The effect of endplate permeability on the internal mechanics of the intervertebral disc , as determined by poroelastic modelling
D. McNally,N. Dhillon,P. Riches,A. W. Woods
Abstract:Introduction: The cartilage end-plate (CEP) is the structural unit situated between the intervertebral disc (IVD) and the vertebral body, consisting of hyaline cartilage and approximately 0.6mm in thickness (Roberts et al., 1989). The CEP is important for the mechanical functioning of the spine, since under compressive loads the CEP deforms under high stresses and bulges into the vertebral bodies (Holmes et al., 1993). Moreover, since the ‘normal’ CEP is porous, it provides one pathway via which metabolites can enter the disc, the other being through the outer annulus fibrosus (Nachemson et al., 1970). Whilst the main mechanism for transport of small solutes, such as oxygen and glucose, is diffusion, convective transport of large enzymes and hormones may be of some importance in maintaining a healthy disc (Urban et al., 1982). However, with increasing age and in some diseased states, such as scoliosis, the CEP can decrease in permeability by calcification, hindering the transport of both large and small solutes (Roberts et al., 1996). This phenomenon has been previously associated with IVD degeneration (Urban et al., 1977). A poroelastic model of the IVD can describe the convective transfer of fluid within the disc under application of an applied load. We derive our own formulation of the poroelastic equation in terms of pore pressure, based of the equations of Biot (Biot, 1941, 1956), and include strain dependent permeability and osmotic potential. We validate our model by fitting it to cyclic compression/expansion loading of non-degenerate discs. The CEP is incorporated into the model by a suitable boundary condition that assumes the CEP is a porous rigid layer. By varying the permeability of the CEP, we investigate theoretically the effect of a calcified CEP on the internal mechanics of the IVD. Methods: The experimental methodology has been previously published (Dhillon et al., 2001), and it is briefly reported here. Four human lumbar spines were harvested and muscular and ligamentous structures, facet joints, transverse processes and posterior elements were removed. From these four lumbar spines, six non-degenerated IVDs were obtained. Four of these discs were freshly tested and underwent repeat testing. Two discs were frozen, thawed, and subsequently tested, since it has been shown that freezing has an insignificant effect on the elastic and time-dependent properties of human IVDs (Dhillon et al., 2001). Before mechanically loading the discs, the vertebrae adjacent to the disc were embedded in polymethylmethacrylate. All specimens were tested while submerged in a bath of normal saline (0.15M NaCl) at room temperature. Each specimen was subjected to five consecutive cycles of axially compressive creep loading and recovery. The applied force was adjusted based on the cross-sectional area of each specimen such that the applied stress was 1MPa. A creep time of 20 minutes was used, and the discs were allowed to recover for 40 minutes at 0N between creep cycles. The heights of the IVDs were constantly measured and were normalised with respect to their original height, averaged, and a standard error calculated. The mechanics of the intervertebral disc can be described by the following 1d equation (Riches et al., 2001)
Medicine,Engineering