Importance of depth-wise distribution of collagen and proteoglycans in articular cartilage--a 3D finite element study of stresses and strains in human knee joint.

Proteoglycans and collagen fibrils are distributed inhomogeneously throughout the depth of articular cartilage, providing the tissue with its unique depth-dependent properties and directly influencing local tissue deformations and stresses in in vitro/in situ. The aim of this study was to investigate the importance of the proteoglycan and collagen distributions for cartilage stresses and strains resulting from dynamic joint loading (i.e., a simulated gait cycle) and mechanical equilibrium in a knee joint. A 3D finite element model of a human knee joint including femoral and tibial cartilages and menisci was created. In order to characterize the effects of collagen orientation, collagen distribution and proteoglycan distribution on knee joint stresses and strains, five fibril-reinforced poroviscoelastic models with different depth-wise tissue structure were created. For each model strains and stresses were evaluated at four different depths in the medial tibial compartment during a gait cycle (simulating walking) and at mechanical equilibrium (simulating standing). The model with arcade-like collagen fibril architecture predicted substantially lower stresses than the homogeneous model, especially during dynamic joint loading. The depth-wise proteoglycan gradient caused a substantial increase in stresses and axial strains in the superficial layer, and reduced stresses and strains in the deep layer under static loading. The effect of fibril volume density distribution was minor during both dynamic joint loading and at mechanical equilibrium. The present study emphasizes the importance of the arcade-like collagen fibril orientation for cartilage function in a human knee joint. However, we suggest that, for practical reasons, a constant fibril volume density may be used in 3D models of knee joints, whereas a realistic depth-wise proteoglycan distribution should be applied when simulating the cartilage response during mechanical equilibrium.