c SBP (mmHg) 133.42 ± 14.59 136.43 ± 13.56 0.36 a DBP (mmHg) 87.42 ± 10.93 85.11 ± 11.15 0.37 a Hypertension (n) 10 14 0.31 b Hyperlipidemia (n) 13 14 0.75 b TIV (cm 3 ) 1444.3±130.9 1473.3±126.5 0.33 c WMHVF (%) 0.46 ± 0.32 0.48 ± 0.45 0.40 a GMVF (%) 41.0 ± 1.48 41.3 ± 1.75 0.33 c Fazekas-PVWMH (range) 1-2 1-2 0.65 a Fazekas-DWMH (range) 0-2 0-2 0.08 a Cognitive performance * MMSE 28.68±1.21 28.34±1.28 0.23 AVLT-immediate 6.72 ± 2.06 6.12 ± 1.41 0.08 AVLT-5min recall 6.97± 1.88 6.81 ± 2.18 0.73 AVLT-20min recall 6.53 ± 2.15 5.92 ± 2.56 0.27 DST-forward 7.42 ± 1.59 6.94 ± 1.22 0.15 DST-backward 4.55 ± 1.44 4.03 ± 0.91 0.03 CDT 3.43 ± 0.55 3.20 ± 0.63 0.10 CFT-copy 34.78 ± 1.51 34.24 ± 2.36 0.42 CFT-delay 17.63 ± 6.10 14.21 ± 4.93 0.01* Time in TMT-A 62.11 ± 13.68 68.38 ± 23.37 0.16 Time in TMT-B 144.11 ± 50.43 177.08 ± 55.10 0.01* Data are presented as mean ± standard deviation or the number of patients unless otherwise stated. * Comparison was adjusted for age, sex and education level. a P values for group differences obtained using Mann-Whitney U test; b p values obtained using Chi-squared test; c p values obtained using two-sample t-test. Bold means p value less than 0.05. Abbreviations: BMI, body mass index; FPG, fasting plasma glucose; 2h-PPG, 2-hour postprandial glucose; HbA1c, Hemoglobin-A1c; HOMA-IR, Homeostatic Model Assessment for Insulin Resistance; LDL, Low-density lipoprotein; HDL, High-density lipoprotein; SBP, systolic blood pressure; DBP, diastolic blood pressure; TIV, total intracranial volume; WMHVF, WMH volume fraction; GMVF, GM volume fraction; MoCA, The Montreal Cognitive Assessment; AVLT, Auditory Verbal Learning Test; DST, digit span test; CDT, clock drawing test; CFT, complex figure test; TMT, trail-making test. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7859506","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":549321035,"identity":"e82c742c-e905-4f71-8993-8ed47566ea15","order_by":0,"name":"Ying Cui","email":"","orcid":"","institution":"Zhongda Hospital, Medical School of Southeast University","correspondingAuthor":false,"prefix":"","firstName":"Ying","middleName":"","lastName":"Cui","suffix":""},{"id":549321036,"identity":"9532645f-f1f7-4f42-9bd0-a17cd9968116","order_by":1,"name":"Yu Wang","email":"","orcid":"","institution":"Zhongda Hospital, Medical School of Southeast University","correspondingAuthor":false,"prefix":"","firstName":"Yu","middleName":"","lastName":"Wang","suffix":""},{"id":549321037,"identity":"ec1977ef-265a-4e76-a46e-ccf58b056dbf","order_by":2,"name":"Ying Luan","email":"","orcid":"","institution":"Zhongda Hospital, Medical School of Southeast University","correspondingAuthor":false,"prefix":"","firstName":"Ying","middleName":"","lastName":"Luan","suffix":""},{"id":549321038,"identity":"9834e209-e33c-4054-ac22-232fc99d8c72","order_by":3,"name":"Tianyu Tang","email":"","orcid":"","institution":"Zhongda Hospital, Medical School of Southeast University","correspondingAuthor":false,"prefix":"","firstName":"Tianyu","middleName":"","lastName":"Tang","suffix":""},{"id":549321039,"identity":"c3507e1b-665a-40fd-8621-b29abb29b509","order_by":4,"name":"Lijuan Zheng","email":"","orcid":"","institution":"Zhongda Hospital, Medical School of Southeast University","correspondingAuthor":false,"prefix":"","firstName":"Lijuan","middleName":"","lastName":"Zheng","suffix":""},{"id":549321040,"identity":"86a6229c-8c02-423f-a2e9-231b519be767","order_by":5,"name":"Chunqiang Lu","email":"","orcid":"","institution":"Zhongda Hospital, Medical School of Southeast University","correspondingAuthor":false,"prefix":"","firstName":"Chunqiang","middleName":"","lastName":"Lu","suffix":""},{"id":549321041,"identity":"049c0faa-d300-4fd2-9899-e7f3923778e0","order_by":6,"name":"Yuancheng Wang","email":"","orcid":"","institution":"Zhongda Hospital, Medical School of Southeast University","correspondingAuthor":false,"prefix":"","firstName":"Yuancheng","middleName":"","lastName":"Wang","suffix":""},{"id":549321042,"identity":"d8fcfe43-0b40-478a-8865-9bf31d8c7a0f","order_by":7,"name":"Shenghong Ju","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAArUlEQVRIiWNgGAWjYBACxgYexgcQZgLxWpgNSNPCwMDDJkGaFub+s8cqf1QcZuBnzzFg+LmDGIfNyEu7zXPmMINkzxsDxt4zRGnhMbvN2HaYweBGjgEzYxsxWvrPmBX+/HeYwZ54LQ05Zgy8DUBbJIjWMiPHWJrnWDqPxJlnBQd7idFi2H/G8OOPGms5/vbkjQ9+EqWlAULzgIgDRGhgYJAnStUoGAWjYBSMbAAA98cyem7iZY0AAAAASUVORK5CYII=","orcid":"","institution":"Zhongda Hospital, Medical School of Southeast University","correspondingAuthor":true,"prefix":"","firstName":"Shenghong","middleName":"","lastName":"Ju","suffix":""}],"badges":[],"createdAt":"2025-10-14 14:08:23","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7859506/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7859506/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96694412,"identity":"f57c10ec-f0b6-4240-a1f9-00b668331c2f","added_by":"auto","created_at":"2025-11-25 07:17:22","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":5290106,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBetween-Group Differences in MRI Measurements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe Violin and box plots illustrating the mean ALPS index, FW-WM, PVSVF-BG, and PVSVF-WM in healthy controls and T2DM patients. Group differences and \u003cem\u003ep\u003c/em\u003evalues were determined using general linear models with covariates adjustment. Statistical significance set at FDR-corrected \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05.\u003c/p\u003e","description":"","filename":"figure1MRcompare.png","url":"https://assets-eu.researchsquare.com/files/rs-7859506/v1/ec22953cb0dbb5bb3f071032.png"},{"id":96694410,"identity":"09c63dcd-f06b-48a9-8065-bd2551f7603f","added_by":"auto","created_at":"2025-11-25 07:17:21","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":2163747,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eComparison of MRI Measurements Across Age and Sex Subgroups\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBar plots showing the adjusted mean ALPS in control and T2DM groups, stratified by sex (male/female) and age (younger/older, split by median age of 59 years old). Estimated marginal means and standard errors were derived from general linear models with covariates adjustment. \u003cem\u003eP\u003c/em\u003e values indicate group differences within each stratified subgroup, with statistical significance set at uncorrected \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05.\u003c/p\u003e","description":"","filename":"figure2subgroup.png","url":"https://assets-eu.researchsquare.com/files/rs-7859506/v1/eb07777a62acc1b4502d71e0.png"},{"id":96711409,"identity":"7c5743b0-81e5-4ed5-bdef-3aa87ce25410","added_by":"auto","created_at":"2025-11-25 10:11:58","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":6662518,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCorrelations among MRI-derived glymphatic markers\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eScatter plots illustrate the correlations between key MRI-derived glymphatic markers. In both healthy controls and T2DM patients, higher white matter free water (WM FW) levels were associated with greater perivascular space volume fraction (PVSVF) in the basal ganglia (BG), and with lower mean ALPS index values, indicating potential glymphatic dysfunction. Blue dots indicate control subjects; red dots indicate T2DM subjects.\u003c/p\u003e","description":"","filename":"figure3MRIcorrelations.png","url":"https://assets-eu.researchsquare.com/files/rs-7859506/v1/0f85a675e7be68f8340115c1.png"},{"id":96694415,"identity":"b414d181-7b80-404a-8261-b3154b1701c5","added_by":"auto","created_at":"2025-11-25 07:17:26","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":9647227,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCorrelations between MRI measurements and in T2DM group\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePartial correlation scatter plots are shown for the associations between imaging indices and clinical measures or cognitive scores, after adjusting for covariates. Residuals from linear regression models are plotted. The fitted regression line and the corresponding β and uncorrected p values are displayed for each group.