ABSTRACT
BACKGROUND: Long-term survivorship and accurate characterization of revision reasons in unicompartmental knee arthroplasty (UKA) are limited by a lack of long-term data and standardized definitions of revision. The aim of this study was to identify survivorship, risk factors, and reasons for revision in a large cohort of medial UKAs with long-term follow-up (up to 20 years). METHODS: Patient, implant, and revision details for 2,015 primary medial UKAs (mean follow-up, 8 years) were recorded following systematic clinical and radiographic review. Survivorship and risk of revision were analyzed using Cox proportional hazards. Reasons for revision were analyzed using competing-risk analysis. RESULTS: Implant survivorship at 15 years was 92% for cemented fixed-bearing (cemFB), 91% for uncemented mobile-bearing (uncemMB), and 80% for cemented mobile-bearing (cemMB) UKAs (p = 0.02). When compared with cemFB, the risk of revision was higher for cemMB implants (hazard ratio [HR] = 1.9, 95% confidence interval [CI] = 1.1 to 3.2; p = 0.03). At 15 years, cemented implants had a higher cumulative frequency of revision due to aseptic loosening (3% to 4%, versus 0.4% for uncemented; p < 0.01), cemMB implants had a higher cumulative frequency of revision due to osteoarthritis progression (9% versus 2% to 3% for cemFB/uncemMB; p < 0.05), and uncemMB implants had a higher cumulative frequency of revision due to bearing dislocation (4% versus 2% for cemMB; p = 0.02). Compared with the oldest patients (≥70 years), younger patients had a higher risk of revision (<60 years: HR = 1.9, 95% CI = 1.2 to 3.0; 60 to 69 years: HR = 1.6, 95% CI = 1.0 to 2.4; p < 0.05 for both). At 15 years, there was a higher cumulative frequency of revision for aseptic loosening in these younger groups (3.2% and 3.5% versus 2.7% for ≥70 years; p < 0.05). CONCLUSIONS: Implant design and patient age were risk factors for revision of medial UKA. The findings from this study suggest that surgeons should consider using cemFB or uncemMB designs because of their superior long-term implant survivorship compared with cemMB designs. Additionally, for younger patients (<70 years), uncemMB designs had a lower risk of aseptic loosening than cemFB designs at the expense of a risk of bearing dislocation. LEVEL OF EVIDENCE: Prognostic Level III . See Instructions for Authors for a complete description of levels of evidence.
