Using the biphasic theory for hydrated soft tissues (Mow et al., 1980) and a transversely isotropic elastic model for the solid matrix, an analytical solution is presented for the unconfined compression of cylindrical disks of growth plate tissues compressed between two rigid platens with a frictionless interface. The axisymmetric case where the plane of transverse isotropy is perpendicular to the cylindrical axis is studied, and the stress-relaxation response to imposed step and ramp displacements is solved. This solution is then used to analyze experimental data from unconfined compression stress-relaxation tests performed on specimens from bovine distal ulnar growth plate and chondroepiphysis to determine the biphasic material parameters. The transversely isotropic biphasic model provides an excellent agreement between theory and experimental results, better than was previously achieved with an isotropic model, and can explain the observed experimental behavior in unconfined compression of these tissues.

The tensile properties of distal femoral growth plates from 12-month-old cows were determined on uniformly prepared straight bone-growth plate-bone specimens (7 x 7 mm2 in cross-section) from predetermined anatomical sites on the physis. Each specimen was distracted to failure using a computer-controlled servo-hydraulic testing machine at a strain rate of 0.004 s-1. It was found that the exponential constitutive law, using finite deformation formulation for strain, provides an excellent description of the stress-strain behavior of all the specimens up to the point of failure. The ultimate stress and both tangent moduli (i.e., the toe region tangent modulus and the tangent modulus calculated at 75% of ultimate strain) varied with anatomical site. The anterior region was the strongest, followed by the posterior/lateral. The anterior and posterior/lateral regions were also the stiffest, whereas the posterior/medial and center regions were the weakest and most compliant. The bone-growth plate-bone specimen exhibited a low ultimate strain (13.8% +/- 6%) that did not vary significantly throughout the growth plate. This result suggests that disruption of the physis may occur in vivo even at the lower distractions currently recommended for the clinical chondrodiatasis procedure for leg lengthening. The biochemical composition of the growth plate in the anatomical regions correlated well with the tensile properties. There was a greater collagen content in the regions that were the stiffest and strongest. The gross morphology of the growth plate of the bovine distal femur is also described in this study. There is a regular pattern to the undulations of the physis at several dimensional levels. Histologic findings showed that orientation of the hypertrophic cell columns and transphyseal septa are aligned nearly parallel to the longitudinal axis of the diaphyseal shaft. This column orientation is not affected by the undulation of the primary contour of the physis, which at certain locations may be inclined as much as 60 degrees relative to the diaphyseal axis. The orientation of the hypertrophic cell columns appears to be one of the dominant microstructural features influencing the tensile behavior of the bone-growth plate-bone specimens.

The compressive behavior of the bovine distal femoral growth plate was studied in vitro. Strain-rate controlled, compression stress-relaxation experiments were performed on cylindrical bone-growth plate-bone specimens from the interior and periphery of the growth plate. The questions addressed in this study were (a) Can the nonlinear biphasic theory, one with strain-dependent permeability, be used to represent the compressive stress-relaxation behavior of bovine growth plate? (b) How do different assumptions concerning the permeabilities of the chondro-osseous interfaces influence the inferred material properties of the growth plate? and (c) Are there any differences in these properties between the periphery and the interior of the growth plate? Intrinsic biphasic material properties--aggregate modulus (HA), Poisson's ratio (v), and nonlinear strain-dependent permeability coefficients (ko and M)--were calculated from the compression stress-relaxation data with use of a finite element model and a least squares curve-fitting procedure. To verify this constitutive model for the growth plate, an independent set of finite element analyses was performed with use of the determined intrinsic biphasic properties, and comparisons were made between these finite element predictions and two additional sets of experimental data subsequently obtained for the same specimens with use of two slower rates of compression. Excellent agreement was achieved between these finite element predictions and the latter two sets of data. The aggregate modulus was found to be insensitive to the permeability of the chondro-osseous interface. The permeability coefficients were very sensitive to, and the Poisson's ratio was only slightly sensitive to the interface permeability condition. Therefore, the periphery of the growth plate is more compliant and permeable than the interior.

Our biomechanical data indicate that the percutaneous compression plating provides adequate bending stiffness and torsional stability. With sliding capability this fixation leads to controlled fracture impaction, which is further enhanced by cyclic loading.

No abstract