\u003c/p\u003e","description":"","filename":"figure4correlaitons.png","url":"https://assets-eu.researchsquare.com/files/rs-7859506/v1/d1dc778ede5067d4b088e682.png"},{"id":96694411,"identity":"f84abd74-5ad5-4d0c-8d2d-44c22330ab1a","added_by":"auto","created_at":"2025-11-25 07:17:21","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":2156540,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMultivariable regression analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eForest plots illustrating standardized regression coefficients (β) and 95% confidence intervals (CIs) derived from multivariable linear regression models. Each horizontal line represents one imaging marker, with the dot indicating the standardized β. The size of each dot reflects the partial R², indicating the proportion of variance in the outcome independently explained by that variable after adjustment for covariates.\u003c/p\u003e","description":"","filename":"figure5forest.png","url":"https://assets-eu.researchsquare.com/files/rs-7859506/v1/ff0cdee875b81d24d7c41142.png"},{"id":106394645,"identity":"dd8e8dfb-7762-4e29-ac76-ad3882b2f930","added_by":"auto","created_at":"2026-04-08 07:44:37","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":20908718,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7859506/v1/a1568e36-d11f-44a1-b76c-8ea9f44967b1.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"MRI Markers of Perivascular Fluid Dynamics in Type 2 Diabetes and Their Associations with Glycemic Control","fulltext":[{"header":"Background","content":"\u003cp\u003eType 2 diabetes mellitus (T2DM) is characterized by insulin resistance (IR) and abnormal glucose metabolism. In addition to its related metabolic consequences, T2MD is received increasing attention for its adverse impact on brain health [1]. Epidemiological evidence shows that cognitive decrements are more prevalent and progresses more rapidly among T2DM patients [2]. T2DM nearly doubles the risk of developing dementia compared to age-matched non-diabetic subjects, especially in elderly subjects [3]. Due to the increasing burden of T2DM worldwide, potential causative pathways in T2DM-associated cognitive dysfunction need to be identified in order to develop course-modifying therapies.\u003c/p\u003e\n\u003cp\u003eThe brain perivascular network is recently identified that is responsible for clearing metabolic waste products through cerebrospinal fluid (CSF) and interstitial fluid (ISF) exchange, a process facilitated by aquaporin-4 (AQP4) on astrocytic endfeet [4]. Disruption of this fluid dynamics has been increasingly implicated in neurodegenerative disease, particularly Alzheimer’s disease (AD), where impaired clearance of amyloid-β and tau is observed [5]. Importantly, the perivascular pathway is emerging as a therapeutic target, as pharmacologic and lifestyle interventions can reduce waste accumulation and improve cognition in preclinical models [6, 7]. \u003c/p\u003e\n\u003cp\u003eAlthough the exact mechanism of T2DM-related cognitive decline is still under debate, perivascular dysfunction has been proposed as an important contributor. In diabetic animal models, CSF tracer clearance is slowed, accompanied by reduced AQP4 expression/polarization and greater amyloid burden that correlate with cognitive deficits [8, 9]. Despite these evidences, robust in-vivo human evidence of disturbed perivascular fluid dynamics in T2DM remains limited, underscoring the need for further investigation.\u003c/p\u003e\n\u003cp\u003ePromising contrast‐free MRI approaches have been proposed to indirectly probe perivascular function [10, 11]. The diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) index is among the most used, which computes a ratio of diffusivities sampled in periventricular white matter (WM) near deep medullary veins at the lateral ventricle body [12]. Notably, ALPS has been recently challenged due to its sensitivity to ROI placement, local fiber geometry and acquisition parameters [13], and should therefore be regarded as an indirect proxy instead of a measure of waste clearance. White matter free water (WM-FW), quantifiable via diffusion MRI, suggests extracellular fluid accumulation and reflects water-content changes that may arise from attenuated CSF-ISF exchange [14]. The visible perivascular space (PVS) burden is often interpreted as a drainage system, the enlargement of which has been hypothesized to be secondary to impaired interstitial drainage [15]. While none of these metrics alone constitutes a definitive readout of glymphatic activity, their combined measurements may provide a more comprehensive signature of perivascular fluid dynamics and potential waste-clearance deficits [11]. In T2DM, however, systematic evaluations of ALPS, WM-FW, and PVS, and their relationships to metabolic dysfunction and cognitive decline remain limited.\u003c/p\u003e\n\u003cp\u003eThis study evaluates MRI-derived metrics of perivascular fluid dynamics in T2DM, and tests their associations with metabolic dysregulation and cognitive performance. We hypothesized that T2DM would show alterations in these perivascular markers and that worse metabolic profiles would relate to more adverse imaging signatures and poorer cognition. Clarifying these links may enhance the understanding of diabetes-related brain disturbances and inform earlier risk stratification.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eParticipant Recruitment\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe inclusion criteria for all participants included: (1) aged between 50 to 75 years old; (2) right handedness; (3) a minimum education of 6 years; (4) no contraindications of MRI scanning, screened using MRI safety questionnaire; (5) previous routine CT/MRI examination of the head showed no severe brain abnormality or only mild leukoaraiosis, which will be reassured after the following MR scan. T2DM patients were recruited from the department of Endocrinology or the local community, and met the the clinical diagnostic criteria for T2DM recommended by the American Diabetes Association (ADA) [16]. Each patient had a documented disease duration of at least one year and was under regular self-monitoring of blood glucose. A group of healthy control subjects (HCs) was recruited through community advertisements, who were matched to the T2DM pateints with respect to age, sex, and educational level.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical Assessments and Biochemical Measurements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOn the study day, all participants arrived at the hospital at 7:00 a.m. following an overnight fast of at least 8 hours. Fasting venous blood samples were collected for the assessment of key metabolic indicators, including fasting plasma glucose (FPG), glycated hemoglobin (HbA1c), fasting insulin, and a comprehensive lipid profile (triglycerides, total cholesterol, low-density lipoprotein [LDL] cholesterol, and high-density lipoprotein [HDL] cholesterol). To assess glucose metabolism in the HC group, a standard oral glucose tolerance test (OGTT) was administered, consisting of 75 grams of dextrose monohydrate dissolved in 250 mL of water. At 9:00 a.m., two hours after OGTT administration in the HC group and two hours after a typical breakfast in the T2DM group, additional blood samples were collected to evaluate the 2-hour postprandial glucose (2-h PG). Insulin resistance was measured by Homeostatic Model Assessment for Insulin Resistance (HOMA-IR), calculated as (fasting insulin\u0026nbsp;×\u0026nbsp;fasting glucose)/22 [17]. Blood pressure was measured three times throughout the visit at standardized intervals, and the average of these readings was recorded. During the visit, a structured questionnaire was also administered to all participants to collect detailed information on medical history, current medication use, systemic illnesses, and family history. In addition, weight, height, waist and hip circumference were measured and recorded.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAfter the clinical assessments and biochemical measurements, participants were excluded if they met any of the following criteria: (1) a documented history of neurological or psychiatric disorders unrelated to T2DM; (2) prior brain lesions (such as tumor, stroke or traumatic injury) or intracranial surgeries; (3) serious medical emergencies (such as heart, lung, liver and kidney failure); (4) current/past history of major physical diseases (such as thyroid/parathyroid dysfunction, malignant tumors, autoimmune diseases and other endocrine disorders); (5) HCs who met the criteria for prediabetes or T2DM; (6) subjects with severe WMH, defined as a Fazekas score \u0026gt;2 (i.e., grade 3) in either periventricular WMH (PVWMH) or deep WMH (DWMH).