Survivorship of the humeral component in shoulder arthroplasty.

Abstract

Loosening of the humeral component is rarely a cause for revision. This study was conducted to determine long-term survivorship of humeral components and investigate the risk factors associated with humeral component removal or revision. From 1984 to 2004, 1423 patients underwent 1584 primary Neer and Cofield shoulder arthroplasties. The Kaplan-Meier method was used to estimate implant survival. Cox proportional hazards regression was used to assess the effects of age, gender, etiology of the disease, surgery type (hemi vs total), fixation (cemented vs noncemented), and the humeral component design (Neer II, Cofield 1 or 2) with survival free of revision or removal of the humeral component. There were 108 revisions and 17 removals of the humeral component. Estimates of survivorship free of revision or removal of the humeral component for any reason was 94.8% (95% confidence interval [CI], 93.6-96.0) at 5 years, 92.0% (95% CI, 90.4-93.6) at 10 years, 86.7% (95% CI, 84.2-89.4) at 15 years, and 82.8% (95% CI, 78.5-87.5) at 20 years. Younger age, male gender, replacement due to posttraumatic arthritis, an uncemented component, and use of a metal-backed glenoid component increased the likelihood of humeral component failure. The need for revision of the humeral component is commonly related to glenoid or glenoid component issues. Patient and diagnostic factors play a role in implant survival; implant type and method of fixation are less important. Level 4; Case series, treatment study.