Comparison of the clinical outcomes and complications of total shoulder arthroplasty and reverse total shoulder arthroplasty: A systematic review and meta-analysis.

Abstract

Total shoulder arthroplasty (aTSA) and reverse shoulder arthroplasty (rTSA) are commonly used to treat shoulder joint diseases, but differences in clinical outcomes and complications remain unclear. This study aimed to compare the efficacy and safety of aTSA versus rTSA. We systematically searched PubMed, EMBASE, and the Cochrane Library, identifying 27 studies for meta-analysis. Random-effects models were used to pool data. Odds ratios (ORs) with 95 % confidence intervals (CIs) were calculated for binary outcomes, and mean differences (MDs) with 95 % CIs for continuous outcomes. Forest plots, funnel plots, and Egger's test were employed to assess effect sizes and publication bias. Subgroup and sensitivity analyses explored sources of heterogeneity and result stability. A total of 195,413 participants from 27 studies were included. aTSA showed superior outcomes in active abduction ≥100° (SMD = 0.28, 95 % CI: 0.10-0.46), active forward flexion ≥120° (SMD = 0.17, 95 % CI: 0.01-0.33), external rotation ≥10° (SMD = 0.60, 95 % CI: 0.50-0.71), internal rotation score ≥4 (SMD = 0.40, 95 % CI: 0.25-0.54), and simple shoulder test ≥8 (SMD = 0.16, 95 % CI: 0.08-0.23). Regarding complications, aTSA had lower rates of failure (OR = 0.16, 95 % CI: 0.03-0.83), revision (OR = 0.71, 95 % CI: 0.55-0.91), and loosening (OR = 0.40, 95 % CI: 0.23-0.71), whereas rTSA was associated with fewer pulmonary embolisms (OR = 1.69, 95 % CI: 1.26-2.28), infections (OR = 1.55, 95 % CI: 1.09-2.20), fractures (OR = 2.09, 95 % CI: 1.03-4.07), and hematomas (OR = 1.96, 95 % CI: 1.56-2.54). Our meta-analysis indicates that, for shoulder degenerative diseases, aTSA generally offers superior postoperative efficacy compared with rTSA. aTSA was associated with lower rates of failure, revision, and loosening, while rTSA had fewer cases of pulmonary embolism, infection, fracture, and haematoma. Given the small effect sizes and low complication rates, these results should be interpreted cautiously. Nonetheless, they may aid clinicians in choosing the most suitable surgical approach to optimize patient outcomes.