Why do primary reverse total shoulder arthroplasties fail today? A systematic review and meta-analysis

Reverse total shoulder arthroplasty (rTSA) offers good clinical outcomes, with varying complication rates. Despite the technological advances and increasing popularity of shoulder arthroplasty in the last 2 decades, a comprehensive understanding of the specific causes for failure after primary rTSA remains limited.

This PROSPERO-registered systematic review and meta-analysis was conducted following Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. A literature search was performed on PubMed, Embase, and Cochrane covering studies from 2014 to 2024 to identify studies reporting outcomes in primary rTSA. Data extraction focused on failure modality and subgroup analyses was performed. Proportional meta-analysis was performed using random-effects models, and data were reported in estimated proportions with 95% confidence intervals (CIs). It is important to emphasize that data consist of a proportion of failures, not failure rates of rTSA.

A total of 42 articles involving 69,222 rTSA procedures were identified, and 2,086 failures were included. Instability was the most common failure mode, which accounted for 21.4% of failures (95% CI [15.2%-29.2%]). Aseptic loosening accounted for 19.4% (14.3%-25.8%), followed by infection (18.6% [14.1%-24.0%]) and periprosthetic fracture (10.6% [7.0%-15.7%]). Of the total aseptic loosening failures, glenoid loosening accounted for 64.9% (43.9%-81.4%), and humeral loosening accounted for 31.4% (16.6%-51.4%) of the failures. Failures in studies with short-term follow-up (859 failures) were most commonly due to instability (24.9% [16.4%-35.9%]), followed by infection (21.8% [16.8%-27.8%]) and aseptic loosening (20.2% [11.7%-32.7%]). Failures in studies with long-term follow-up (286 failures) were most commonly due to aseptic loosening (24.1% [19.5%-29.4%]), followed by infection (21.5% [11.2%-37.1%]) and periprosthetic fracture (16.9% [7.5%-34.0%]). Failures for onlay humeral stems (66 failures) were most commonly due to instability (36.7% [20.5%-56.4%]), whereas failures for inlay stems (70 failures) were most commonly due to aseptic loosening (32.8% [13.6%-60.3%]). Failures for cemented humeral stems (20 failures) were most commonly due to periprosthetic fracture (30.0% [14.1%-52.7%]), whereas failures in cementless humeral stems (257 failures) were most commonly due to aseptic loosening (21.2% [8.6%-43.7%]).

This study identified instability (21.4%) being the most common cause of failure after primary rTSA, followed by aseptic loosening (19.4%) and infection (18.6%), and that differences exist for duration of follow-up, time of publication, implant design, and use of cement, although some subgroup analyses were limited by number of failures. These findings emphasize the need for more research focused on mitigating these predominant causes of failure to improve long-term outcomes.