Abstract
Introduction
Aseptic glenoid loosening is a common failure mode of reverse shoulder arthroplasty (rTSA). Achieving initial glenoid fixation can be a challenge for the orthopedic surgeon since rTSA is commonly used in elderly osteoporotic patients and is increasingly used in scapula with significant boney defects. Multiple rTSA baseplate designs are available in the marketplace, these prostheses offer between 2 and 6 screw options, with each screw hole accepting a locking and/or compression screw of varying lengths (between 15 to 50mm). Despite these multiple implant offerings, little guidance exists regarding the minimal screw length and/or minimum screw number necessary to achieve fixation. To this end, this study analyzes the effect of multiple screw lengths and multiple screw numbers on rTSA initial glenoid fixation when tested in a low density (15pcf) polyurethane bone substitute model.
Methods
This rTSA glenoid loosening test was conducted according to ASTM F 2028–17; we quantified glenoid fixation of a 38mm reverse shoulder (Equinoxe, Exactech, Inc) in a 15 pcf low density polyurethane block (Pacific Research, Inc) before and after cyclic testing of 750N for 10k cycles. To evaluate the effect of both screw fixation and screw number, glenoid baseplates were constructed using 2 and 4, 4.5×18mm diameter poly-axial locking compression screws (both n = 5) and 2 and 4, 4.5×46mm diameter poly-axial locking compression screws (both n = 5). A two-tailed unpaired student's t-test (p < 0.05) compared prosthesis displacements to evaluate each screw length (18 vs 46mm) and each screw number (2 vs 4).
Results
All glenoid baseplates remained well-fixed after cyclic loading in the low density bone substitute block, regardless of screw length or screw number. As described in Table 1, the average pre- and post-cyclic displacement for baseplates with 18mm long screws was significantly greater than that of baseplates with 46mm long screws in both the A/P and S/I directions, with exception of displacements for 4 screws S/I-pre cyclic and 2 screws A/P-post cyclic loading. As described in Table 2, the average pre- and post-cyclic displacement for all baseplates with 2 screws was significantly greater than that of all baseplates with 4 screws, regardless of screw length in the A/P and S/I directions.
Discussion and Conclusions
These results of this study demonstrate that rTSA glenoid baseplate fixation is impacted by both the number of screws and by the length of screws, with longer screws and more screws associated with significantly better initial fixation. However, it should be noted that none of the tested devices catastrophically failed in this non-defect/low-density model, demonstrating that adequate fixation can be achieved with as little as 2×18mm screws for some baseplate types. Care should be made when extrapolating these results to that of other designs. This study is limited by its use of only one implant design and by its use of a polyurethane substrate without any defect; future work should evaluate the effect of screw length and screw number in with multiple different prostheses in different densities of bone with and without defects.
