JSES - 2026-08-01 - Journal Article; Comparative Study
Anatomic shoulder replacement has superior functional outcomes with equivalent survivorship compared with reverse shoulder replacement in patients with inflammatory arthritis-New Zealand registry results.
Newton BN, Frampton CMA, Kejriwal R
Topics
Key Takeaway
In 714 inflammatory arthritis shoulder arthroplasties, aTSA achieved higher Oxford Shoulder Scores than rTSA at all time points (41.9 vs. 35.4 at 10 years, P=.021) with equivalent 10-year revision-free survival (90.8% vs. 92.0%).
Summary Depth
Choose how much analysis to show on this article page.
Summary
This study compared functional outcomes and revision-free survival between aTSA and rTSA for inflammatory arthritis using New Zealand Joint Registry data from 2000–2022. aTSA patients were younger (65.7 vs. 70.3 years) with higher BMI, yet demonstrated statistically superior Oxford Shoulder Scores at 6 months, 5 years, and 10 years, though differences did not exceed the MCID. Revision rates were equivalent (0.99 vs. 0.92 per 100 component-years), but revision etiology differed: infection predominated in rTSA (33.3% vs. 7.4%), while glenoid loosening and cuff failure drove aTSA revisions.
Key Limitation
The differential mean follow-up (9.8 years aTSA vs. 6.0 years rTSA) combined with younger aTSA patients introduces substantial survivorship and selection bias that limits direct comparison of revision rates and long-term functional scores.
Original Abstract
BACKGROUND
Reverse shoulder arthroplasty (rTSA) has become increasingly common for inflammatory arthritis (IA) due to concerns regarding rotator cuff failure and glenoid loosening in anatomic total shoulder arthroplasty (aTSA). Comparative data between these procedures in IA, however, remains limited. This study aims to compare functional outcomes and revision-free survival between aTSA and rTSA in IA using data from the New Zealand Joint Registry.
METHODS
All primary aTSA and rTSA procedures performed for IA between 2000 and 2022 were identified in the New Zealand Joint Registry. Demographic data, revision events, causes of revision, and Oxford Shoulder Scores were analyzed. Revision rates were expressed per 100 component-years. Cox regression models estimated the hazard of revision, adjusted for age, sex, and American Society of Anesthesiologist score.
RESULTS
A total of 714 arthroplasties were included (433 rTSA and 281 aTSA) with mean follow-up of 6.0 years for rTSA and 9.8 years for aTSA. Patients with aTSA were younger (65.7 vs. 70.3 years, P < .001) and had higher body mass index (30.3 vs. 28.2, P = .018). aTSA demonstrated significantly higher Oxford Shoulder Scores at all time points: 38.7 vs. 36.4 at 6 months (P = .012), 41.9 vs. 38.0 at 5 years (P = .002), and 41.9 vs. 35.4 at 10 years (P = .021); however, this did not exceed the reported minimal clinically important difference. Revision rates were equivalent: 0.92 per 100 component-years for rTSA and 0.99 for aTSA (P = .529). Ten-year revision-free survival was 92.0% for rTSA and 90.8% for aTSA. Causes of revision differed, with infection more frequent in rTSA (33.3% vs. 7.4%, P = .02), whereas aTSA was more often revised for glenoid loosening and cuff failure.
CONCLUSION
Compared with rTSA, aTSA was associated with statistically higher functional outcomes and equivalent revision-free survival. IA is not a contraindication to aTSA, with it being a viable option for appropriately selected patients, rTSA should be considered for those with rotator cuff deficiency or severe glenoid pathology.