Background/Purpose: To compare corpus callosum (CC) volume and diffusion tensor imaging in systemic lupus erythematosus according to age of disease-onset. 

Methods: We selected 75 patients with childhood-onset (c)SLE [mean age of 24.6 years (SD 4.6) and disease duration of 11.6 years (SD 4.8)] and two control groups: 1. Matched by disease duration, consisting of 51 patients with adult-onset (a)SLE [mean age of 32.4 (SD 3.2)] and disease duration of 11.5 years (SD 4.2)] and 2. Matched by age consisting of 77 healthy controls (HC) [mean age of 28 years (DP 3.6)]. A complete clinical, laboratory and neurological evaluation was performed in all subjects. Cognitive impairment were screened with Montreal Cognitive Assessment (MoCA), mood and anxiety disorders were determined through Beck Inventory. SLE patients were further assessed for disease activity [SLE Disease Activity Index (SLEDAI)], damage [Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI)] and current and cumulative drug exposures. All subjects underwent MRI examination with a Philips 3T scanner. Sagittal T1‐weighted images were used for CC segmentation using FreeSurfer. Diffusion Weighted Images (DWI) were analyzed in FSL tool and scalar maps of the Fractional Anisotropy (FA), Mean Diffusivity (MD), Axial Diffusivity (AD) and Radial Diffusivity (RD) were calculated. The CC was parcellated based on DTI. Using the scalar maps we obtained the average and standard deviation of FA, MD, AD and RD for the segmented CC and each parcel. P values < 0.05 were considered statistically significant.

Results: We observed similar CC volume in cSLE [mean volume = 6410.19mm³ (SD 2026.85)] when compared to aSLE [mean volume = 6658.61mm3 (SD = 2121.50)]; p=0.465] and HC [mean volume=7216.40mm3 (DP 1991.61)]; p=0.175]. When CC was divided into different portions, we observed significant smaller volumes in the mid-anterior region, the central portion, the mid-posterior portion (p˂0.001) in cSLE and HC. No significant difference was observed between CC parcels in cSLE and aSLE and between aSLE and HC. We observed that FA values were significantly lower cSLE (0.63603; SD±0.0665) when compared with the aSLE (0.677624 SD ± 0.08920; p=0.002) and HC (0.699624;SD±0.07920; p=0.001). Increased MD (0.00096,SD±0.00014), RD (0.00057, SD±0.00015) and AD  (0.0017, SD± 0.00014) were observed in cSLE when compared to aSLE [(MD:0.00096, SD ± 0.00014; p = 0.002); (RD:0.00048, SD ± 0.00019; p = 0.002);  (AD:0.0016, SD±0.00011; p=0.024)] and HC [(MD:0.00116, SD±0.00024; p< 0.001); (RD:0.00052, SD± 0.00023; p< 0.001);  (AD:0.0020, SD±0.00018; p=0.014)]. MoCA scores were significantly lower in cSLE [20.82(SD±4.6)] when compared to aSLE [23.27(SD±4.4); p=0.005] and HC [26.52(SD±2.6); p< 0.001]. No significant differences between BDI and BAI scores were observed (p >0.05). When comparing MoCA scores with diffusion parameters we observed a correlation with FA (r=0.65; p=0.03), MD (r=0.7; p=0.03) and RD (r=0.647; p=0.04).

Conclusion: We observed more frequently microstructural changes in the corpus callosum in adults with cSLE when compared to aSLE with similar disease duration and HC. Microstructural changes can explain worse neurocognitive outcome in cSLE.  

« Back to ACR Convergence 2020

ACR Meeting Abstracts - https://acrabstracts.org/abstract/microstructural-damage-is-associated-with-age-at-disease-onset-and-cognitive-impairment-in-systemic-lupus-erythematosus/