CORR Insights®: No Differences in Outcomes Scores or Survivorship of Unicompartmental Knee Arthroplasty Between Patients Younger or Older than 55 Years of Age at Minimum 10-year Followup

Where Are We Now?

Medial unicompartmental knee arthroplasty (UKA) indications have provided an enduring source of controversy since the 1970s. Indeed, on this topic, orthopaedic surgeons generally can be divided into two groups. One group considers UKA a niche intervention, reserved only for the “ideal” patient, one with radiologically pristine lateral and patellofemoral compartments with the patient pinpointing their pain to the medial compartment. This group of surgeons follow the recommendations of Kozinn and Scott [5], who advised against offering UKA to patients who are too young, too heavy, or too active. The second group, typified by the Oxford group [3], suggests that the indications for UKA are broader, and can include patients who have substantial patellofemoral wear, a lateral osteophyte, or a degree of generalized knee pain. Surgeons in this camp might offer UKA to patients whose knees have correctable varus with full thickness medial cartilage loss and intact cruciates, regardless of their age, weight, or activity level [3].

Because of the advancements in implant design and polyethylene sterilization (in particular the introduction of ultra-high molecular weight polyethylenes and the discontinuation of the practice of sterilizing polyethylenes using gamma irradiation in air) wear has become a less-common cause of revision following UKA [1], and perhaps as a result, we have seen a growing body of evidence favoring the Oxford philosophy [3, 4, 6, 9, 10]. Successive studies have demonstrated that patellofemoral wear [4], lateral osteophytes [10], pain location [4], obesity [3], high activity levels [3], and now age [6], are unimportant in determining a patient’s suitability for UKA. This is supported by the finding that those who offer the broadest selection criteria obtain the best results with UKA [9].

Age is of particular interest. Younger patients have poorer long-term implant survival following all types of arthroplasty surgery, but this result appears to be most notable in UKA when observational data are considered [7]. The suspicion is that in nonspecialist practices, UKA has been seen as a stop-gap intervention for patients with partial-thickness cartilage loss who are too young or not ready for TKA. Such patients fare poorly whenever joint replacement is used [2]. The great strength of the current study by Lee and colleagues [6] is that robust and appropriate selection criteria were applied, with all patients having full-thickness cartilage loss on the medial side and with quite severe patellofemoral changes being tolerated. And thus, we can conclude that disease state is more important than age in identifying those who will do well after UKA.

Where Do We Need To Go?

Despite the growing evidence that favors the use of broad indications in UKA [3, 4, 9, 10], gaps in our knowledge still remain. It has previously been shown that surgeons who perform a small proportion of UKAs relative to their overall knee arthroplasty practice (and whom we can infer have narrow indications for UKA) have worse implant survival compared to those who perform a higher proportion (and hence have broader indications) [9]. However, the idea that extending UKA indications to a population with proportionally fewer “ideal” patients will lead to better results appears contradictory. Surgeons are unlikely to respond to poor results of any procedure in their hands by doing it in more patients. Most of the studies [3, 4, 10] on broadening indications for UKA are cohort studies from high-volume, expert surgeons. It could be suggested that their expertise leads to good results in spite of, rather than because of, their broad indications.

The more logistically difficult group to study appears to be surgeons who perform few UKAs. Why do these surgeons report such poor results when they would appear to have the strictest indications? Around a quarter of all surgeons who perform UKA in the United Kingdom will implant only one UKA per year [8]. Are these surgeons adhering to the strictest of indications as per the teaching of Kozinn and Scott [5] or, perhaps just as likely, are they picking the young patients with partial thickness disease only apparent on arthroscopy or MRI? Are the poor results reported by this group of surgeons caused by technical errors relating to a lack of experience, or are they simply operating on patients who are destined to fare poorly with the intervention? If the indications are narrow but correct, can a surgeon achieve good results despite performing fewer UKAs per year than is considered optimal on the basis of the published literature?

To those of us who believe that UKA confers a benefit to patients that is greater than what is possible with TKA, these are important questions. Is achieving a high caseload an end in itself, or is it simply a reflection of the fact that the surgeon understands the indications and is competent at performing the procedure? If a surgeon is implacably opposed, for instance, to offering UKA to patients with a degree of wear of the patellofemoral joint, (s)he is unlikely to achieve the high levels of usage associated with the best results in observational studies. However, if this surgeon is technically excellent and only offers UKA to patients with bone-on-bone disease, should (s)he be discouraged from offering UKA on the basis of not doing enough such procedures each year to be proficient?

How Do We Get There?

These issues boil down to two questions: (1) Why do UKA procedures performed by surgeons who offer UKA to only a small proportion of their patients yield poor results in registry studies? (2) How might it be possible for a surgeon with narrow indications to achieve good results with UKA?

Although large observational datasets have told us much about this subject, answering these questions will require more detailed study of the indications, pre and postoperative radiographs, and the results of surgeons who perform few UKAs. Gathering this information will be a challenge. Perhaps a place to start would be within a large institution that has a number of surgeons with different indications contributing to a shared database. For the first question, we need to determine whether these surgeons are operating on the wrong patients; whether their lack of experience with UKA results in technical errors; or whether they are choosing to revise UKAs earlier than a high-volume surgeon might. For the second question, we need to study those surgeons—identified from joint registries—who obtain excellent results in spite of operating on a small number of patients each year.

The suspicion is that the main determinant of success for these patients is the severity of the disease being treated. The best way to obtain reliable results with UKA, as was the case in in the current study, is to operate only on those with severe anteromedial disease. This is likely to be far more important than the age, weight, or activity level of the patient.