The aim of this thesis was to establish the long-term outcomes of the Oxford medial Unicompartmental Knee Arthroplasty (OUKA), define patient selection criteria and to develop and externally validate an evidence based method of patient selection for this procedure. In the hands of the developer surgeons, outcomes following medial OUKA were found to be good with an implant survival of 94% (95%CI 92 to 96) at ten-years and 91% (95%CI 83 to 98) at fifteen-years. Across the published literature, however, variation in outcomes was observed with a meta-analysis of published series of OUKA finding estimates of ten-year survival ranging from 57% to 100%, mean 88% (95%CI 85 to 90). It was identified that both increased surgical caseload (volume) and increased surgical usage (proportion of primary knee arthroplasty that are OUKA), a surrogate marker of indications, were associated with improved outcomes. Surgical usage, however, was more important, with good results following OUKA seen with high surgical usage, representing broad indications, independent of the surgical volume. This finding, coupled with differences in patient demographics and failure mechanisms between usage groups, highlighted that differences in indications for OUKA may explain the variability in outcomes observed. One reason surgeons may have a low usage is if they apply previously recommended patient factor contraindications based on age (<60 years), weight (&GE;82kg) and activity level (high activity). When disease factors are standardised, however, it was found that patients with these previously reported contraindications often actually did better than those without, and outcomes of knees implanted where all these factors were present were as good as where none were present. Therefore, the decision to proceed with OUKA should be based on the pathoanatomy of disease. Optimal candidates for OUKA should have full-thickness cartilage loss, with bone on bone arthritis, in the medial compartment, as knees with partial thickness cartilage loss were found to have worse functional outcomes and almost three-times the reoperation rate, predominantly for unexplained pain. Provided there was full-thickness preserved cartilage laterally and functionally normal ligaments, the presence of lateral osteophytes and the macroscopic status of the anterior cruciate ligament was not found to influence outcomes, nor did the presence of patellofemoral joint disease (with the exception of lateral facet disease with bone loss and grooving) or anterior knee pain. The pathoanatomy of disease can be identified radiologically, however, standing knee radiograph were found to perform poorly. To identify medial compartment full-thickness cartilage loss either a varus stress radiograph or fixed flexion radiograph, both at 20° flexion and aligned to the joint surface, were identified as the optimum views. To confirm preserved lateral compartment full-thickness cartilage a valgus stress radiograph at 20° flexion, aligned to the joint surface, was identified as the most appropriate technique. As stress radiographs are time and resource consuming, a novel stress device was developed in line with the IDEAL-D framework and validated against the gold standard of manual, clinician performed stress radiographs, as well as independently tested in clinical practice. Finally, to simplify patient selection, an atlas based Decision Aid, combined with a structured radiographic assessment, was developed and externally validated with an accuracy of over 90% at identifying suitability for OUKA. The routine use of this approach would be expected to standardise patient selection and ultimately translate into improved long-term outcomes.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:748795 |
| Date | January 2017 |
| Creators | Hamilton, Thomas |
| Contributors | Murray, David ; Mellon, Stephen ; Pandit, Hemant |
| Publisher | University of Oxford |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://ora.ox.ac.uk/objects/uuid:6a36e359-defb-4ff3-a6b5-6f459c4c40f2 |
Page generated in 0.0019 seconds