• Hip and knee surgery

The medial unicompartmental knee arthroplasty (UKA) was introduced in the 1970s, but has struggled with higher revision rates compared to the alternative; total knee arthroplasty (TKA). In the last decade its popularity in Denmark has grown in popularity as published evidence has increased our understanding of who benefit from it and which surgical practice factors influence the outcome. We have also seen an increased emphasis on outcomes other than revision drawing the attention to the many benefits of UKA compared to TKA. UKA now accounts for 20 % of the annual primary knee replacements performed in Denmark. The nationwide implementation of UKA as standard treatment for end-stage AMOA is the basis for this thesis.
The aim was to describe the UKAs performance in three scenarios; First, its development over time. Secondly, how it is performing during the implementation both nationally and at a unit level. Thirdly, to describe its performance in an optimized set-up, and give us an indication of what can be achieved if current recommendations for practice is followed.

Two propensity score matched registry studies on 20 years of registry data aimed to describe development in UKA revision risk over the last 20 years compared to TKA. They were designed to identify any changes in practice patterns which could be the cause of any potential changes to risk of revision, and to determine any correlation between these practice changes and changes to revision indication patterns for UKA. Two additional clinical comparative studies were performed on a unit level. They compared UKA patients to TKA patients with pre-operative radiographic AMOA. The first study monitored the UKAs performance during implementation, at a unit with no prior history of UKA usage, and compared it to that of TKA patients from prior to UKA implementation. Evaluating performance as; length-of-stay, complication- and readmission rates and revisions and joint specific patient reported outcome measurements; Oxford Knee Score and Forgotten Joint Score at 3, 12 and 24 months follow-up. Lastly a 2-centre propensity score matched study comparing UKA and TKA as strategies in their optimal set-up, using change in OKS at 1 year follow-up as out primary outcome, and determined difference in likeliness of reaching the patient acceptable symptom state. We included UKA patients from an experienced UKA design centre and TKA patients with AMOA on preoperative radiographs from an experienced TKA centre with no history of UKA usage. In total data was collected from almost 100 000 (10 000 UKA) primary knee replacements performed from 1997 to 2018 in Denmark and the UK.

We found a significant decrease in UKA revision risk over the last 20 years, and a corresponding reduction in difference between UKA and TKA revision rates. The decrease was correlated to an increase in high usage units and the use of cementless fixation for UKA, and was due to a decrease in revisions for pain and aseptic loosening. The implementation study found significantly better outcome for UKA compared to TKA. Lastly, we found larger than previously reported difference OKS between UKA and TKA when investigated as optimized strategies and significantly larger likeliness for UKA patients of categorizing themselves as being well one year after surgery.

In conclusion, we found UKA to be safe to implement without any significant performance drop. The revision rates for UKA are decreasing significantly nearing those reported for TKA. A change which is correlated to the resent changes in practice. The decrease is primarily caused by fewer revision on unspecific indication, showing that surgeons are adhering to evidence based practice. Lastly, we found patients are more likely to categorize themselves as well after receiving a UKA rather than a TKA.

Download afhandling