Outcomes of shoulder arthroplasty by year of index procedure: Are we getting better?
These authors sought to determine if postoperative patient-reported outcomes improved over time following anatomic (TSA) and reverse (RTSA) total shoulder arthroplasty.
3,467 patients underwent either primary TSA, RTSA, or a revision procedure, while the remaining
patients underwent shoulder hemiarthroplasty (n = 228) and insertion of an antibiotic spacer (n =
21). 2,952 (85%) patients completed baseline American Shoulder and Elbow Surgeons (ASES) score
surveys. 1,899 (55%) patients completed postoperative ASES surveys at 2 and/or 5 years were included in the retrospective analysis. There were 1,323 anatomic total shoulder arthroplasties and 576 reverse total shoulder arthroplasties; 1786 were primary arthroplasties.
The average preoperative to postoperative difference in ASES scores for the patients responding at 2 years did not improve significantly with more recent dates of surgery for either TSA or RTSA.
The average preoperative to postoperative difference in ASES scores for the patients responding at 5 years after surgery did improve with more recent dates of surgery for TSA and RTSA.
Patient sex, ASA classification, rotator cuff status, primary diagnosis, Walch classification, and revision procedures were significant factors affecting the ASES score.
Comment: Consistent with a prior study (see this link and this link) this analysis did not provide evidence that technological innovations (such as newer versions of implants, 3D preoperative planning, or patient specific instrumentation) led to improved incomes. Instead it identified changes over time in the diagnoses for arthroplasty: the percentage of shoulders with osteoarthritis having anatomic or reverse arthroplasty increased with more recent dates of surgery while the percentage shoulders having rheumatoid arthritis, avascular necrosis, cuff tear arthropathy, fracture, post-traumatic deformity, cuff tear, instability, infection or an “other” diagnosis decreased. There were also changes for more recent dates of surgery in the percentages of patients with different glenoid types. Thus it seems likely that selection bias may have had an effect on the trend in outcomes.
The arthroplasties were performed by 25 different surgeons. The implants used in this study are not specified. The effect of surgeon or implant on outcomes were not presented.
The lack of improvement in two year outcomes is not explained. It would seem that if outcomes are improving, the two year outcomes for arthroplasties performed more recently would be better than the two year outcomes for those performed earlier.
Five-year outcomes were available for 634 arthroplasties (492 anatomic and 142 reverses). The response rate (# of arthroplasties with five year outcome data/the total # of arthroplasties performed five or more years prior to the study) is unknown, so the risk of non-response bias cannot be determined (see this link).
Finally, while it has been shown that conditions such as depression, narcotic use, smoking, medical comorbidities, and prior surgery have negative impacts on arthroplasty outcomes, these were not assessed in thus study.
Careful clinical research is needed to determine the factors influencing arthroplasty outcomes and to inform our efforts to achieve the best outcomes for our patients.
Follow on twitter: https://twitter.com/shoulderarth
Follow on facebook: https://www.facebook.com/frederick.matsen
Follow on LinkedIn: https://www.linkedin.com/in/rick-matsen-88b1a8133/
How you can support research in shoulder surgery Click on this link.
Here are some videos that are of shoulder interest
