In this section, we first summarise the 20-year and longer results of the Oxford Knee. These results relate primarily to the Phase 1 and Phase 2 Oxford Knee because the Phase 3 was introduced only in 1998. We then summarise the 10 and 15-year results of the Phase 3 Oxford Knee. There are numerous short term studies of the Oxford Knee. We will not discuss these in depth but we will present the results of a meta-analysis of all published Phase 3 studies (Hamilton et al, 2016).
Dr Ulf Svard is a surgeon who works in Skövde in Sweden. In 1983, he visited Oxford and learned how to use the Oxford Knee. He has been using it since then and following his patients. Initially a group of four surgeons did the operations but more recently he has done them alone so the majority of the patients had their surgery done by him. Currently he has implanted Oxford Knees in more than 1100 patients (mean age 69). At the time of his various reviews, none were lost to follow up. His 20-year data was first presented in 2006 (Price & Svard, 2006). At that stage, there were 638 OUKA and the 20-year survival was 92% (95% CI 77 – 100). At 10 years when 187 patients were reviewed, 90% had good or excellent HSS scores. This data was updated and subsequently published in 2011 (Price & Svard, 2011) with a 20-year survival of 91%. There had been 29 revisions, 10 for lateral arthrosis, nine for component loosening, five for infection, two bearing dislocations and three for unexplained pain.
The next review was based on the first 1000 implants (125 Phase 1, 271 Phase 2, and 604 Phase 3), and at 20 years the survival rate was 87% (95% CI 79 – 95). The survival was also 87% at 22 years. Dr Svard also reviewed his 125 Phase 1 implants (Svard, 2012). These had been implanted between 1983 and 1988. At the time of review, 80% were dead and the remainder were reviewed with an average follow up of about 25 years. At time of death, or last review, 90% had not been revised and had a good or excellent Hospital for Special Surgery (HSS) Score. This demonstrates that, when used correctly with the correct indications, the Oxford Knee could be considered to be a definitive knee replacement. We are not aware of any other implant, either unicompartmental or total, that has achieved as good, or better, results.
At 20 years, about 2% of cases had failed from progression of the disease in the lateral compartment (Price & Svard, 2011). It is generally believed that arthritis will inevitably progress in the lateral compartment after medial unicompartmental replacement. The very low incidence of progression demonstrates that this is not the case and that progression should be considered to be a rare event. The incidence of progression was higher in the Phase 2, than the Phase 1, OUKA. Analysis of the post-operative radiographs demonstrated that, following Phase 2, the knees were in slightly more valgus than Phase 1. The Phase 2 tibial resection guide had an extension that passed around the medial side of the tibia (Fig. 10.11). To insert this, a small medial release was required. In contrast, the Phase 1 did not have this extension so no medial release was required. It was thought that this medial release might have contributed to the higher incidence of lateral compartment arthritis.
Figure 10.11 Plan view of OUKA (Phase 2) tibial saw guide.
Barrington and Emerson (2010) also presented 20 year results of the Oxford Knee and in this study, which was originally started as part of an Investigational Device Exemption (IDE) study in the USA for the Phase 2, there were 54 knees in 48 patients. At 20 years, the survival was 85%. Nine knees in seven patients had been revised, six for disease progression. As part of the surgical technique, a medial release was undertaken to allow for insertion of a retractor. This may have contributed to the incidence of progression of lateral compartment arthritis.
The 15-year results of the first consecutive 1000 cemented medial Oxford Phase 3 UKA implanted by two designer surgeons have been reviewed with a mean follow up of 10 years (Pandit et al, 2015). At 10 years, the mean Oxford Knee Score (OKS) was 40 (SD 9) with 79% of the knees having excellent or good outcome. There were 52 implant related re-operations at a mean of 5.5 years (range 0.2 – 14.7 years). Progression of arthritis in the lateral compartment (2.4%) followed by bearing dislocation (0.7%) and unexplained pain (0.7%) were the most common indications for revision. When implant-related re-operations are considered failures, the 15-year survival is 91% (95% CI 83 – 98). When the end point is revision for failure of the tibial or femoral components, the 15-year survival is 99% (95% CI 96 – 100). When revision requiring TKA components are considered failures, the 15-year survival is 99.7% (95% CI 98 – 100). There were no cases of a revision performed for wear, progression of PFJOA or bearing fracture. When all implant-related re-operations are considered to be failures with 3 or 4 mm bearings (n = 712), the 15-year survival was 94.3% (95% CI 87.1 – 100), with 5 mm or greater (n = 86) 15-year survival was 75.0% (95% CI 28.5 – 100). Although the reasons for these differences are not fully understood, it seems sensible to aim for a 3 or 4 bearing. If this is achieved, the 15-year survival is as good as the best TKA.
