A two-dimensional morphometry-based model of interstitial and transcapillary flow in rabbit synovium

The synovial lining of a joint is a layer of specialized connective tissue, containing fenestrated capillaries, that regulates the volume and composition of fluid in the joint cavity. The hydraulic conductance of the synovial lining and of the plasma-to-cavity pathway is increased by high intra-articular pressures (IAP greater than 9 cmH2O) and this is accompanied by an increase in the interstitial pathway's area/pathlength ratio. In order to assess the contribution of the altered pathway geometry to the conductance changes, and also to evaluate the conductivity of the synovial interstitial matrix (Ki) and local flow patterns, a two-dimensional model of trans-synovial flow was developed from morphometric data at low IAP (5 cmH2O) and high IAP (25 cmH2O). Darcy's law was applied to finite elements within a 'unit cell' of synovium to compute a steady-state pressure field. Elements adjacent to limited discrete porous regions of the capillary wall were additionally subjected to the Starling principle. Appropriate values of Ki and capillary wall conductance were obtained by iteration to match the experimentally measured net conductances. The value of Ki that matched the data at low IAP was 1.4-2.1 x 10(-12) cm4 s-1 dyn-1 (1.4-2.1 x 10(-15) m4 s-1 N-1). This is in the range reported for some other fluid-confining tissues, namely scleral stroma and aortic wall. Comparison between a capillary model with localized porous regions and one with a uniformly distributed permeability showed that the uniformity assumption (commonly used in mathematical modelling) leads to severe underestimation of the local pericapillary pressure gradients. At high IAP, synovial deformation was found to account for 24-50% of the increased hydraulic conductances. To explain the remainder, it was necessary to postulate a rise in Ki, especially in the zone between the joint cavity and capillary wall. Possible mechanisms involving tissue hydration and/or glycosaminoglycan wash-out at high IAP are discussed. The model highlights a need for quantitative biochemical analysis.