Humeral Stem Loosening Following Reverse Shoulder Arthroplasty: A Systematic Review and Meta-Analysis.

Abstract

Aseptic stem loosening following reverse shoulder arthroplasty (RSA) is an uncommon complication. The timing and the factors contributing to aseptic stem loosening remain poorly understood. We performed a systematic review that identified 75 articles; 65 of the included articles were case series (Level-IV evidence), 8 were cohort studies (7 Level III, 1 Level II), and 2 were prospective randomized controlled trials (Level I). A meta-analysis of 1,660 cemented and 805 uncemented RSA stems was performed. We compared the rates of aseptic stem loosening, humeral radiolucent lines, and revision for stem loosening (1) between cemented and uncemented stems in cohorts with short and long mean follow-up periods (<5 and ≥5 years) as well as in all cohorts combined, and (2) among different etiological subgroups. Aseptic stem loosening occurred more commonly in the cohort with long compared with short follow-up time (2% versus 0.8%, p = 0.01). When comparing cemented with uncemented stems, there was no significant difference in the rates of aseptic stem loosening or revision for stem loosening in either the short or long-term follow-up groups. Humeral radiolucent lines were more common with cemented compared with uncemented stems (15.9% versus 9.5%, p = 0.002). Analyzed by etiology, the highest rate of aseptic stem loosening occurred in the tumor subgroup (10.8%), followed by RSA as a revision procedure after a failed arthroplasty (3.7%). No stems in the acute fracture or fracture sequelae subgroups developed aseptic stem loosening. Aseptic stem loosening occurred more commonly in cohorts with longer follow-up. There were no differences in the rates of aseptic stem loosening or revision for stem loosening between cemented and uncemented stems. Patients treated with RSA following excision of proximal humeral tumors and RSA as a revision procedure after a failed arthroplasty were at greater risk of aseptic stem loosening. Therapeutic Level IV. See Instructions for Authors for a complete description of levels of evidence.