Radiographic joint inflammation in rheumatoid arthritis (RA) can persist or progress despite clinical remission[1], suggesting that local joint inflammation is not fully reflected by systemic inflammatory markers. Adipose tissue is increasingly recognized as an active immunometabolic organ, and the infrapatellar fat pad (IPFP), an intracapsular adipose structure adjacent to the synovium[2], may contribute to local joint inflammation in RA[3]. While Magnetic resonance imaging (MRI) allows non-invasive quantification of fat fraction[4], the relationship between MRI-derived IPFP fat fraction and local joint inflammation in RA remains unclear. This study aimed to quantify the fat fraction of the IPFP in RA patients using MRI and to explore its association with joint inflammation.

Fifty patients with RA presenting with knee symptoms were recruited from the Third Affiliated Hospital of Sun Yat-sen University between July 2022 and December 2023 to undergo contrast-enhanced 3.0 T MRI scans. Three consecutive sagittal slices in the central portion of the IPFP were selected as regions of interest (ROIs) on fat fraction (FF) maps. FF values (%) within the ROIs were quantified using the Iterative Decomposition of Water and Fat with Echo Asymmetry and Least-Squares Estimation (IDEAL-IQ) sequence (Figures 1a and 1c). The RA MRI scoring system (OMERACT 2016 RAMRIS) was used to evaluate knee joint pathologies, including inflammation (synovitis, bone marrow edema) and structural impairment (joint space, bone erosion) (Figures 1b and 1d). Differences in focal inflammation among subgroups were assessed using the Kruskal–Wallis test, and associations between IPFP fat fraction and systemic inflammation, joint structural damage, and disease activity were evaluated using Spearman’s rank correlation coefficients. All MRI parameter evaluations were manually conducted using the ITK-SNAP 3.5 workstation.

The results showed that FF values of IPFP measured by MRI differed across synovitis and bone marrow edema grades, with higher FF observed in knees with severe synovitis and moderate-to–severe bone marrow edema (P < 0.01 for both; Figures 2A and 2B). FF was not correlated with structural joint damage, including bone erosion and joint space width. Additionally, FF showed no correlation with systemic inflammatory markers including ESR, CRP, or disease activity score (DAS28).

MRI-measured FF of the IPFP is associated with local knee inflammation in RA, but not with structural damage or systemic inflammation, and may be involved in local inflammatory processes.