Patella Resurfacing: Rarely, If Ever Necessary

This week’s Orthopaedic Crossfire® debate was part of the 35th Annual Current Concepts in Joint Replacement® (CCJR®), Winter meeting, which took place in Orlando. This week’s topic is “Patella Resurfacing: Rarely, If Ever Necessary.” For is Robert L. Barrack, M.D., Washington University School of Medicine, St. Louis, Missouri. Opposing is Steven B. Haas, M.D., Hospital for Special Surgery, New York, New York. Moderating is Robert E. Booth, Jr., M.D., Jefferson Health 3B Orthopaedics, Philadelphia, Pennsylvania.

Dr. Barrack: Anterior knee pain is very common. Maybe the most common symptom after knee replacement. The misconception is that it’s from an unresurfaced patella. In my experience, it is far more common that this patient has a resurfaced patella that isn’t quite articulating well. Knee pain—this is something we see day in and day out and it’s usually with a resurfaced patella.

The bigger problem is when you go beyond just having asymmetric loading and anterior knee pain to where you get fragmentation of the patella. You look at the AP x-ray, the coronal alignment looks good. Something’s not quite right and because of that you get loosening on one side and fracture and fragmentation on the other side. This is the type of thing we see with regularity. These are very serious complications and the knee is never quite the same.

On CT, the knee is a little malrotated, but a fragmented, fractured patella—I see this on a regular basis.

The data I see on resurfacing studies is varied and contradictory. Dozens of studies say not resurfacing is equivalent, occasionally even better. Other studies say that routine resurfacing is equivalent or better. But they are usually underpowered, so you have to go to meta-analyses and registry data. If you look at the meta-analyses, you find that the reoperation rate is higher if you don’t resurface the patella. The incidence of anterior knee pain is contradictory.

The problem is the meta-analyses include old components that don’t do well with the non-resurfaced patella. Registry data says about the same. The reoperation rate is higher when the patella is not resurfaced. The clinical results of a subsequently resurfaced patella are generally unpredictable because there is an underlying problem. If you have malrotation, resurfacing the patella isn’t going to solve it.

Clinical results are similar. Indications for revision are unclear. If the patient has diffuse pain and you resurface their patella, it’s going to make your revision rate look higher, but they’re actually just problems with the indications.

What about the resurfaced patella? The problem is all the complications are not reported to a registry because they’re not implant related, in that you’re not putting in an implant. Usually you’re just taking out an implant. And the results vary by type.

A paper from the Mayo Clinic showed that lateral facet pain can require revision so they looked at 15-20 cases, but 99% of surgeons are not aware of this option (Nikolaus, et al., JOA). They don’t perform this procedure. If I re-operated on everyone who had lateral facet pain, the reoperation rate would go through the roof.

The second most commonly implanted total knee in the Swedish Knee Registry had the lowest risk ratio—0.44. But only when it was not resurfaced. That’s half the revision rate of all the knees. The best performing implants in the Swedish Knee Registry, over 90% of the knees, are not resurfaced.

So, if you are concerned with the incidence of reoperation and anterior knee pain being higher, why not resurface routinely? Well, a lot of things can go wrong. And these things all do go wrong – loosening, fracture, fragmentation, AVN, restricted motion. Ritter’s group published a paper which documented fragmentation in 4% over three years (Clin Orthop, 2001). Oblique resurfacing…20 years ago they reported at the Mayo Clinic 7% of cases have oblique resurfacing, half required knee reoperations (Pagnano, et al., Clin Orthop, 1999).

You can make it too thick; you can make it too thin.

The problem is it’s the last step in a total knee. The instrumentation is variable. And the complications are underreported. There are a lot of advantages to not resurfacing the patella. It’s faster. It’s less expensive. Lower risk of major complications. If a problem does develop you have a lot more options. End results are very similar.

A little know trend in meta-analyses and registry reviews that I’m sure you’ll hear from Steve Haas, is that there is tremendous variability and discretion involved in the methodology. Studies have shown that the results agree with the practice patterns of the authors over 100% of the time. It’s best to rely on impartial sources of data, like the AAOS Clinical Practice Guidelines. And they say that there is strong evidence that pain and function are the same. There’s moderate evidence that the reoperation rate is higher, but that’s only because the reoperation rate for resurfacing is underreported.

Now in the Swedish Registry, at 10 years, the reoperation rate is higher in resurfaced patellae (Fraser, et al., JOA, 2017).

So, in conclusion, this continues to be a highly debated topic. You can get good results from either. U.S. is an outlier in the world. Most surgeons resurface the patella. Outside the U.S. most surgeons do not. The difference, if it exists at all, is very small and I predict that the fastest trend that we’ll see in the next decade is cementless knees without patella resurfacing.

Dr. Haas: No resurfacing of the patella is a great idea, if you like to re-operate on your patients. If I re-operate on a patient, they’re mad at me because they said, “Why didn’t you get rid of the arthritis you said was there? It was bone-on-bone. You said that was bad. Why didn’t you get rid of it?”  So, they have lawyers climbing up my butt. Or at least if they don’t have a lawyer, they’re mad at me. And I don’t like to have patients mad at me.

So, what are the data? First of all, the cartilage may not be normal in the first place. Cushner showed that there are abnormal pathologic changes in the cartilage. Patella resurfacing, in data, shows that the rate of complications is actually pretty low. And you may recognize some of those authors (Boyd, et al., JBJS, 1993; Barrack, et al., JBJS, 2001; Erak, et al., Clin Orthop, 2009).