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNeuropsychological tests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eEach participant underwent a comprehensive battery of neuropsychological tests to assess general cognitive status and multiple specific domains. Global cognitive functioning was evaluated using the Mini-Mental State Examination (MMSE). Visuospatial processing was assessed using the Clock-Drawing Test (CDT) and the copy trial of the Complex Figure Test (CFT-copy). Episodic memory was assessed using the Auditory Verbal Learning Test (AVLT), working memory with the Digit Span Test (DST), and spatial memory with the 20min delayed recall trial of the Complex Figure Test (CFT-delay). Processing speed and attention were evaluated using Part A of the Trail-Making Test (TMT-A), and executive function with Part B of the Trail-Making Test (TMT-B). All tests were administered in a fixed sequence, and the entire battery required approximately 60 minutes to complete. Participants scoring below 24 on the MMSE were considered to meet the criteria for dementia and were excluded from the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMRI data acquisition\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSubjects were scanned using a 3 Tesla MR instrument (MAGENETOM Trio, Siemens Medical Solutions, Erlangen, Germany) with a birdcage head coil. Foam padding was used to minimize head motion and earplugs were used to reduce the scanner noise. All subjects were instructed to rest with their eyes closed, not think of anything in particular, not fall asleep and avoid any head motion during fMRI scanning. High-resolution three-dimensional T1-weighted structural images were acquired using a magnetization-prepared rapid acquisition gradient-echo (MPRAGE) sequence (repetition time [TR]/echo time [TE] = 1900ms/2.48 ms, flip angle = 9°, acquisition matrix size = 256 × 256, field of view [FOV] = 250 mm × 250 mm, slice thickness = 1 mm, 176 slices in the sagittal orientation). Diffusion tensor images (DTI) were obtained using a spin echo-based echo planar imaging sequence (TR/TE = 10000 ms/95 ms, FOV = 256 mm × 256 mm, matrix size = 128 × 128, section thickness = 2 mm). Images were obtained with both 30-direction encoding (b-value = 1000 s/mm\u003csup\u003e2\u003c/sup\u003e for each direction) and no diffusion encoding (b = 0 s/mm\u003csup\u003e2\u003c/sup\u003e), consisting of 70 contiguous axial slices covering the whole brain. T2-weighred images based on spin echo sequence were obtained: TR/TE = 6000 ms/95 ms, field of view = 212mm\u0026nbsp;×\u0026nbsp;229 mm, slice thickness = 5 mm. Finally, T2 fluid-attenuated inversion recovery (T2-FLAIR) images were acquired to evaluate white matter lesions:\u0026nbsp;TR/TE = 8500 ms/94 ms, flip angle = 150°, matrix size = 256 × 256, FOV = 230 mm × 207 mm, slice thickness = 5 mm, 20 slices.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMR data analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eData preprocessing\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll T1w images were preprocessed using FreeSurfer v7.4.1 (https://surfer.nmr.mgh.harvard.edu), including subject motion correction, denoising, nonuniform intensity normalization, and skull stripping. The preprocessed images were used for the segmentation of cerebral gray matter (GM) and white matter (WM), and for the calculation of total intracranial volume (TIV).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eDiffusion data were preprocessed using the FSL toolbox, including conversion to 4D NIFTI, visual inspection for image quality, and AC-PC alignment via rigid-body FLIRT registration. The transformation matrix was retained for gradient correction. Skull stripping was performed using the Brain Extraction Tool (BET) on the b0 image to obtain a brain mask. Further preprocessing included PCA-based denoising, Gibbs ringing removal (MRtrix), eddy current correction (FSL’s eddy), and N4 bias field correction (ANTs). Subsequently, diffusion tensors were fitted using FSL’s dtifit to \u0026nbsp;yield fractional anisotropy (FA) and three diffusivity maps (Dxx, Dyy, and Dzz, corresponding to the x-, y-, and z-axes in image space, respectively). Individual FA map was warped to standard JHU-ICBM-FA template. The transformation matrix was obtained to transform three diffusivity maps into standard space.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eALPS Index Calculation\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe DTI-ALPS index was calculated following the pipeline previously proposed by Taoka et al [12]. Four 3-mm spherical masks were placed at the lateral ventricle body level using the fsleyes viewer, covering both the bilateral projection fibers (i.e., superior and posterior corona radiata) and association fibers (i.e., superior longitudinal fasciculus). The position of the regions of interests (ROIs) was verified for each subject by two Radiologists with over 10-year experience (Y.C. and Y.W.) and any misplacement was corrected by manual adjustment. These ROI masks were then superimposed on individual diffusivity maps to extract the diffusivity values along the three axes (i.e., Dxx, Dyy, Dzz). The ALPS index was then calculated using the following formula:\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003c/p\u003e\n\u003cp\u003ewhere Dxx_proj and Dxx_assoc represent diffusivity along the x-axis within projection and association fiber ROIs, respectively, and Dyy_proj and Dzz_assoc represent diffusivity perpendicular to the respective fibers. The ALPS index was independently calculated for the left and right hemispheres, and their average (mean ALPS) was used in subsequent statistical analyses.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003ePVS and WMH Segmentation\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003ePVS segmentation was performed on T1-weighted images using an automated and reliable quantification method described previously [18]. The Frangi filter was applied to preprocessed T1w images to estimate the vesselness measure at each voxel [19], derived from the eigenvectors of the image Hessian matrix, using the Quantitative Imaging Toolkit (QIT, https://cabeen.io/qitwiki/). The default parameters (α = 0.5, β = 0.5) of the Frangi filter were used, while parameter c was set to half the value of the maximum Hessian norm according to a previous study. Vesselness measures were calculated across multiple scales ranging from 0.1 to 5 voxels, providing optimal sensitivity for capturing various sizes of PVS structures. Following this procedure, the volumes of PVS within the WM and basal ganglion (BG) regions, using the masks derived from T1 images, were computed. In the meantime, WM hyperintensity (WMH) mask was constructed from FLAIR images using the Lesion Segmentation Toolbox implemented in Statistical Parametric Mapping 12 (https://www.fil.ion.ucl.ac.uk/spm/software/spm12/), which was then registered to the T1w image space by applying the FLAIR-to-T1w image transformation matrix. The WMH mask were then subtracted from the PVS mask to eliminate PVS mis-segmentation caused by WMH. The PVS volume was measured in the WM, BG and the sum of these structures. The PVS volume fraction (PVSVF; PVSVF = PVS volume/intracranial volume) was then obtained to eliminate interindividual variability in brain size.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e\u003cem\u003eFW Calculation\u003c/em\u003e\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFW maps were generated from DTI data using the single-shell FW elimination model implemented in the Diffusion Imaging in Python (DIPY, version 1.4.0) software package. For each subject, the mean cerebral FW value was extracted by applying a WM mask derived from the individual’s T1 image. To ensure accurate alignment, the T1w-based WM mask was linearly registered to the diffusion data using affine transformation, resulting in a WM mask in diffusion space. To minimize confounding from PVS, regions identified as PVS on T2WI images were subtracted from the WM mask by applying the T2-to-T1w transformation matrices. In addition, WMH regions segmented on T2-FLAIR images were also excluded from the final WM mask using the same transformation procedure (FLAIR-to-T1w). This process ensured that the calculated FW values reflected normal-appearing WM while minimizing contamination from PVS and WMH lesions.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll statistical analyses were conducted using R software (version 4.4.1). Group differences in demographic and biochemical variables were assessed using independent two-sample t-tests for normally distributed continuous variables and Mann-Whitney U tests for non-normally distributed data. Categorical variables were compared using the chi-squared test.