Anatomic versus reverse shoulder arthroplasty for post-traumatic sequelae of operatively and nonoperatively treated proximal humerus fractures

Abstract

BackgroundThis study aimed to determine whether there is a long-term difference in outcomes between anatomic total shoulder arthroplasty (aTSA) and reverse shoulder arthroplasty (rTSA) performed for proximal humerus fracture (PHF) sequelae. Hypotheses were as follows: (1) patients undergoing aTSA would have improved functional outcomes but a greater incidence of adverse events (AEs) and reoperation than those undergoing rTSA and (2) patients undergoing shoulder arthroplasty after open reduction internal fixation (ORIF) would have worse outcomes with more AEs and reoperations than those undergoing shoulder arthroplasty for sequelae of nonoperatively managed PHF. MethodsA prospectively collected database was queried for patients with PHF sequelae undergoing aTSA or rTSA between 2007 and 2020 with minimum 2-year follow-up. Baseline demographics, perioperative data, postoperative AEs, functional outcomes, and range of motion (ROM) were compared between aTSA and rTSA groups. A secondary analysis was performed to compare patients treated with prior PHF ORIF vs. those treated nonoperatively. ResultsThere were 17 patients in the aTSA group and 83 patients in the rTSA group. Type I PHF sequelae predominated among patients who underwent aTSA (71% vs. 40%, P = .026). Incidence of total postoperative AEs was greater after aTSA than that after rTSA (12% vs. 5%, P = .277), with a significantly higher rate of glenoid aseptic loosening after aTSA (6% vs. 0%, P = .026). All AEs required reoperation except one in the rTSA group. The mean follow-up was 66 months for aTSA compared with 45 months for rTSA (P = .002). No differences in functional outcomes or ROM between aTSA and rTSA persisted beyond 3 months or at the final follow-up except external rotation, which favored aTSA for 3 years postoperatively. In the secondary analysis, there were 33 patients in the ORIF group and 67 in the non-ORIF group. One (3%) postoperative AE occurred in the ORIF group vs. five (7%) in the non-ORIF group (P = .385). At a mean follow-up of 4 years, there were no differences in functional outcome scores or ROM between ORIF and non-ORIF groups, except for patient-reported shoulder function (6.3 vs. 7.4, respectively, P = .037). ConclusionFor treatment of PHF sequelae, aTSA may result in a higher incidence of postoperative AEs and reoperation than rTSA, particularly due to glenoid aseptic loosening. No difference in functional outcome scores between aTSA and rTSA persists beyond 3 months or at the final follow-up. Although active external rotation is significantly improved after aTSA for the first 3 years postoperatively, no differences in ROM exist beyond 4 years at the final follow-up. Patients undergoing shoulder arthroplasty for PHF sequelae have comparable outcomes regardless of prior ORIF or nonoperative management. Level of evidenceLevel III; Retrospective Cohort Design; Treatment Study