AUTHOR=Liu Yong-Lin , Yin Han-Peng , Wang Feng , Bao Ting-Yu , Yao Min-Yi , Yuan Yi-Mian , Liang Man-Qiu , Wang Fang , Chen Yang-Kun , Huang Li-An 

TITLE=Deep medullary veins disruption in cerebral small vessel disease: links to AI-quantified lesions and cognitive decline

JOURNAL=Frontiers in Neurology

VOLUME=Volume 16 - 2025

YEAR=2025

URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2025.1647684

DOI=10.3389/fneur.2025.1647684

ISSN=1664-2295

ABSTRACT=BackgroundThe relationship between cerebral small vessel diseases (CSVD) and deep medullary veins (DMV) has been the subject of only a limited number of studies to date, with the majority of research focusing on the correlation between DMV and cognitive function in patients with CSVD. The present study aims to quantitatively evaluate the relationship between DMV and CSVD imaging markers, utilizing an artificial intelligence neuroimaging system to achieve this objective. Furthermore, an investigation was conducted into the correlation between DMV disruption and cognitive impairment.MethodsThis cross-sectional study enrolled patients with CSVD imaging markers who were admitted to the Department of Neurology at our hospital. The quantitative assessment of CSVD imaging markers, encompassing white matter hyperintensities (WMH), lacunar infarction (LI), and cerebral microbleeds (CMB), was conducted employing the AccuBrain® neuroimaging system. The DMV was evaluated using susceptibility-weighted imaging (SWI) with a semi-quantitative scale. The Montreal Cognitive Assessment (MoCA) score was evaluated for each patient.ResultsThe study’s sample population comprised 171 patients. Multivariate ordinal logistic regressions indicated that DMV score was significantly associated with WMH volume, LI count, and CMB count (p < 0.05). In the context of multivariate linear regression analysis, a significant negative correlation was observed between the DMV score and the MoCA score, with the latter being adjusted for confounding variables such as age, education, gender, and smoking history (p = 0.003). The application of path analysis revealed a significant correlation between the DMV score and the MoCA score, which indicates that WMH volume plays a mediating role in this relationship, thereby offering a novel perspective on cognitive function and neurodegenerative processes.ConclusionDMV score is associated with the severity of WMH, CMB, and LI, as well as cognitive performance decline, implicating that cerebral venule damage may play a role in the development of CSVD and related cognitive impairment.