Abstract
While no one would argue the necessary role for the medical management of patients with early knee arthritis, significant controversy remains regarding the ideal treatment for a patient with bone-on-bone osteoarthritis who could equally be treated with a high tibial osteotomy, a unicompartmental total knee, potentially a patello-femoral replacement if dealing with isolated patello-femoral disease or lastly, a complete total knee replacement. While clearly to date there has not been consensus on this issue, a review of the arguments, both pro and con, should be used as a guide to the surgeon in making this clinical judgment.
Many ardent supporters of unicompartmental knee replacements espouse one of the principle benefits of the uni knee as much greater patient satisfaction. Unfortunately, what is never taken into account is the pre-selection bias that occurs in this patient population. Patients with the most minimal amount of arthritis and those with the greatest range of motion are pre-selected to undergo a unicompartmental knee replacement compared to the more advanced arthritic knee with mal-alignment and more significant pre-operative disability that will undergo a total knee replacement. Additionally the sources of data to draw the conclusions must be carefully analysed. We must avoid using data from small series with unblinded patients performed by surgeons expert in the technique. Instead registry data, with its broad based applicability, is a much more logical source of information. Of significance, when over 27,000 patients were assessed regarding satisfaction following knee surgery; there was no difference in proportions of satisfied patients whether they had a total knee or a unicompartmental knee.
Once again large prospective cohort data in the form of arthroplasty registries strongly favors total knee arthroplasty over unicompartmental knee arthroplasty. The Swedish Knee Arthroplasty Registry demonstrated higher revision rates with uni's as compared with total knee replacements. In the Australian Joint Replacement Registry the cumulative 11 year percent revision rate for primary total knee replacements is 6.1% and for unicompartmental knee replacements is 16.3%. The 10 year cumulative percent revision rate for patella femoral replacements 29.9% and lastly bicompartmental replacements are at 10.3% after only 3 years. Higher failure rates in unicompartmental knee replacements seen in Australia has correlated to a significant decrease in the number of uni's being performed, which peaked at 14.6% in 2003 and in 2012 has reduced to 9.0%. There is a direct correlation to age, with younger patients having a significantly higher percentage of revision following unicompartmental knee replacements (25% failure rate at 11 years if less than 55 years old). There is also tremendous variability in the success rate of the uni in the Australian Registry depending on the implant design (5 year cumulative revision rate range 5.0% to 18.9%), which is simply not seen in the total knee replacement population (5 year cumulative revision rate range 1.6% to 7.7%).
While one can perform the philosophical exercise of debating the merits of a total knee versus unicompartmental knee, the evidence is overwhelming that in the hands of the masses a total knee replacement patient will have equal satisfaction to a unicompartmental patient, and will enjoy a much lower probability of revision in the short term and in the long term.