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eBetween-group differences in cognitive measures and MRI-derived measures of perivascular function (DTI-ALPS, WM-FW, PVSVF) were estimated with general linear models (ANCOVA), with group as the fixed factor and age, sex, education (years), periventricular and deep WMH Fazekas grades, and GM volume fraction (GMVF = GM/TIV) as covariates. To assess potential multi-collinearity among covariates, we computed variance inflation factors (VIF). All adjusted VIF values were \u0026lt;1.5, below the commonly used threshold of 5, indicating no concerning collinearity.\u003c/p\u003e\n\u003cp\u003eBecause perivascular metrics may be influenced by age and sex, we tested group\u0026nbsp;×\u0026nbsp;age and group\u0026nbsp;×\u0026nbsp;sex interactions in the covariate-adjusted GLMs. To address residual imbalance in age and sex, we applied propensity score (PS) methods using age and sex to model group assignment, through both 1:1 nearest-neighbor PS matching and stabilized inverse probability of treatment weighting (IPTW). The same covariate-adjusted models were re-fit in the matched and IPTW samples. Finally, subgroup analyses stratified by median age (younger vs older) and by sex (male vs female) were further conducted to explore potential effect heterogeneity in the imaging metrics.\u003c/p\u003e\n\u003cp\u003ePrimary partial correlations (controlling for the covariates used in group comparisons) were restricted to MRI parameters and clinical/cognitive outcomes that showed significant between-group differences, to limit multiple-comparison burden. To assess the independent contribution of each MRI parameter, we also fit multivariable linear regression models including all four MRI metrics simultaneously, with the same covariates entered. In addition, exploratory correlations were examined for MRI metrics without group differences and clinical variables that have been previously indicated in modulating perivascular dynamics.\u003c/p\u003e\n\u003cp\u003eStatistical significance was set at two-tailed \u003cem\u003ep\u0026nbsp;\u003c/em\u003e\u0026lt; 0.05. Multiple testing in group contrasts (four MRI metrics) and in correlation analyses was controlled using the Benjamini–Hochberg false discovery rate (FDR). For correlation analyses, FDR correction was applied separately within metabolic outcomes and within cognitive outcomes.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eDemographic and Clinical Characteristics of the Study Population\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eInitially a total of 40 T2DM patients and 39 HCs were recruited. After excluding participants with excessive head motion (two T2DM and one HC), Fazekas scale \u0026gt; 3 (three T2DM and two HC), unqualified FLAIR images (one T2DM) and prediabetes (two HC), this study finally included 34 healthy controls (12 men and 22 women; mean age, 57.3 \u0026plusmn; 6.67 years), and 34 patients with T2DM (18 men and 16 women; mean age, 59.9 \u0026plusmn; 7.23 years) (Table 1). Age, sex, education years, BMI, SBP and DBP, blood lipid parameters, the presence of hypertension and hyperlipidemia, GMV ratio and Fazekas scales did not vary significantly between the controls and T2DM patients. Compared with HCs, T2DM patients showed significantly higher HbA1C, fasting plasma glucose (FPG), 2-hour postprandial glucose (2-h PG) and HOMA-IR levels (obtained in all HCs and 25 patients who are not dependent on insulin treatment nor insulin ) (all \u003cem\u003eP\u003c/em\u003e values \u0026lt; 0.05, Table 1).\u003c/p\u003e\n\n\u003cp\u003eIn terms of cognitive performance, patients exhibited significantly longer completion time in TMT-B, and lower scores in CFT-delay and DST-forward, reflecting impairments in executive functioning, long-term visuospatial memory and working memory, respectively. In the meantime, patients also performed worse in the other cognitive tests, but the results did not reach statistical significance (\u003cstrong\u003eTable 1\u003c/strong\u003e). \u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eGroup Differences in MRI Measurements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn unadjusted analyses, T2DM group showed a lower ALPS index (p \u0026lt; 0.001). This difference remained after covariate adjustment (general linear model controlling for age, sex, education years, Fazekas grades, and GMVF) (\u0026beta; = -0.091, 95% CI -0.155 to -0.027, \u003cem\u003ep\u003c/em\u003e = 0.006, FDR-adjusted \u003cem\u003ep\u003c/em\u003e = 0.025) (\u003cstrong\u003eFigure 1A\u003c/strong\u003e). WM-FW was higher in T2DM unadjusted (p = 0.007), but the between-group difference was not significant after adjustment (\u0026beta; = 0.003, \u003cem\u003ep\u003c/em\u003e = 0.53) (Figure 1B). For PVSVF in BG and WM, as well as their sums, group means were higher in T2DM group, but none reached significance before or after adjustment (all \u003cem\u003ep\u003c/em\u003e values \u0026gt; 0.05) (Figure 1C and D).\u003c/p\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSubgroup Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo examine whether group differences were modified by age or sex, we next tested the age \u0026times; group and sex \u0026times; group interaction, controlled for all covariates. None of the interaction effects reached statistical significance (all \u003cem\u003ep\u003c/em\u003e values \u0026gt; 0.1). Sensitivity analyses using age- and sex-balanced samples, via PS matching and stabilized IPTW, yielded similar, non-significant between-group differences, consistent with the primary analyses. These results indicated that the influence of age or sex on MRI-derived perivascular metrics was consistent across groups, and the observed effects were not driven by imbalances in age or sex. \u003c/p\u003e\n\n\u003cp\u003eParticipants were then stratified by median age of 59 (determined based on all participants, younger vs. older) and by sex (male vs. female). Within each stratum, imaging parameters were compared between groups using the same covariate-adjusted model. In the younger subgroup, no significant between-group differences were observed for any MRI parameter (all \u003cem\u003ep\u003c/em\u003e values \u0026gt; 0.05). In contrast, among older participants, the ALPS index was significantly reduced in T2DM compared with HC (\u0026beta; = -0.117, \u003cem\u003ep\u003c/em\u003e = 0.013, FDR-corrected \u003cem\u003ep\u003c/em\u003e = 0.05), suggesting that perivascular dysfunction was more prominent in older diabetic individuals. No significant differences were detected for other parameters within the older subgroup. When stratified by sex, male patients showed a significantly lower ALPS index than male controls (\u0026beta; = -0.125, \u003cem\u003ep\u003c/em\u003e = 0.016, FDR-corrected \u003cem\u003ep\u003c/em\u003e = 0.06), whereas females showed no significant group difference. Subgroup comparisons of ALPS are presented in Figure 2.\u003c/p\u003e\n\u003cp\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCorrelational Analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe first examined interrelationships among MRI measures (ALPS index, PVSVF, and WM-FW). In the overall cohort, WM-FW showed a significant positive correlation with PVSVF in the BG regions (\u003cem\u003er\u003c/em\u003e = 0.521, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001), and was negatively correlated with the mean ALPS index (\u003cem\u003er\u003c/em\u003e = -0.493, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001). These correlations remained significant in either patient or control group, indicating robust associations independent of group status (Figure 3). \u003c/p\u003e\n\n\u003cp\u003eIn the T2DM group, primary correlation analyses (covariate-adjusted) were performed between DTI-ALPS and four metabolic measures/three cognitive tests that showed group differences. Specifically, average ALPS index was negatively correlated with 2-h PG (partial \u003cem\u003er\u003c/em\u003e = -0.46, \u003cem\u003ep\u003c/em\u003e = 0.048) and HbA1c (partial \u003cem\u003er\u003c/em\u003e = -0.37, \u003cem\u003ep\u003c/em\u003e = 0.052). But the correlations did not survive FDR correction (corrected \u003cem\u003ep\u003c/em\u003e = 0.11). Exploratory analyses were next performed for the other MR metrics and interested outcomes including disease duration, BMI and BP control. Among the MR metrics, PVSVF in BG were both positively correlated with SBP (partial \u003cem\u003er\u003c/em\u003e = 0.40, \u003cem\u003ep\u003c/em\u003e = 0.053) and DBP (partial \u003cem\u003er\u003c/em\u003e = 0.49, \u003cem\u003ep\u003c/em\u003e = 0.015). None of these \u003cem\u003ep\u003c/em\u003e value remained significant after FDR correction. In contrast, the HC group exhibited no significant associations between the above imaging parameters and clinical variables (Figure 4). \u003c/p\u003e\n\n\u003cp\u003eMultivariable regression analyses including all four MRI metrics (ALPS, WM-FW, BG PVSVF, and WM PVS) were further performed for metabolic outcome with covariate adjustment. In the T2DM group, the ALPS index was the only metric that remained negatively associated with HbA1c (partial \u003cem\u003er\u003c/em\u003e = -0.43; \u003cem\u003ep\u003c/em\u003e = 0.029) and a trend for 2-h PG (partial \u003cem\u003er\u003c/em\u003e = -0.50; \u003cem\u003ep\u003c/em\u003e = 0.050). Associations for WM-FW and PVS metrics with glycemic outcomes were not significant (all \u003cem\u003ep\u003c/em\u003e \u0026gt; 0.05). After FDR correction (4 MRI metrics \u0026times; 4 outcomes), no association remained significant. These adjusted associations of MRI-derived perivascular metrics with glycemic outcomes in both T2DM and HC groups are summarized in forest plots (Figure 5), where each line represents the standardized \u0026beta; coefficient and 95% confidence interval.