A comparison of hemiarthroplasty with aortograft versus reverse total shoulder arthroplasty following proximal humeral oncologic resection.

BACKGROUND Reconstruction after proximal humerus tumor resection poses challenges in restoring stability and function due to loss of rotator cuff attachments and soft-tissue integrity. Hemiarthroplasty (HA) with synthetic mesh augmentation has traditionally been used, whereas reverse total shoulder arthroplasty (rTSA) has emerged as an alternative. Comparative data between these techniques in oncologic settings remain limited. This study compared functional outcomes and complication profiles between HA with aortograft mesh (HA-aortograft) and rTSA endoprosthetic reconstructions following proximal humeral oncologic resection.

METHODS A single-center retrospective cohort study was performed on 68 patients who underwent oncologic proximal humerus resection and reconstruction with HA-aortograft (n = 58) or rTSA (n = 10) from 2000 - 2025. Functional outcomes, including forward elevation (FE), external rotation (ER) lag, and internal rotation (IR) to the hip or less, were assessed preoperatively, 6 and 12 months postoperatively, and at final follow-up. Complication, recurrence, and revision rates were recorded. Between-group comparisons used t-tests or Fisher's exact tests; within-group changes used paired tests.

RESULTS rTSA had shorter mean final follow-up (21.9 vs 43.5 months, P = .001) and shorter resection lengths (7 vs 12.7 cm, P < .001). rTSA demonstrated superior FE at 6 months (76° vs 34°, P = .004) and final follow-up (87° vs 41°, P = .018); this superiority persisted when restricting analysis to those with preserved deltoid insertions. Longitudinally, HA-aortograft lost significant FE function from baseline at 6 months (-36°, P = .003) and 12 months (-25°, P = .042). In contrast, rTSA achieved significant gains at 6 months (+50°, P = .015) and final follow-up (+60°, P = .023). A greater proportion of rTSA patients achieved the minimal clinically important difference for FE (≥ 12°) at 6 months (80% vs 20.6%, P = .001) and final follow-up (77.8% vs 26.5%, P = .008). Rotational outcomes were largely comparable cross-sectionally, though rTSA showed a trend toward greater improvement in ER lag and IR limitation. Rates of implant-related complication (rTSA 10% vs HA-aortograft 12.1%) and revision (rTSA 10% vs HA-aortograft 5.2%) were similar between groups.

CONCLUSION rTSA endoprosthetic reconstruction provided superior restoration of FE and overall functional recovery compared with HA-aortograft following proximal humeral oncologic resection, without increased complication or revision rates.