Article Text


Anterior knee pain and patellofemoral osteoarthritis: what should we do?
  1. C Niek van Dijk
  1. Correspondence to Professor C Niek van Dijk, Academic Medical Center Amsterdam, Amsterdam 1105AZ, The Netherlands; C.NiekvanDijk{at}

Statistics from

With primary total knee arthroplasty, there are three schools-of-thought about the patella: always resurface, never resurface or selectively resurface (bearing in mind the patient's articular cartilage, and patellofemoral congruence at the time of surgery).1 In Scandinavia, about 2% of patellas are resurfaced (according to knee arthroplasty registries), while in USA more than 90% of patients get a patellar button. Is this a genuine difference of opinion? Or is resurfacing the defensive option in a litigious climate, where a persistently painful knee might end up in court? A recent meta-analysis of 16 randomised controlled trials has shown no difference between resurfaced or non-resurfaced groups, as regards anterior knee pain and functional scores.2 We do not, it seems, yet understand why some patients experience anterior knee pain and others do not, regardless of their preoperative peripatellar conditions and patella resurfacing. Meanwhile, always resurface is doing a lot of unnecessary surgery.

But what about isolated patellofemoral arthritis? Such patients present a particular difficulty. If conservative treatment proves unsuccessful, we have various other options, such as realignment procedures, arthroscopic debridement, patellofemoral arthroplasty (PFA), total knee arthroplasty (TKA) and patellectomy. Recently, and because of second-generation prostheses, there has been renewed interest in patellofemoral joint replacement, for which there are good mid-term results. But it remains unclear, whether an isolated PFA does offer any advantage over a TKA? In this issue, Vasta et al3 make a systematic comparison between PFA and TKA.

The overall survivorship rate of inlay design PFAs is 82%, with a mean follow-up of 89 months (all figures rounded). The survivorship rate of the latest generation onlay design PFAs is 95%, with a mean follow-up of 46 months. This seems to be an improvement, and compares favourably with TKA, but we still lack the requisite long-term data.

In those patients where a PFA has been deemed correct, a major concern remains the later progression to all-compartment knee osteoarthritis. For the first generation of inlay design PFAs, progression to all compartment knee osteoarthritis occurred in 8% of knees. These patients would have been better off with an initial TKA instead of a PFA.

Initial patient selection is important here, as are factors which affect the disease's progress. Patellofemoral osteoarthritis develops faster, for example, where there had been a pre-existing trochlear dysplasia as its underlying cause.4

For these reasons, TKA remains the ‘gold standard’, when operating on isolated patellofemoral joint osteoarthritis. This seems confirmed by Vasta et al's3 survey, at least for older patients. For younger patients, however, where more is demanded of the functioning knee, PFA remains a favourable option. Preserving the femorotibial compartments and cruciate ligaments will improve the gait and knee kinematics. And, overall, PFA is less invasive, quicker in the OR and faster for rehabilitation.

Having said all this, the literature about PFA remains scarce, compared with the TKP resurfacing versus non-resurfacing debate. What can we learn from the latter? Several studies have shown that patellar denervation improves the visual analogue scores for pain, and the knee society scores. As for knee function scores, patellar denervation performs even better than patellar resurfacing5

For younger patients with isolated patellofemoral osteoarthritis, the best solution might well be patellar debridement with denervation, combined with a trochlear prosthesis. This would obviate maltracking, overstuffing and other patellar implant complications. To the best of my knowledge, however, this combination has not been attempted.

At this stage, I am afraid, we need more trials, more long-term evidence, etc.


View Abstract


  • Competing interests None declared.

  • Provenance and peer review Not commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.