Abstract
Failure of reverse total shoulder arthroplasty (rTSA) due to loosening of the metaglene remains a concern. The metaglene is typically affixed to the glenoid via four peripheral bone screws, and the orientations of these screws can affect the stability of the metaglene. The purpose of this finite element analysis (FEA) study was to investigate whether screw orientations should be considered on a patient-specific basis to maximise early fixation.
Three-dimensional geometries of four scapula specimens were obtained by segmenting from CT data in 3D Slicer. A metaglene and four rigidly attached 4.5 mm diameter, 18 mm long cylinders representing screws, were placed on each reamed glenoid. Each screw was placed at one of four orientations, 15° or 7.5° toward or away from the central axis of the metaglene face, while all others were held in the baseline (BL) configuration, where all screws were perpendicular to the metaglene face. Finite element models were created by meshing with linear tetrahedral elements. Material properties of titanium (E=113.8 GPa, v=0.34) were applied to the metaglene and screws. Cortical bone material properties were considered uniform (E=17.5 GPa, ν=0.3) while cancellous bone material properties were non-uniform and mapped on an element-by-element basis using CT attenuation data. The scapula was fully constrained, and a 252 N superiorly oriented shear force was applied to the inferior portion of the metaglene. Contact was modelled at bone-implant and bone-screw interfaces. Displacements of the metaglene with respect to the glenoid were measured. The orientations of each screw that minimised in-plane displacement were used for specimen-specific (SS) configurations. A global (GL) configuration was also defined based on the averages of SS orientations. FE model-predicted metaglene displacements of the SS, GL, and BL screw configurations were compared using paired t-tests.
The average in-plane metaglene displacements for the SS, GL, and BL configurations were 4.8 ± 1.2, 6.5 ± 3.7, and 5.3 ± 1.5 um, respectively. SS configurations significantly decreased displacements by −0.4 ± 0.3 um (−8.5%, p = 0.024) when compared to BL, but the difference of −1.6 ± 3.1 um (25.3%, p = 0.187) was not significant when compared to the GL configuration.
In general, the SS configurations resulted in smaller metaglene displacements than the GL configurations, however the difference was not statistically significant. In one specimen, the GL configuration resulted in abnormally large displacements. These results indicate that, while on average, patient-specific orientations won't yield significantly greater fixation than global configurations; non-patient-specific configurations can, in some cases, yield poor results. Therefore, to ensure optimal fixation for all patients, screw orientations should be considered on a patient-specific basis.
