Conservation of the elastic and flexural moduli of osteopenic femoral cortical bone in experimental inflammatory arthritis in the rabbit*

Abstract

Experimental inflammatory arthritis (EIA) produced by carrageenan injection provokes a rapid bone remodeling state with cortical and cancellous bone loss. The objective of this study was to determine whether changes in cortical mechanical properties and/or geometry occur in long bones, either near or remote to the site of inflammation. EIA was induced in the right tibio-femoral joint of rabbits over 56 days. The right humerus and right femur from 15 normal and 25 arthritis group animals were excized. Semi-cylindrical specimens of the medial cortical shaft were subjected to non-destructive four-point bending tests. Transverse sections at the four contact sites of the loading jig were photographed and digitized to obtain average cross-sectional area (A) and moment of inertia (I). Moment of inertia and slope of the load/deflection curve permitted calculation of modulus of elasticity (E) for each specimen. Load/time curves were also used to calculate per cent stress remaining in relaxation experiments. Per cent stress remaining, E, A, I and $.......CONVERSION........$\sqrt {{\text{I/A}}} (radius of gyration) were examined for differences by bone (humerus, femur) and by treatment (N,A) using two way ANOVA. The induction of inflammatory arthritis did not significantly alter the modulus of elasticity in either the femur or humerus; however, arthritis reduced the moment of inertia from ${\text{34}}{\text{.54}} \pm {\text{2}}{\text{.88}} \times {\text{10}}^{ - {\text{12}}} {\text{m}}^{\text{4}} {\text{ to 25}}{\text{.06}} \pm {\text{1}}{\text{.80}}^{ - 12} {\text{m}}^{\text{4}} ({\text{mean}} \pm {\text{SEM,}}p < 0.05)$. This was observed in the femur (near the arthritic joint), but not in the humerus (remote from arthritic joint). Analysis of area and ratio I/A demonstrated that this geometric effect of treatment was due to reduced area without gross cross-sectional shape changes. Per cent stress remaining in the femur (but not in the humerus) was higher in the arthritis specimens than in the normal specimens $\left( {{\text{N:}}\;{\text{80}}{\text{.86}} \pm {\text{0}}{\text{.97}}\% ;\;{\text{A:}}\;{\text{83}}{\text{.25}} \pm {\text{0}}{\text{.71}}\% ,\;{\text{p}} < 0.05} \right)$. Thus, in this arthritis model, the principal mechanical or geometric effect on cortical bone was reduction of the cross-sectional area and moment of inertia. The viscoelastic relaxation response of bone was also altered, perhaps due to loss of water or collagen degradation.