Abstract
Introduction
Unicompartmental knee arthroplasty (UKA) offers significant advantages over total knee arthroplasty (TKA) but is reported to have higher revision rates in joint registries. In both the New Zealand and the UK national registry the revision rate of cementless UKR is less than cementless. It is not clear whether this is because the cementless is better or because more experienced surgeons, who tend to get better results are using cementless. We aim to use registry data to compare cemented and cementless UKA outcomes, matching for surgical experience and other factors.
Methods
We performed a retrospective observational study using National Joint Registry (NJR) data on 10,836 propensity matched Oxford UKAs (5418 cemented and 5418 cementless) between 2004 and 2015. Logistic regression was utilized to calculate propensity scores to match the cemented and cementless groups for multiple confounders using a one to one ratio. Standardised mean differences were used before and after matching to assess for any covariate imbalances. The outcomes studied were implant survival, reasons for revision and patient survival. The endpoint for implant survival was revision surgery (any component removal or exchange). Cumulative patient and implant survival rates were determined using the Kaplan-Meier method. Patients not undergoing revision or death were censored on the study end date. The study endpoints implant and patient survival were compared between cemented and cementless groups using Cox regression models with a robust variance estimator.
Results
The 5-year implant survival for cemented and cementless Oxford UKA were 95.4% (95%CI 94.6–96.1%) and 96.5% (95%CI 95.8–97.1%) respectively. Implant revision rates were significantly lower in cementless Oxford UKA than cemented, HR 0.8 (CI 0.64–0.99); (p=0.04). The most common reasons for revision in the cemented Oxford UKA group were aseptic loosening (n=44, 0.8%), pain (n=37, 0.7%) and osteoarthritis progression (n=37, 0.7%) compared with osteoarthritis progression (n=28, 0.5%), pain (n=24, 0.4%), aseptic loosening (n=23,0.4%) in the cementless group. Patient survival 5-year survival rates for cemented and cementless Oxford UKA were 96.1% (95%CI 95.2–96.9) and 96.3% (95%CI 95.4–97.1) respectively and were not significantly different HR 0.91 (95%CI 0.71–1.15); (p = 0.42).
Conclusion
This is the first study comparing the outcomes of the cemented and cementless UKA from the largest arthroplasty register in the world. Our work shows the cementless Oxford UKA has superior implant survivorship to the cemented implant at 5 years follow up. Cementless implants also had half the risk of requiring revision for aseptic loosening, which may be related to the decreased incidence of tibial radiolucent lines with cementless fixation. Patient survival did not significantly differ between the implant types.