Not resurfacing the patella is associated with a higher revision rate. That’s just clear. Patients just don’t like being revised, and the pain is about equal or more. So, I’ll give you that it’s equal. However, if it’s equal, then it may have to get operated on more. Knees don’t get better when you re-operate on them.

Almost all meta-analyses show a similar thing. The non-resurfaced get re-operated on more and they have anterior knee pain in both groups. (Nizard, et al., Clin Orthop, 2005).

In most of the data…14 studies, non-resurfaced patellae have a greater incidence of anterior knee pain. Non-resurfaced patellae resulted in less patient satisfaction (Parvizi, et al., Clin Orthop, 2005).

Ten studies, 1,223 knees, again, patellar resurfacing reduces risk of reoperation and anterior knee pain.  (Pakos, et al., JBJS Am, 2005).

Reoperation rate significantly higher in non-resurfaced group. Resurfacing 1.9% compared to 4% non-resurfacing. (Fiddian & Murray, KAT Group, Interim KS 2009)

In fact, a study by Dr. Barrack—patellar resurfacing versus non-resurfacing. No significant difference in knee pain, range of motion, patella function. And they didn’t even have a preference in bilateral knees (Barrack, et al., JBJS, 2001).

If you look at this large study of 6,000 patients, the incidence of patellofemoral pain is higher in the non-resurfaced group than the resurfaced group (Tubutin, et al., KSSTA, 2005).

Same again, anterior patella resurfacing, less anterior knee pain, less patella-related reoperation. And, in fact, lower satisfaction scores when the patella is not resurfaced (Calvisi, et al., AOTS, 2009).

We go through the Australian Orthopaedic Association National Joint Replacement Registry (Clements, et al., Acta Orthop, 2010) you’re going to see the same thing. I think this is one of the more well studied entities in orthopedic surgery.

In conclusion, it’s really pretty simple. People hate being re-operated. Since our primary goal is relieve pain and there’s a higher incidence of anterior knee pain and more reoperations, I favor to resurface the patella, except in selected cases.

Moderator Booth: Robert, it says in your abstract that St. Louis is now the largest American city with unresurfaced patellae. You’ve obviously persuaded the local community.

Dr. Barrack: I’ve only been in St. Louis for 15 years, Leo Whiteside has been there 120 years (laughter), so I have to credit him. He’s influenced some high-volume surgeons in the community that never resurfaced a patella, so we have a data set of tens of thousands of patients. I’ve never had to subsequently resurface one of Leo’s knees, or his former fellows, who do unresurfaced patellae.

As far as why Americans are behind, I think it’s the I-95 axis which is the road between Boston, New York and Philadelphia. Seriously, many of our leading surgeons were brainwashed…I mean trained in New York or Boston or Philadelphia. You’re partly to blame Dr. Booth. (laughter)

Moderator Booth: I accept that with honor.

Dr. Haas: There’s a Philly axis there. The issue though is I tried it, I said I’m not going to resurface the patella and I had patients who were not happy. The ones that I re-operated on were not happy campers.

Moderator Booth: So, everybody’s talks now begin with a point that 28% of patients or something approaching that are unhappy with their total knee. It’s because of anterior knee pain. And that’s resurfaced or not, correct? What percentage of that embarrassing number do you think are from the patella and what are from other issues. Things we can’t define?

Dr. Haas: I actually think they are getting it from the patella, whether resurfaced or unresurfaced. There probably are better ways to resurface the patella, but that doesn’t mean that throwing away resurfacing is necessarily the answer. Maybe there are better ways to do it.

Moderator Booth: What percentage of your anterior knee pain would you say are unresurfaced?

Dr. Haas: About 10%.

Moderator Booth: To my eye a bunch of those cases you showed had internally rotated femurs. Over a year, 10-15% have the patellar button on the lateral side of the patella. It seems to be such a simple thing to do, to cut an asymmetrical piece away to be left with a symmetrical piece, but I think it’s something we don’t do well for some reason.

Dr. Barrack: Most surgeons do it freehand with a saw. They take one try and this is what happens. The thing is if you’re going to mal-rotate the components, wouldn’t you rather have a fragmented, fractured patella to work with when you do your revision?

Moderator Booth: Why do you think we’re blaming the unresurfacing and the admittedly lamentable history of metal-on-cartilage, uni spacers and hip protheses? Why do you think it is that when you replace a patellar button or put one on and the knee is painful, they don’t get better?

Dr. Haas: I think because the loads in the patella are so high. Knee replacements probably load the patella more than the natural knee. So, some of them, I think, are just biomechanical. I’m not exactly sure why they get better. But most of them…

Dr. Barrack: They don’t get better because there is something wrong with the underlying knee replacement. We talk about the patella being like the canary in the coal mine. If you have anterior knee pain, there’s a problem in the knee. And it’s not the patella.

Moderator Booth: I know when I see a painful anterior knee, the first thing I look at is things other than the patella itself.

Dr. Barrack: Frequently you have sagittal plane laxity that will increase the loads, or you have an internally rotated component, but it’s a problem with the underlying knee, not the patella.

Please visit www.CCJR.com to register for the 2020 CCJR Winter Meeting — December 9 – 12 in Orlando, Florida.