Introduction Successful glenoid component fixation in shoulder arthroplasty is dependent on the quality of the underlying bone. The quantity of trabecular bone available for fixation is small and its properties are critical for both fixation and load bearing. Indentation testing has been used previously to determine regional changes in the mechanical properties of the glenoid surface [1]. However, there has been no attempt to relate these properties to the quality of the surrounding bone. The aim of this study was to investigate the relationship between the mechanical properties of the surface with both the trabecular bone volume fraction and the cortical thickness of the underlying bone. Materials and

Methods Nineteen embalmed glenoids were obtained from human cadavers (mean age 82 years). Previous work had shown that embalming had minimal impact on the mechanical properties of bone derived using indentation testing [2]. Indentation tests were performed using a 2.95 mm flat cylindrical indenter, with a speed of 2 mm/min, at 11 pre-selected grid points, up to a depth of 3 mm. Care was taken to ensure that the indenter surface was perpendicular to the local surface of the glenoid. The stiffness and maximum load following mechanical properties were measured from the resulting load-displacement curve. The Young’s modulus and strength were derived using the formula given in [3] and normalising with respect to the indenter cross section, respectively. Each of the glenoids was scanned using a large sample microCT (Scanco uCT 80) at a resolution of 78 microns. The cortical thickness and bone volume fraction (BV/TV) local to each of the grid points was determined from the 3-D reconstructions of these scans.

Results The mean strength and elastic modulus of each of the 11 indentation sites ranged from 26 to 67 MPa and 83 to 184 MPa, respectively. The largest value of BV/TV was found at the posterior edge (0.41%) and the lowest at the inferior edge (0.14%). The measured cortical thickness ranged from 0.68mm to 0.88mm with the thickest at the superior edge. Multiple regression analysis found, in the main, a significant correlation between strength and BV/TV for data derived from each of the indentation sites. The elastic modulus had only a weak correlation with BV/TV. Cortical thickness was found to have only a very marginal influence on both the elastic modulus and strength.

Discussion The indentation and uCT analysis have been used for the first time to relate the glenoid’s mechanical properties to bone morphology. The distribution of the BV/TV data is similar to that found by Frich et al [4] and for BMD measurements for BMD [5]. However, the cortical thickness measurements differ from those of Frich [4]. The local bone volume fraction strongly influenced the strength at the glenoid surface. Further investigations are ongoing to determine more fully the morphological factors important in the properties of the glenoid surface and whether such factors can be a predictor of clinical success.

Loosening of the glenoid component after Total Shoulder Arthroplasty is an established phenomenon with long-term follow-up studies showing radiolucency in 65% of glenoid components at 10 years (Stewart and Gray, 1997). Glenoid component designs are based on anthropometric measurements of normal shoulder joints. The purpose of this study was to study the surface anatomy of both bony and cartilaginous layers of the normal glenoid fossa in more detail.

We have developed a reproducible and inexpensive technique of surface shape assessment using laser morphometric analysis and applied this to thirty normal glenoid fossae mounted in the scapular plane. Surface analysis was carried out before and after removal of the glenoid labrum and after papain digestion of the articular cartilage allowing assessment of the skeleton alone allowing comparison with other studies using bony or cartilaginous landmarks in assessment of glenoid version. Using a specially designed program, five equi-distant lines were placed across the glenoid from which analysis of the orientation of the fossa was determined.

None of the scapulae presented a single surface that could be judged anteverted or retroverted by an amount representable by a single figure. All scapulae demonstrated a twist about the vertical axis. Two main types were identified.

Type 1 – Superior retroversion (mean 16.0 degrees) becoming progressively less to the lower pole (mean 3.0 degrees)

Type 2 – Retroverted superiorly, twisting to reach the position of maximal anteversion in the lower half of the glenoid, twisting back into more retroversion towards the lower pole. None of the specimens were morphologically equivalent to currently available glenoid prostheses. The surface shape of the glenoid is so variable that we should explore the relationship between this and the kinematics of the shoulder joint. There may be implications for the design of shoulder replacements and possibly custom prostheses.