Total Shoulder Arthroplasty Outcome for Treatment of Osteoarthritis: A Multicenter Study Using a Contemporary Implant
In this article, we present clinical results of primary total shoulder arthroplasty for osteoarthritis using an implant that provides dual eccentricity and variable neck and version angles for reconstruction of proximal humeral anatomy.
Two hundred one patients with symptomatic osteoarthritis underwent 218 total shoulder arthroplasties with a fourth-generation anatomical shoulder and a replicator plate at 4 centers between August 1, 2006, and December 31, 2010. Fourth-generation implants allow for varying humeral neck and version angles and have dual eccentricity so as to be consistently able to cover the humeral head cut.
At a mean follow-up of 3 years (minimum, 2 years), there was an 81% follow-up rate. At final follow-up, 3 objective measures were significantly (P < .05) improved over preoperative levels: average active elevation (preoperative, 92°; postoperative, 137°), active external rotation (pre, 15°; post, 42°), and active internal rotation (pre, S3; post, L2).
The functional outcome scores that were significantly (P < .05) improved at final follow-up were Constant normalized (pre, 39; post, 79), Shoulder Pain and Disability Index (pre, 86; post, 20), Simple Shoulder Test (pre, 3.3; post, 10), UCLA Shoulder Rating Scale (pre, 13; post, 31), and American Shoulder and Elbow Surgeons Shoulder Assessment (pre, 33; post, 85). Complications were noted in 11% of the shoulders. The most common complications were rotator cuff failure (13, 6%) and infection (5, 2%).
In 21 shoulders, these complications were treated with revision shoulder arthroplasty (16 shoulders), arthroscopic capsular release (3), evacuation of postoperative hematoma (1), and débridement of suture abscess (1). The 16 revision shoulder arthroplasties performed were conversion to reverse shoulder arthroplasty (11 shoulders) and placement of an antibiotic spacer for infection (5). The stem was left in place for all revisions, excluding those for infection. This is a significant advantage of the modular platform stem. Details of the complications and treatments are listed in the Table. There was no difference in health status between patients with a complication (ASA, 2.57) and those without one (ASA, 2.56).
Discussion
The implant described in this article consists of a metaphyseal press-fit stem, a replicator plate, multiple eccentric humeral heads, and a glenoid of multiple sizes with 2 radii of curvatures used to match the patient’s native anatomy and still maintain the appropriate radius of curvature mismatch between the humeral head and the glenoid. Between the eccentricity in the replicator plate and the eccentricity in the humeral head, almost any humeral head cut can be covered, more than 99% of the time.1 However, it remains to be seen if a versatile implant that comes close to matching the patient’s native anatomy will make a difference clinically.
The objective and functional outcomes in this study compare well with those of other, large TSA studies using older prostheses.1-4 There are few reports on contemporary implants with sufficient follow-up numbers for the single diagnosis of OA. Norris and Iannotti2 reported on a multicenter study of 176 patients with a Depuy Global TSA. The design of their study comes closest to that of our clinical outcome study. Nineteen surgeons were involved in their study. The follow-up rate is not clear. Their outcomes (with ours in parentheses for comparison) were active external rotation of 45° (42°), active elevation of 138° (137°), ASES of 84 (85), and SST of 9.2 (10). Norris and Iannotti2 noted an overall complication rate of 13% (12% in our series). Their most common postoperative complications were RCF and glenoid loosening; ours were RCF and infection. Another multicenter study with short-term results using a contemporary prosthesis included 268 shoulders followed for a minimum of 12 months.1 At final follow-up, Constant score was 97, active elevation was 145°, and the complication rate was 8.6%. Godenèche and colleagues1 also noted a glenoid lucent-line rate of 58% and reported that rotator cuff pathology adversely affected outcome.
Although the overall clinical outcome results are encouraging and the complication rate is in the reported range, we believe that a focus on the major complication categories may have a significant positive impact on our patients. The present article places significant importance on reporting complications prospectively, which is more accurate than retrospective reporting. The rates of both RCF and infection, the most common complications in our study, need to be decreased. Aldinger and colleagues7 reported a 12% complication rate in 485 primary shoulder arthroplasties—a rate identical to ours here. In their study, nerve injuries and humeral fractures were both more common than rotator cuff tears. We think that rotator cuff deficiency after TSA is underreported because it is often based on revision surgery alone. It is also interesting that the majority of the cuff deficiencies were through the upper subscapularis rotator interval and were not a complete failure of the subscapularis repair. Not all these patients will undergo revision surgery. In the future, the RCF rate may drop with the increasingly common use of reverse shoulder arthroplasty for substandard rotator cuffs.
Use of this contemporary variable neck angle, variable version angle, dual eccentric shoulder arthroplasty with a replicator plate provides satisfying short-term clinical outcomes. Patients with less than optimal health (mean ASA, 2.57) seem to tolerate the procedure well. Continued focus on RCF and infection will have the greatest impact on the overall complication rate.