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn the current study, we assessed perivascular fluid dynamics in T2DM patients using multimodal MRI-derived markers, including the DTI-ALPS index, WM-FW, and PVS burden, and highlighted their potential associations with glycemic control and blood pressure. Among these markers, T2DM patients exhibited significantly reduced ALPS, especially in older and male individuals, even after adjusting for key demographic and intracranial covariates. Notably, the mean ALPS index was associated with glycemic control, as measured by HbA1c and postprandial glucose. While no group-level difference was observed in PVSVF, its burden in BG was positively correlated with DBP and SBP in diabetic cohorts. \u003c/p\u003e\n\u003cp\u003eConsistent with prior work, we observed a lower DTI-ALPS index in T2DM compared with controls. Recent clinical studies similarly report reduced ALPS in T2DM and link this reduction to cognitive performance and diffuse WM injury, supporting a diabetes-related perturbation of perivascular dynamics [20-22]. Similarly, animal models supported such findings, which observed retention of imaging contrast and CSF-ISF tracer in important regions such as hippocampus and hypothalamus [9], leading to cognitive worsening. Biologically, chronic hyperglycemia and insulin resistance may comprise astrocytic AQP4 polarization, stiffen small vessels, and promote low-grade inflammation, each of which may hinder CSF-ISF exchange and lower ALPS [23, 24]. \u003c/p\u003e\n\u003cp\u003eAlthough neither the age/sex×group interaction effects nor the sensitivity analyses using age- and sex-balanced samples reached statistical significance, the stratified analyses revealed that older and male T2DM participants tended to exhibit lower ALPS indices compared with their matched controls. Such apparent discrepancies likely reflect limited statistical power caused by the moderate sample size. Nevertheless, such observations align with prior evidence that ALPS declines with advancing age in healthy adults [25], reflecting age-related reductions in glymphatic efficiency. Similarly, sex-related differences in ALPS have been variably reported, with several studies showing higher ALPS values in females than in males [25, 26]. Male T2DM patients often experience greater vascular stiffness, metabolic burden, and brain atrophy, which could jointly impair perivascular fluid transport and contribute to lower ALPS values [27]. Therefore, while the interaction effects were not statistically significant, the trends observed across subgroups may indicate potential additive or parallel effects of diabetes with aging and sex on perivascular function. These exploratory findings underscore the need for replication in larger cohorts to clarify such effects.\u003c/p\u003e\n\u003cp\u003eOur findings that lower ALPS relates to poorer glycemic control are consistent with population-based evidence showing that greater cumulative blood-glucose exposure is associated with lower ALPS values [28], suggesting a potential vulnerability of perivascular fluid transport to chronic hyperglycemia. A previous T2DM study have likewise reported inverse correlations between ALPS and HbA1c, supporting a possible dose-response relationship between long-term glycemic burden and perivascular dysfunction [29]. Mechanistically, sustained or postprandial hyperglycemia can promote microvascular stiffness, low-grade inflammation, and disrupt astrocytic AQP4 polarization [30], which would be expected to impede CSF-ISF exchange and lower ALPS. Beyond perivascular function, higher glucose levels have also been linked to greater dementia risk and cerebral alterations, underscoring the broader brain vulnerability to dysglycemia [31, 32]. Taken together, both chronic (HbA1c) and fluctuating (postprandial) hyperglycemia may impose stress on perivascular clearance pathways, contributing to the observed ALPS reductions. As DTI-ALPS remains an indirect metric of perivascualr fluid dynamics, further mechanistic validation is warranted.\u003c/p\u003e\n\u003cp\u003eFW represents extracellular water molecules and serves as a diffusion-based marker of interstitial fluid content [33]. Elevated WM-FW generally indicates increased extracellular fluid accumulation and has been associated with WM injury and cognitive decline, particularly in aging and in individuals with MCI and AD [34, 35]. In our study, the loss of statistical significance after covariate adjustment makes it difficult to isolate the independent effect of diabetes within a moderate sample size. Nevertheless, the observed associations between WM-FW and other MRI measures, including the ALPS index and PVS burden, are consistent with recent studies [36, 37]. These findings suggest that altered perivascular dynamics may underlie the concurrent patterns of lower ALPS and higher WM-FW, where impaired interstitial drainage along perivascular pathways could contribute to extracellular water accumulation within WM.\u003c/p\u003e\n\u003cp\u003eAlthough we observed no between-group difference in PVS burden, within the T2DM cohort BG-PVS volume correlated positively with both systolic and diastolic BP. This pattern is consistent with evidence that higher BP load and vascular stiffness relate to greater BG-PVS burden. For example, higher ambulatory SBP was independently associated with enlarged BG-PVS [38]. A large-sample study suggested cumulative BP exposure as an independent risk factor for enlarged PVS, with BG-PVS partially mediating BP-cognition relationships [39]. Individuals with both T2DM and hypertension show larger PVS volumes, suggesting that metabolic and hemodynamic stressors jointly amplify PVS changes [40]. Taken together, our T2DM-restricted BP-PVS association likely reflects a diabetes-related vulnerability, where perivascular structures are more sensitive to hemodynamic stress.\u003c/p\u003e\n\u003cp\u003eSeveral limitations need to be acknowledged. First, this single-center study had a modest sample, limiting generalizability and statistical power. Longitudinal and multi-center samples should clarify the temporal sequence between metabolic disturbances, perivascular dynamics changes and cognitive impairment in T2DM. Second, the adopted MRI indices are indirect proxies of fluid regulation, their use as markers still warrants validation and interpretation should be cautious. Third, complementary biomarkers (e.g., Aβ/tau, inflammatory markers) should be included to probe the underlying mechanisms of T2DM-related perivascular dysfunction. \u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, patients with T2DM showed lower DTI-ALPS indices than controls even after adjustment for demographic and intracranial covariates, and lower ALPS was related to poorer glycemic profiles. Although overall PVS burden did not differ by group, basal-ganglia PVS volume correlated positively with blood pressure within the T2DM cohort. Together, these findings suggested altered perivascular dynamics in T2DM and highlight glycemic and blood-pressure control as clinically relevant correlates. Because ALPS and PVS are indirect MRI metrics, their causal links with these metabolic profiles still require confirmation in longitudinal and interventional studies.