Comparative outcomes of anatomic and reverse total shoulder arthroplasty for rotator cuff-intact glenohumeral osteoarthritis: a propensity score-matched, age-stratified exploratory analysis.

Background: The optimal implant selection for patients with rotator cuff-intact primary glenohumeral osteoarthritis remains controversial, particularly with respect to patient age. Although reverse total shoulder arthroplasty (rTSA) has been increasingly utilized in older patients, it is unclear whether chronological age alone modifies the relative effectiveness of anatomic total shoulder arthroplasty (aTSA) versus rTSA when baseline characteristics are adequately controlled. Methods: This retrospective comparative study included patients with rotator cuff-intact primary glenohumeral osteoarthritis who underwent aTSA or rTSA with a minimum follow-up of 3 years. Propensity score matching was performed in a 1:1 ratio to minimize baseline differences between groups, including age, sex, pre-operative clinical outcomes, range of motion, and rotator cuff muscle quality assessed by fatty infiltration. Clinical outcomes were compared using the American Shoulder and Elbow Surgeons (ASES) score as the primary outcome measure. Secondary outcomes included the Constant score, Single Assessment Numeric Evaluation, and post-operative range of motion (forward elevation, external rotation, and internal rotation), as well as complications including scapular notching and acromial or scapular spine stress fracture in the rTSA group and glenoid loosening or rotator cuff failure in the aTSA group. Age-based subgroup analyses (≤70 years vs. > 70 years) and age-procedure interaction analyses were conducted to evaluate whether age modified the association between implant type and outcomes. Results: After matching, 44 well-balanced pairs (88 patients) were identified. There were no significant differences between the aTSA and rTSA groups in post-operative ASES scores or secondary clinical outcomes. Age-stratified analyses demonstrated consistent results across both age groups, and no significant age-procedure interaction effects were observed. From a clinical perspective, both procedures achieved post-operative ASES and Constant scores exceeding established minimal clinically important difference thresholds. Although the overall complication rate was numerically higher in the rTSA group, these events were predominantly procedure-specific and were largely managed with conservative treatment. Revision rates were low and comparable between groups. Conclusion: When baseline characteristics are adequately balanced, aTSA and rTSA provide comparable short-term to midterm clinical outcomes in patients with rotator cuff-intact primary glenohumeral osteoarthritis. Chronological age alone does not appear to modify the relative effectiveness of implant type, suggesting that implant selection should be guided by rotator cuff integrity, glenoid morphology, and patient-specific functional demands rather than age alone.