Ultrasound viscoelasticity and attenuation imaging for rheumatoid arthritis-associated muscle quality changes: a...

Rheumatoid arthritis (RA) is associated with loss of muscle quality and strength, but accessible quantitative tools for detecting these changes remain limited. We evaluated whether shear wave viscoelastic imaging (SWVI), acoustic attenuation imaging (AAI), and isokinetic strength testing could characterize rectus femoris muscle involvement in RA, and used public synovial transcriptomic data to provide hypothesis-generating biological context.

Twenty-nine patients with RA (early RA, n = 7; active RA, n = 11; treated low/moderate-activity RA, n = 11) and 27 healthy controls underwent rectus femoris ultrasound and knee isokinetic strength testing. The main ultrasound variables were rectus femoris thickness and cross-sectional area, mean Young's modulus (Emean), viscosity (Vimean), speed of sound (SoS), and AAI. Group differences, Spearman correlations, receiver-operating-characteristic (ROC) curves, and exploratory logistic models were analyzed. Because the RA cohort was small and diabetes was more common in RA than in controls, ROC, nomogram, and multivariable results were interpreted as exploratory rather than confirmatory. Public GSE55235 synovial transcriptome data were reanalyzed to identify inflammatory and extracellular-matrix pathways that may plausibly link synovitis with systemic muscle impairment.

Rectus femoris thickness and cross-sectional area showed limited between-group separation, whereas Emean, Vimean, SoS, AAI, and isokinetic strength differed across groups (all p < 0.001). AAI increased from controls to early, active, and treated low/moderate-activity RA groups and correlated with disease duration in the full cohort (Spearman rho = 0.864); this association remained strong in RA-only adjusted sensitivity analyses. SoS decreased across the same gradient (rho =  - 0.875 with duration in the full cohort), whereas Vimean was highest in active RA and correlated with DAS28 in the full cohort (rho = 0.643). Extensor peak torque was lower in RA and correlated inversely with DAS28 in the full cohort (rho =  - 0.844); the corresponding RA-only association was more moderate but remained significant (rho =  - 0.642). In exploratory ROC analyses, AAI, SoS, and extensor peak torque showed high apparent discrimination between RA and controls; however, the small overall and subgroup samples, baseline differences in diabetes and height, and absence of external validation make these estimates potentially unstable and preclude conclusions about clinical applicability. Synovial transcriptomic analysis showed enrichment of TNF, NF-κB, JAK-STAT, cytokine-receptor, and extracellular-matrix remodeling pathways in RA synovium, supporting a plausible inflammatory background rather than direct evidence of muscle pathology.

In this exploratory clinical-transcriptomic study, SWVI/AAI and isokinetic strength testing detected rectus femoris muscle-quality and functional differences associated with RA. AAI and SoS may be candidate ultrasound parameters for RA-associated muscle involvement, but their disease specificity, independence from diabetes, and tissue-level interpretation require confirmation in larger, externally validated studies with reference-standard muscle-composition assessment and reliability testing.

• This study integrates meta-analysis, Mendelian randomization, and single-cell RNA sequencing to comprehensively evaluate treatment efficacy, genetic influences, and immune alterations in late-onset rheumatoid arthritis (LORA). • LORA patients exhibit distinct treatment responses compared with younger-onset RA, including lower clinical remission rates with biologics/tsDMARDs and higher residual disease activity (DAS28). • Genetic variants in the IL-6R and TYK2 pathways are identified as key modifiers of drug response and disease susceptibility in LORA. • The findings underscore the need for age- and genetics-informed personalized treatment strategies in elderly patients with rheumatoid arthritis.