Abstract
The reverse ball and socket shoulder replacement, employing a humeral socket and glenosphere, has revolutionized the treatment of patients with arthritis and rotator cuff insufficiency. The RSP (DjO Surgical, Inc., Austin, Texas) is one such device, characterized by a lateral center of rotation and approved for use in the United States since 2004. Multiple studies by the implant design team have documented excellent outcomes and low revision rates for the RSP, but other published outcomes data are relatively sparse. The objective of this study is to report on the complications and early outcomes in the first consecutive 60 RSPs implanted in 57 patients by a single shoulder replacement surgeon between 2004 and 2010. Forty-four patients were female and mean age at the time of reverse shoulder arthroplasty was 75 years (range 54 to 92 years). The RSP was used as a primary arthroplasty in 42 shoulders and to revise a failed prosthetic shoulder arthroplasty in 18 shoulders. During the study period, 365 shoulder replacements were implanted so that the RSP was used selectively, accounting for only 17% of all shoulder arthroplasties (8.4% for 2004-2007, 24.2% for 2008-2010). Most patients had pseudoparalysis and profound shoulder dysfunction so that mean pre-operative active forward elevation was to 45°, active abduction to 43°, active internal rotation to the buttock, and the mean pre-operative Simple Shoulder Test (SST) score was 1 out of 12. At final follow-up, mean active forward elevation had improved to 101° (p<0.0001), active abduction to 91° (p<0.0001), active internal rotation to the lumbosacral junction (p<0.001), and the mean final SST score was 7 out of 12. There were 16 complications in 14 patients, including 7 reoperations in 6 patients (11%): 3 closed reductions for dislocation, 2 open revisions for instability and for a dissociated liner in the same patient, one evacuation of a hematoma, and one open reduction and internal fixation of a post-operative scapular spine fracture. Two additional scapular spine or acromion fractures and one acromioclavicular joint separation developed postoperatively that impacted outcome adversely but did not require re-operation. None of the glenoid baseplates or humeral stems has been revised and no deep infections have occurred. Experience with reverse shoulder arthroplasty appears to influence the reoperation rate, as 3 of the reoperations occurred following the first 15 reverse shoulder arthroplasties. Overall improvements in active motion and self-assessed shoulder function were comparable to those reported previously. Final active motion results were somewhat lower than those reported previously, which may relate to the selection of predominately pseudoparalytic patients for reverse shoulder arthroplasty in this series. Use of the RSP device for reverse shoulder arthroplasty leads to improved motion and function in carefully selected older patients with pseudoparalysis or a failed shoulder replacement. Re-operations and complications occur but the learning curve may not be as steep as previously reported. This may relate to specific features of the implant system used in this series, as well as to surgeon experience.
