Abstract
Osteocytes play a critical role in bone mechanobiology, sensing and responding to mechanical loading through fluid flow within the lacunar-canalicular network (LCN). Experimental measurements of interstitial fluid flow in bone are difficult due to the embedded nature of osteocytes in the dense mineralized matrix. Therefore, accurate computer simulations of these processes are essential for understanding bone mechanobiology. Two computational approaches have mostly been used to characterize convective interstitial fluid flow in bone: poroelastic finite element (FE) models, which treat bone as a homogenized porous medium, and fluid-structure interaction (FSI) models, which incorporate explicit LCN microarchitecture. However, these approaches have predicted fluid velocities that differ by three to four orders of magnitude. Here, we investigate the reasons for this discrepancy and demonstrate how imposed pressure gradients influence the predicted fluid velocities. Using an FSI model of a single osteocyte embedded in the mineralized matrix, we show that when an imposed pore pressure gradient is smaller than that generated by bone matrix deformation under mechanical loading, the convective fluid velocities in the canaliculi reach ∼100 nm/s and scale with the applied strain. In contrast, applying higher pressure gradients decouples fluid flow from the solid bone matrix deformation, resulting in fluid velocities bigger than 100 μm/s that are insensitive to loading conditions. Future studies investigating the effect of load-induced convection flow on osteocyte mechanobiology should therefore apply small imposed pressure gradients to avoid overestimating interstitial flow and more realistically capture load-induced convective flow.