Background and Objectives Hallmarks of rheumatoid arthritis (RA) are joint synovitis and subchondral bone erosions, resulting in loss of joint function. However, occurrence and kinetics of periarticular bone loss remain unclear in RA. So, we used the rat adjuvant-induced arthritis (AIA) model to characterise finely bone loss kinetics for 4 different bones from inflamed ankle; then, we investigated early bone microarchitecture changes, at the time of arthritis onset.

Materials and Methods Seven-weeks-old female Lewis were divided into two groups: AIA and control (CTRL) rats. Arthritis was induced (AIA) or not (CTRL) by injection at the base of the tail of the adjuvant Mycobacterium butyricum defined as day 0. Rats were monitored clinically by arthritic score and ankle perimeter. Erosion occurrence and bone microarchitecture damages were followed longitudinally using micro-Computed Tomography (micro-CT: vivaCT 40, Scanco, Switzerland). All animals were evaluated at microarchitecture level three times: at baseline, at arthritis onset day, and at day 17 after AIA induction. Navicular, astragalus, cuboid, and calcaneus bones were analysed at cortical site, trabecular site, and together. Analysed microarchitecture parameters included bone volume (BV/TV) and surface/volume (BS/BV), trabecular thickness (Tb. Th), cortical porosity (Ct. Porosity), and cortical thickness (Ct. Th). We performed non-parametric statistics. P-values ​​under 0.05 were considered statistically significant.

Results The clinical signs of arthritis appeared 10 days post-induction in AIA rats. In micro-CT at day 17, we observed that BV/TV, Tb. Th, and Ct. Th parameters were reduced in AIA compared to CTRL rats (p < 0.01). Ct. Porosity, and BS/BV parameters enhanced in AIA rats compared to CTRL rats (p < 0.01). At day 10 (the day of arthritis onset), erosions were observed only in AIA rats, associated with a lower BV/TV in both trabecular and cortical compartments (p<0.01) in AIA compared to CTRL rats. The trabecular bone loss was correlated with a higher BS/BV (p < 0.01) and a lower Tb. Th (p < 0.05) in AIA rats, while the trabecular loss was associated with a higher Ct. Porosity (p < 0.01). Tarsal bones both reacted similarly, but arthritis affected strongly the navicular bone.

Conclusions A quantified and more accurate profile of bone loss in the AIA model was obtained. The main finding is early bone microarchitecture changes in rat AIA, detectable by micro-CT, inside subchondral bone. These changes occur at the same day of inflammation onset in synovia, and both cortical and trabecular bone losses are present, reinforcing the inside-outside concept.