Abstract

The water apparent diffusion coefficient (ADC) in rabbit Achilles tendon is anisotropic, diffusion-time dependent, and changes as a function of tensile load. Water ADC changes of tendon under mechanical load are thought to be due to the extrusion of water from the more restricted tendon core to a relatively unrestricted bulk phase at the periphery (rim) of the tendon. Tensile loading may influence water ADC values by changing the spatial separation of restricting barriers (e.g., increasing the tendon fibril packing density). To explore this issue, we have applied porous-media theory to the investigation of water ADC changes in rabbit Achilles tendon under two different mechanical loading conditions (a baseline condition with a minimal tensile stress and a second in which the tensile stress was approximately 1 MPa). Diffusion sensitivity was applied in directions parallel and perpendicular to the long axis of the tendon. The short diffusion-time behavior of the resulting time-dependent ADC curves was used to indirectly infer information regarding the average surface area to volume ratio of the space available for molecular diffusion. From these values, we estimated a 40% reduction in volume available for diffusion in the perpendicular direction after tensile loading, but only a 10% reduction in the parallel direction. These differences are consistent with the known geometry of the tendon microstructure and suggest an increase in fibril packing density upon loading. The long diffusion-time behavior of the time-dependent ADC curves was used to indirectly infer the tortuosity of the diffusion pathways through the interstitial space. The tortuosity in the direction perpendicular to the tendon long axis was approximately 2.5 times greater than that in the parallel direction. Stimulated-echo measurement of the ADC values at longer diffusion times resulted in T1 spin editing of water with shorter T1 values (and correspondingly lower ADC values). The resulting increase in water ADC with increasing diffusion time was attributed to multiple components arising from the (overlapping) distribution of T1 values in the core and rim regions of the tendon.