The paper titled, “Complication rates following reverse and anatomic shoulder replacement in treatment of glenohumeral arthritis: a 10-year Medicare review study” by Givens et al. is a 2024 retrospective cohort study (Level III evidence) published in the Journal of Shoulder and Elbow Surgery that aims to evaluate the complication rates of reverse shoulder arthroplasty (RSA) and anatomic total shoulder arthroplasty (aTSA).
The studied population was derived from a 5% subset of Medicare data that is collected from a random sample of beneficiaries. Therefore, this database review was able to capture patient data from multiple centers and surgeons to better represent outcomes of shoulder arthroplasty in the general population. A total of 5,935 aTSA and 2,911 RSA subjects were included, and these subjects were limited to those patients with primary glenohumeral arthritis as the diagnosis for the index aTSA or RSA claim. The measured outcomes included postoperative infection or inflammatory reaction, instability (dislocation), scapula fracture, peripheral nerve injury, postoperative hematoma, or any other mechanical complications. Each of these outcomes were assessed at three months, six months, one-year, two-years, and 5-years.
The results demonstrated statistically significant higher rates of postoperative infection, instability, scapula fractures, and early revision (at three months) in the RSA group compared to the aTSA group. Hazard ratios were utilized to compare the occurrence of each complication between groups. The authors concluded higher rates of complications among patients receiving RSA when compared to aTSA, particularly in regard to early instability and scapular fractures.
This study succeeds in its inclusion of a large subset of the general population and adequately examines outcomes that may have significant clinical implications. However, the study only includes patients with a primary diagnosis of glenohumeral joint osteoarthritis. This parameter may limit the breadth of the patient population being studied who receive RSA for rotator cuff arthropathy. Also, there is also no way to verify the diagnostic sub-category of “rotator cuff intact osteoarthritis” from this coding methodology, and therefore the study relies heavily on accurate surgeon coding, which is known to be imprecise. Additionally, there is an asymmetric rate of increase between procedures in the timeframe that is being investigated. The study reports that from 2010 to 2019, the prevalence of aTSA increased from 494 to 623 (26% increase), whereas the prevalence for RSA increased from 15 to 600 (390%). This drastic increase in the annual rate of surgeons performing RSA may confound the complications which may have been due to its novelty a decade prior.
The study does include a subset analysis comparing overall rates of complications between 2009-2016 and 2017-2019 and it shows significantly lower rates of infections, instability, scapular fractures, mechanical complications, and hematoma in the later 2017-2019 group, suggesting a “learning curve effect” that was difficult to isolate. However, the authors do not further distinguish RSA and aTSA in this sub-group analysis. Lastly, while the results show that the RSA group endured higher rates of revision at three months, although this was not maintained at one year or five years, the inherent follow-up period limits findings of hardware survivorship as those rates of revisions would likely be seen at ten years or more. This final limitation, which is acknowledged by the authors, substantially limits the epistemological value of this study and our ability to extrapolate the data beyond the included study time period.
Ultimately, this paper provides a valid perspective on the need to balance short term benefits of reverse TSA from a speed of rehabilitation standpoint with the potential for greater mid-term complication rates in the first 5 years. It also underscores the fact that more studies are needed regarding the use of RSA in cuff-intact osteoarthritis, technical factors related to deltoid tensioning and soft tissue balance in RSA as it relates to post-op complications, and the extent to which patient co-morbidities should factor into surgical indications for various diagnostic categories.