Correlation between perivascular spaces and clinical features in patients with rheumatoid arthritis

This study aims to compare perivascular space (PVS) parameters across different rheumatoid factor (RF) titer subgroups within a rheumatoid arthritis (RA) cohort, and to explore their correlations with serological markers.

In this cross-sectional study, 160 RA patients were enrolled and categorized into three groups based on RF titer: the negative group (n = 35), the low-titer group (n = 26), and the high-titer group (n = 99). Demographic, clinical, and laboratory data were systematically collected. PVS parameters across various brain regions were quantitatively assessed using 3.0 T magnetic resonance imaging (MRI). Spearman correlation analysis was employed to explore associations, and multiple linear regression was conducted for multivariate analysis.

Statistically significant differences were identified among the three groups concerning the number of PVS in the right frontal lobe (p = 0.022), the number of PVS in the right parietal lobe (p = 0.032), the volume of PVS in the right parietal lobe (p = 0.004), and the number of PVS in the pons (p = 0.006). The high-titer group exhibited significantly elevated values of the frontal and parietal PVS parameters compared to the negative group (all p < 0.05). Correlation analysis demonstrated that RF-IgA showed weak-to-moderate but statistically significant positive correlations with the number of PVS in the right frontal lobe (R = 0.181), the number of PVS in the right parietal lobe (R = 0.193), and the volume of PVS in the right parietal lobe (R = 0.162) (all p < 0.05). Additionally, the anti-cyclic citrullinated peptide III antibody (Anti-CCPIII) was positively correlated with the number of PVS in the right parietal lobe (R = 0.196), while the anti-mutated citrullinated vimentin antibody (MCV) was positively correlated with the number of PVS in the pons (R = 0.168) (all p < 0.05). Multiple regression analysis revealed that the erythrocyte sedimentation rate (ESR) was independently negatively associated with the number of PVS in the right frontal lobe (β = - 0.142, p = 0.022). In conclusion, autoantibody titers and inflammatory levels in RA patients are linked to PVS enlargement in specific brain regions. These intra-group differences suggest that PVS burden varies with autoimmune load in RA patients, highlighting its potential as a surrogate imaging marker for immune-mediated cerebral microvascular involvement within the RA population.

This study demonstrates that distinct autoantibody profiles in RA correlate with region-specific enlargement of PVS. Perivascular space abnormalities may potentially serve as a non-invasive imaging marker to reflect immune-mediated cerebral small vessel injury. Longitudinal multicenter trials with cognitive assessments are required to further validate the association and explore the therapeutic effects of immune regulation on brain microvascular lesions. Key Points • This study quantitatively analyzed PVS via 3.0 T MRI in 160 RA patients grouped by rheumatoid factor titers. • High RF titer was significantly correlated with increased PVS number and volume in the right frontal and parietal lobes, while pontine PVS peaked in the low-titer group. • RA-related autoantibodies including RF-IgA, anti-CCPIII and MCV showed positive correlations with region-specific PVS enlargement, independent of traditional vascular risk factors. • PVS may be considered a potential non-invasive imaging biomarker for early identification and risk stratification of RA-associated cerebral small vessel injury.