Free water, perivascular spaces, and diffusivity along perivascular spaces dynamics after recent small subcortical
Tiantian Qiu1,2, Hui Hong3, Xiaopei Xu3
1Department of Radiology, Linyi People's Hospital, Linyi, China.
Background And Purpose:
The pathophysiological mechanisms underlying cognitive changes following recent small subcortical infarcts (RSSIs) remain incompletely understood. In this study, we aimed to investigate alterations in brain fluid dynamics-including interstitial fluid (ISF) and perivascular space (PVS) abnormalities-using magnetic resonance imaging (MRI)-based markers, and examined their associations with cognitive outcomes.
Methods:
In this prospective single-center study, patients with RSSIs underwent comprehensive neuropsychological evaluations and multimodal MRI assessments at 3-7 days, 3-6 months, and 1, 2, and 3 years post-stroke. Three MRI-derived proxies of brain fluid dynamics were assessed: free water (FW) fraction, reflecting the ISF fraction; and PVS volume and diffusion tensor imaging analysis along the PVS (DTI-ALPS), both reflecting PVS dynamics. Relative metrics (FW, PVS, and DTI-ALPS) were calculated as (ipsilesional - contralesional)/contralesional values, to quantify hemisphere-specific changes associated with RSSI. Comparative analyses of MRI-derived proxies of brain fluid dynamics were performed between ipsilesional and contralesional hemispheres both cross-sectionally and longitudinally. The associations between these markers and cognitive performance were examined using linear regression and mixed-effects models, with false discovery rate (FDR) corrections for multiple comparisons.
Results:
The study included 66 RSSI patients (mean age: 57.15 ± 7.35 years; 80.3% male). Baseline comparisons revealed significantly higher PVS volume (p = 0.004) and lower DTI-ALPS (p = 0.018) in the ipsilesional hemisphere compared with the contralesional side. While FW did not differ significantly between hemispheres (p = 0.858), the perilesional regions showed significantly higher FW compared with the corresponding contralesional regions (p < 0.05). Furthermore, baseline lower relative PVS was associated with attention improvement (β = 0.513, PFDR = 0.004). Longitudinal analysis revealed bilateral FW increase (p = 0.025) and DTI-ALPS decline (p = 0.016), with no significant interhemispheric differences. However, no significant correlations were observed between relative FW, PVS, and DTI-ALPS slopes and cognitive trajectories (all PFDR > 0.05).
Conclusion:
Our findings demonstrate RSSIs are associated with disrupted MRI-derived proxies of brain fluid dynamics, characterized by increased ISF and PVS dysfunction in the ipsilesional hemisphere. Baseline PVS volume is correlated with cognitive changes in RSSI patients, highlighting the potential of early PVS-targeted interventions to prevent post-stroke cognitive impairment.