\u003c/p\u003e\n"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe current study was approved by the local Medical Research Ethics Committee of Zhongda Hospital (2024ZDSYLL021-P01). Written informed consent was obtained from all participants prior to their inclusion in the study. All procedures were conducted in accordance with the Declaration of Helsinki and relevant institutional guidelines, ensuring that participation was fully voluntary and that ethical standards were strictly upheld.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll the listed authors have reviewed and approved the final manuscript. There are no conflicts of interest to declare.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research was supported by\u0026nbsp;National Natural Science Foundation of China (82471968 to Y.C., 82427803 to S.J., 82202131 to Y.L. and 82472072 to T.T.), Natural Science Foundation of Jiangsu Province (BK20241688 to Y.C.).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eY.C. conceived of the idea and designed the study. Y.C., Y.W. and and Y.L. drafted and revised the manuscript. Y.C., T.T., L.Z and C.L. analyzed the MR data and performed the statistical analyses. Y.C.W. and S.J. provided critical revisions to the manuscript. S.J. supervised the data collection and analyses. S.J. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.\u003cstrong\u003e\u003cbr\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eSrikanth, V., et al., \u003cem\u003eType 2 diabetes and cognitive dysfunction-towards effective management of both comorbidities.\u003c/em\u003e Lancet Diabetes Endocrinol, 2020. \u003cstrong\u003e8\u003c/strong\u003e(6): p. 535-545.\u003c/li\u003e\n\u003cli\u003eBiessels, G.J. and F. Despa, \u003cem\u003eCognitive decline and dementia in diabetes mellitus: mechanisms and clinical implications.\u003c/em\u003e Nat Rev Endocrinol, 2018. \u003cstrong\u003e14\u003c/strong\u003e(10): p. 591-604.\u003c/li\u003e\n\u003cli\u003eDove, A., et al., \u003cem\u003eThe impact of diabetes on cognitive impairment and its progression to dementia.\u003c/em\u003e Alzheimers Dement, 2021. \u003cstrong\u003e17\u003c/strong\u003e(11): p. 1769-1778.\u003c/li\u003e\n\u003cli\u003eHablitz, L.M. and M. Nedergaard, \u003cem\u003eThe Glymphatic System: A Novel Component of Fundamental Neurobiology.\u003c/em\u003e J Neurosci, 2021. \u003cstrong\u003e41\u003c/strong\u003e(37): p. 7698-7711.\u003c/li\u003e\n\u003cli\u003eMestre, H., Y. Mori, and M. Nedergaard, \u003cem\u003eThe Brain\u0026apos;s Glymphatic System: Current Controversies.\u003c/em\u003e Trends Neurosci, 2020. \u003cstrong\u003e43\u003c/strong\u003e(7): p. 458-466.\u003c/li\u003e\n\u003cli\u003eLian, X., et al., \u003cem\u003eTargeting the glymphatic system to promote alpha-synuclein clearance: a novel therapeutic strategy for Parkinson\u0026apos;s disease.\u003c/em\u003e Neural Regen Res, 2026. \u003cstrong\u003e21\u003c/strong\u003e(1): p. 233-247.\u003c/li\u003e\n\u003cli\u003evan Hattem, T., et al., \u003cem\u003eTargeting Sleep Physiology to Modulate Glymphatic Brain Clearance.\u003c/em\u003e Physiology (Bethesda), 2025. \u003cstrong\u003e40\u003c/strong\u003e(3): p. 0.\u003c/li\u003e\n\u003cli\u003eLee, K.S., et al., \u003cem\u003eHyperglycemia enhances brain susceptibility to lipopolysaccharide-induced neuroinflammation via astrocyte reprogramming.\u003c/em\u003e J Neuroinflammation, 2024. \u003cstrong\u003e21\u003c/strong\u003e(1): p. 137.\u003c/li\u003e\n\u003cli\u003eJiang, Q., et al., \u003cem\u003eImpairment of the glymphatic system after diabetes.\u003c/em\u003e J Cereb Blood Flow Metab, 2017. \u003cstrong\u003e37\u003c/strong\u003e(4): p. 1326-1337.\u003c/li\u003e\n\u003cli\u003eKamagata, K., et al., \u003cem\u003eAssociation of MRI Indices of Glymphatic System With Amyloid Deposition and Cognition in Mild Cognitive Impairment and Alzheimer Disease.\u003c/em\u003e Neurology, 2022. \u003cstrong\u003e99\u003c/strong\u003e(24): p. e2648-e2660.\u003c/li\u003e\n\u003cli\u003eLi, H., et al., \u003cem\u003ePerivascular Spaces, Diffusivity Along Perivascular Spaces, and Free Water in Cerebral Small Vessel Disease.\u003c/em\u003e Neurology, 2024. \u003cstrong\u003e102\u003c/strong\u003e(9): p. e209306.\u003c/li\u003e\n\u003cli\u003eTaoka, T., et al., \u003cem\u003eEvaluation of glymphatic system activity with the diffusion MR technique: diffusion tensor image analysis along the perivascular space (DTI-ALPS) in Alzheimer\u0026apos;s disease cases.\u003c/em\u003e Jpn J Radiol, 2017. \u003cstrong\u003e35\u003c/strong\u003e(4): p. 172-178.\u003c/li\u003e\n\u003cli\u003eLi, S., et al., \u003cem\u003eMicrostructural Bias in the Assessment of Periventricular Flow as Revealed in Postmortem Brains.\u003c/em\u003e Radiology, 2025. \u003cstrong\u003e316\u003c/strong\u003e(3): p. e250753.\u003c/li\u003e\n\u003cli\u003ePasternak, O., et al., \u003cem\u003eFree water elimination and mapping from diffusion MRI.\u003c/em\u003e Magn Reson Med, 2009. \u003cstrong\u003e62\u003c/strong\u003e(3): p. 717-30.\u003c/li\u003e\n\u003cli\u003eWardlaw, J.M., et al., \u003cem\u003ePerivascular spaces in the brain: anatomy, physiology and pathology.\u003c/em\u003e Nat Rev Neurol, 2020. \u003cstrong\u003e16\u003c/strong\u003e(3): p. 137-153.\u003c/li\u003e\n\u003cli\u003eAmerican Diabetes Association Professional Practice, C., \u003cem\u003e2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes-2024.\u003c/em\u003e Diabetes Care, 2024. \u003cstrong\u003e47\u003c/strong\u003e(Suppl 1): p. S20-S42.\u003c/li\u003e\n\u003cli\u003eWallace, T.M., J.C. Levy, and D.R. Matthews, \u003cem\u003eUse and abuse of HOMA modeling.\u003c/em\u003e Diabetes Care, 2004. \u003cstrong\u003e27\u003c/strong\u003e(6): p. 1487-95.\u003c/li\u003e\n\u003cli\u003eSepehrband, F., et al., \u003cem\u003eImage processing approaches to enhance perivascular space visibility and quantification using MRI.\u003c/em\u003e Sci Rep, 2019. \u003cstrong\u003e9\u003c/strong\u003e(1): p. 12351.\u003c/li\u003e\n\u003cli\u003eFrangi, A.F., et al. \u003cem\u003eMultiscale vessel enhancement filtering\u003c/em\u003e. 1998. Berlin, Heidelberg: Springer Berlin Heidelberg.\u003c/li\u003e\n\u003cli\u003eTuerxun, R., et al., \u003cem\u003eAssessing interstitial fluid dynamics in type 2 diabetes mellitus and prediabetes cases through diffusion tensor imaging analysis along the perivascular space.\u003c/em\u003e Front Aging Neurosci, 2024. \u003cstrong\u003e16\u003c/strong\u003e: p. 1362457.\u003c/li\u003e\n\u003cli\u003eYu, S., et al., \u003cem\u003eDTI-ALPS index decreased in patients with Type 2 Diabetes Mellitus.\u003c/em\u003e Front Neurosci, 2024. \u003cstrong\u003e18\u003c/strong\u003e: p. 1383780.\u003c/li\u003e\n\u003cli\u003eHu, P., et al., \u003cem\u003eAssociation of diffusion tensor imaging along the perivascular space index with cognitive impairment in type 2 diabetes mellitus.\u003c/em\u003e Quant Imaging Med Surg, 2025. \u003cstrong\u003e15\u003c/strong\u003e(2): p. 1491-1504.\u003c/li\u003e\n\u003cli\u003eKim, Y.K., K.I. Nam, and J. Song, \u003cem\u003eThe Glymphatic System in Diabetes-Induced Dementia.\u003c/em\u003e Front Neurol, 2018. \u003cstrong\u003e9\u003c/strong\u003e: p. 867.\u003c/li\u003e\n\u003cli\u003eBoyd, E.D., et al., \u003cem\u003eThe Glymphatic Response to the Development of Type 2 Diabetes.\u003c/em\u003e Biomedicines, 2024. \u003cstrong\u003e12\u003c/strong\u003e(2).\u003c/li\u003e\n\u003cli\u003eZhang, Y., et al., \u003cem\u003eThe Influence of Demographics and Vascular Risk Factors on Glymphatic Function Measured by Diffusion Along Perivascular Space.\u003c/em\u003e Front Aging Neurosci, 2021. \u003cstrong\u003e13\u003c/strong\u003e: p. 693787.\u003c/li\u003e\n\u003cli\u003eOzsahin, I., et al., \u003cem\u003eDiffusion Tensor Imaging Along Perivascular Spaces (DTI-ALPS) to Assess Effects of Age, Sex, and Head Size on Interstitial Fluid Dynamics in Healthy Subjects.\u003c/em\u003e J Alzheimers Dis Rep, 2024. \u003cstrong\u003e8\u003c/strong\u003e(1): p. 355-361.\u003c/li\u003e\n\u003cli\u003eAntal, B., et al., \u003cem\u003eType 2 diabetes mellitus accelerates brain aging and cognitive decline: Complementary findings from UK Biobank and meta-analyses.\u003c/em\u003e Elife, 2022. \u003cstrong\u003e11\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eZhang, Y., et al., \u003cem\u003eImpaired glymphatic function in relation to cumulative blood glucose exposure: A population-based cohort study.\u003c/em\u003e IBRO Neurosci Rep, 2025. \u003cstrong\u003e19\u003c/strong\u003e: p. 437-444.\u003c/li\u003e\n\u003cli\u003eWang, N., et al., \u003cem\u003eEvaluation of the glymphatic system using the DTI-ALPS index in type 2 diabetes mellitus-induced cognitive impairment.\u003c/em\u003e J Clin Imaging Sci, 2025. \u003cstrong\u003e15\u003c/strong\u003e: p. 31.