MRI-based study of gray matter volume and cortical morphological alterations in patients with rheumatoid arthritis

To investigate alterations in gray matter volume (GMV) and cortical morphology in patients with rheumatoid arthritis (RA) and examine their associations with clinical characteristics using voxel-based morphometry (VBM) and surface-based morphometry (SBM).

41 RA patients and 34 age- and sex-matched healthy controls (HCs) underwent high-resolution structural magnetic resonance imaging (sMRI). VBM was used to compare GMV, and SBM was applied to assess gyrification index (GI), cortical thickness (CT), sulcus depth (SD), and fractal dimension (FD). Whole-brain voxel/vertex-wise multiple linear regression analyses evaluated associations between structural indices and DAS28 disease activity scores, Hamilton Depression Scale (HAMD), Hamilton Anxiety Scale (HAMA), Montreal Cognitive Assessment (MoCA), C-reactive protein (CRP), Visual Analogue Scale (VAS), swollen joint count, and tender joint count.

No significant group diferences were found in sex (χ²= 0.98, P = 0.32) and age (t = 0.48, P = 0.14). Compared with HCs, RA patients showed reduced GMV in the bilateral putamen (P_voxel = 0.001, P_cluster = 0.05; cluster size > 50), decreased SD in the left hemisphere (118 vertices: 79% left inferior frontal gyrus, opercular part; 16% lateral orbital frontal gyrus; 5% superior frontal gyrus; P = 0.0003), decreased GI in the right superior and inferior parietal gyri (65 vertices; P = 0.0004), and reduced CT in the left inferior temporal gyrus (93 vertices; P < 0.0001). Several structural indices were correlated with clinical variables. In the RA group, HAMA scores correlated positively with CT in the left superior frontal gyrus and both inferior and superior parts of the right middle frontal gyrus (P = 0.0001). VAS scores correlated positively with CT in the left fusiform gyrus and entorhinal cortex (P = 0.0001). Swollen joint counts correlated positively with CT in the inferior left middle frontal gyrus (P = 0.0001) and right inferior parietal lobule (P = 0.0004), positively with FD in the left precuneus (P < 0.0001), and negatively with FD in the right inferior parietal gyrus (P = 0.0001). DAS28 (CRP) correlated positively with GI and negatively with SD in the opercular part of the left inferior frontal gyrus and the right superior temporal gyrus (P = 0.0001). HAMD scores correlated negatively with GI and SD in the left inferior parietal gyrus, right supramarginal gyrus, and right precentral gyrus (GI: P = 0.0002, P < 0.0001, P = 0.0001; SD: P = 0.0002, P < 0.0001, P = 0.0003). CRP correlated positively with FD in the right postcentral gyrus and right superior parietal gyrus (P < 0.0001). MoCA scores correlated negatively with FD in the left superior frontal gyrus (P = 0.0001). Tender joint counts correlated negatively with SD in the left inferior parietal gyrus (P < 0.0001) and with FD in the opercular part of the left inferior frontal gyrus (P < 0.0001).

Through multi-parameter joint analysis, this study systematically depicted the multi-dimensional changes in brain structure of patients with RA for the first time and initially established a framework for their association with disease activity, emotional and cognitive impairment, and inflammatory levels.