\u003c/li\u003e\n\u003cli\u003eLashkari, A., \u003cem\u003eThe effects of diabetes on the glymphatic system: recent advances and mechanistic insights.\u003c/em\u003e Cardiovasc Diabetol Endocrinol Rep, 2025. \u003cstrong\u003e11\u003c/strong\u003e(1): p. 14.\u003c/li\u003e\n\u003cli\u003eShin, J., et al., \u003cem\u003ePrediabetic HbA1c and Cortical Atrophy: Underlying Neurobiology.\u003c/em\u003e Diabetes Care, 2023. \u003cstrong\u003e46\u003c/strong\u003e(12): p. 2267-2272.\u003c/li\u003e\n\u003cli\u003eCrane, P.K., R. Walker, and E.B. Larson, \u003cem\u003eGlucose levels and risk of dementia.\u003c/em\u003e N Engl J Med, 2013. \u003cstrong\u003e369\u003c/strong\u003e(19): p. 1863-4.\u003c/li\u003e\n\u003cli\u003eMaillard, P., et al., \u003cem\u003eCerebral white matter free water: A sensitive biomarker of cognition and function.\u003c/em\u003e Neurology, 2019. \u003cstrong\u003e92\u003c/strong\u003e(19): p. e2221-e2231.\u003c/li\u003e\n\u003cli\u003eBerger, M., et al., \u003cem\u003eFree water diffusion MRI and executive function with a speed component in healthy aging.\u003c/em\u003e Neuroimage, 2022. \u003cstrong\u003e257\u003c/strong\u003e: p. 119303.\u003c/li\u003e\n\u003cli\u003eZhu, Z., et al., \u003cem\u003eWhite Matter Free Water Outperforms Cerebral Small Vessel Disease Total Score in Predicting Cognitive Decline in Persons with Mild Cognitive Impairment.\u003c/em\u003e J Alzheimers Dis, 2022. \u003cstrong\u003e86\u003c/strong\u003e(2): p. 741-751.\u003c/li\u003e\n\u003cli\u003eLiu, X., et al., \u003cem\u003eMRI free water mediates the association between diffusion tensor image analysis along the perivascular space and executive function in four independent middle to aged cohorts.\u003c/em\u003e Alzheimers Dement, 2025. \u003cstrong\u003e21\u003c/strong\u003e(2): p. e14453.\u003c/li\u003e\n\u003cli\u003eToro, A.R., et al., \u003cem\u003eAssociation of MRI visible Perivascular Spaces with Early White Matter Injury.\u003c/em\u003e AJNR Am J Neuroradiol, 2025.\u003c/li\u003e\n\u003cli\u003eYang, S., et al., \u003cem\u003eHigher ambulatory systolic blood pressure independently associated with enlarged perivascular spaces in basal ganglia.\u003c/em\u003e Neurol Res, 2017. \u003cstrong\u003e39\u003c/strong\u003e(9): p. 787-794.\u003c/li\u003e\n\u003cli\u003eShi, H., et al., \u003cem\u003eEnlarged Perivascular Spaces in Relation to Cumulative Blood Pressure Exposure and Cognitive Impairment.\u003c/em\u003e Hypertension, 2023. \u003cstrong\u003e80\u003c/strong\u003e(10): p. 2088-2098.\u003c/li\u003e\n\u003cli\u003eZebarth, J., et al., \u003cem\u003ePerivascular spaces mediate a relationship between diabetes and other cerebral small vessel disease markers in cerebrovascular and neurodegenerative diseases.\u003c/em\u003e J Stroke Cerebrovasc Dis, 2023. \u003cstrong\u003e32\u003c/strong\u003e(9): p. 107273.\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table 1","content":"\u003cp\u003e\u003cstrong\u003eTable 1. Demographics and clinical characteristics for T2DM and control groups\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cdiv\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eMeasures\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHC (n=34)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eT2DM (n=34)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003ep\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003e\u0026nbsp;Value\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e57.32\u0026plusmn; 6.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e59.87 \u0026plusmn; 7.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.11\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eSex (male/female)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e11/23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e18/16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.09\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eYears of education\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e10.16 \u0026plusmn; 2.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e9.76 \u0026plusmn; 3.69\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.58\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eDiabetes duration (years)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e/\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e8.95 \u0026plusmn; 4.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e/\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eInsulin treatment (n)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e/\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e/\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eBMI (kg/m2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e24.10 \u0026plusmn; 2.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e24.66 \u0026plusmn; 3.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.40\u003cstrong\u003e\u003csup\u003ec\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eFPG (mmol/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e5.53 \u0026plusmn; 0.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e8.03 \u0026plusmn; 2.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.01\u003csup\u003ec\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003e2h-PPG (mmol/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e6.18 \u0026plusmn; 1.46\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e14.76 \u0026plusmn; 4.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.01\u003c/strong\u003e\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eHbA1c (%, mmol/mol)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e5.61 \u0026plusmn; 0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e7.93 \u0026plusmn; 1.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.01\u003c/strong\u003e\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eHOMA-IR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e2.59 \u0026plusmn; 1.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e3.37 \u0026plusmn; 2.03\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.03\u003csup\u003ea\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eTriglyceride (mmol/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e1.27\u0026plusmn; 0.72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e1.34 \u0026plusmn; 0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.65\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eCholesterol (mmol/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e5.19 \u0026plusmn; 0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e5.36 \u0026plusmn; 1.12\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.46\u003cstrong\u003e\u003csup\u003ec\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eLDL (mmol/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e3.06 \u0026plusmn; 0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e3.25 \u0026plusmn; 0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.25\u003cstrong\u003e\u003csup\u003ec\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eHDL (mmol/L)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e1.34 \u0026plusmn; 0.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e1.36 \u0026plusmn; 0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.76\u003cstrong\u003e\u003csup\u003ec\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eSBP (mmHg)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e133.42 \u0026plusmn; 14.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e136.43 \u0026plusmn; 13.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.36\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eDBP (mmHg)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e87.42 \u0026plusmn; 10.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e85.11 \u0026plusmn; 11.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.37\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eHypertension (n)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.31\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eHyperlipidemia (n)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e13\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e14\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.75\u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eTIV (cm\u003csup\u003e3\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e1444.3\u0026plusmn;130.9\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e1473.3\u0026plusmn;126.5\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.33\u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eWMHVF (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e0.46\u0026nbsp;\u0026plusmn;\u0026nbsp;0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e0.48\u0026nbsp;\u0026plusmn;\u0026nbsp;0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.40\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eGMVF (%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e41.0\u0026nbsp;\u0026plusmn;\u0026nbsp;1.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e41.3\u0026nbsp;\u0026plusmn;\u0026nbsp;1.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.33\u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eFazekas-PVWMH (range)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e1-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e1-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.65\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eFazekas-DWMH (range)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e0-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e0-2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.08\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCognitive performance\u003csup\u003e*\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eMMSE\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e28.68\u0026plusmn;1.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e28.34\u0026plusmn;1.28\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.23\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eAVLT-immediate\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e6.72 \u0026plusmn; 2.06\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e6.12 \u0026plusmn; 1.41\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.08\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eAVLT-5min recall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e6.97\u0026plusmn; 1.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e6.81 \u0026plusmn; 2.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eAVLT-20min recall\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e6.53 \u0026plusmn; 2.15\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e5.92 \u0026plusmn; 2.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.27\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eDST-forward\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e7.42 \u0026plusmn; 1.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e6.94 \u0026plusmn; 1.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eDST-backward\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e4.55 \u0026plusmn; 1.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e4.03 \u0026plusmn; 0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.03\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eCDT\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e3.43 \u0026plusmn; 0.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e3.20 \u0026plusmn; 0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.10\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eCFT-copy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e34.78 \u0026plusmn; 1.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e34.24 \u0026plusmn; 2.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.42\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eCFT-delay\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e17.63 \u0026plusmn; 6.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e14.21 \u0026plusmn; 4.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.01*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eTime in TMT-A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e62.11 \u0026plusmn; 13.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e68.38 \u0026plusmn; 23.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 33.9789%;\"\u003e\n \u003cp\u003eTime in TMT-B\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.6479%;\"\u003e\n \u003cp\u003e144.11 \u0026plusmn; 50.43\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 24.2958%;\"\u003e\n \u003cp\u003e177.08 \u0026plusmn; 55.10\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 17.0775%;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.01*\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eData are presented as mean\u0026nbsp;\u0026plusmn;\u0026nbsp;standard deviation or the number of patients unless otherwise stated. \u003csup\u003e*\u003c/sup\u003eComparison was adjusted for age, sex and education level. \u003csup\u003ea\u003c/sup\u003e\u003cem\u003eP\u003c/em\u003e values for group differences obtained using Mann-Whitney U test; \u003csup\u003eb\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e values obtained using Chi-squared test; \u003csup\u003ec\u003c/sup\u003e\u003cem\u003ep\u003c/em\u003e values obtained using two-sample t-test. Bold means \u003cem\u003ep\u003c/em\u003e value less than 0.05.\u003c/p\u003e\n\u003cp\u003eAbbreviations: BMI, body mass index; FPG, fasting plasma glucose;\u0026nbsp;2h-PPG, 2-hour postprandial glucose;\u0026nbsp;HbA1c, Hemoglobin-A1c; HOMA-IR, Homeostatic Model Assessment for Insulin Resistance; LDL, Low-density lipoprotein; HDL, High-density lipoprotein; SBP, systolic blood pressure; DBP, diastolic blood pressure; TIV, total intracranial volume;\u0026nbsp;WMHVF, WMH volume fraction; GMVF, GM volume fraction;\u0026nbsp;MoCA, The Montreal Cognitive Assessment; AVLT, Auditory Verbal Learning Test; DST, digit span test; CDT, clock drawing test; CFT, complex figure test; TMT, trail-making test.\u003c/p\u003e\n"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Type 2 diabetes, insulin resistance, cognitive impairment, MRI Neuroimaging, glymphatic system","lastPublishedDoi":"10.21203/rs.3.rs-7859506/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7859506/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackgrounds: \u003c/strong\u003ePerivascular fluid dynamics support brain-wide solute transport and have been described to be perturbed in type 2 diabetes mellitus (T2DM) models. In this study, we aimed to evaluate T2DM-related alterations in MRI perivascular markers and their associations with glycemic control and cognitive performances.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eThirty-four T2DM patients and 34 age- and sex-matched controls underwent multimodal MRI to quantify perivascular metrics: the diffusion tensor imaging analysis along the perivascular space (DTI-ALPS) index reflects perivenous diffusion anisotropy, white matter free water (WM-FW) indicates CSF-interstitial fluid exchange, and perivascular space volume fraction (PVSVF) indexes perivascular conduits burden. Cognitive function and metabolic indicators were evaluated. Group comparisons and associations were tested with multivariable linear regression adjusting for key demographic and intracranial covariates.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eCompared with controls, T2DM showed a lower ALPS index (adjusted β = -0.091, 95% CI -0.155 to -0.027, \u003cem\u003ep\u003c/em\u003e = 0.006; FDR-adjusted \u003cem\u003ep\u003c/em\u003e = 0.025). The group differences were more prominent among male and older patients, although formal interactions were not significant. WM-FW and PVSVF showed no group difference after adjusting covariates. Within T2DM, lower ALPS was associated with higher postprandial glucose (partial \u003cem\u003er\u003c/em\u003e = -0.46) and HbA1c (partial \u003cem\u003er\u003c/em\u003e = -0.37) (all \u003cem\u003ep\u003c/em\u003e ≤0.05, FDR-adjusted \u003cem\u003ep\u003c/em\u003e = 0.11). Multivariable linear regression including all MRI markers showed that ALPS was the only imaging metric that remained independently associated with glycemic measures.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusions: \u003c/strong\u003eT2DM is associated with altered perivascular diffusion anisotropy on MRI, indicating perivascular dysfunction. Associations with glycemic profile suggest these modifiable factors may influence perivascular health in T2DM and merit prospective evaluation.\u003c/p\u003e","manuscriptTitle":"MRI Markers of Perivascular Fluid Dynamics in Type 2 Diabetes and Their Associations with Glycemic Control","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-25 07:17:17","doi":"10.21203/rs.3.rs-7859506/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"7a32e371-0590-4604-bd8c-db03a3b32bba","owner":[],"postedDate":"November 25th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-08T07:42:55+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-25 07:17:17","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7859506","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7859506","identity":"rs-7859506","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}