The Role of Total Knee Replacement in Unicompartmental Arthritis

Lecture from the 84th Annual Alumni Symposium: Unicompartmental Arthritis and the Patient Under 50

Richard S. Laskin, MD
Hospital for Special Surgery

I am going to talk about total knee replacement in uni-compartmental disease in younger patients and, just in case you are wondering how many younger patients we have seen, I pulled this data together. Look at the decade from the 80s to 90s and the 2000s (where we have had two years projected over 10), and look at the increasing number of people who are under 60 and 50 as those decades go along. Why is that? Is osteoarthritis becoming more rampant? No, it's because what we are seeing in more patients is what I call post-traumatic arthritis, and that is not only tibial plateau fractures. These are also people who have had total meniscectomies who all went out and played sports, who tore their anterior cruciate ligament, and with all due respect to the sports people, they got fixed, but their mechanics were not quite normal. Then they tore their medial collateral ligament again, and then there was the LCL, and eventually their knee goes, and their knee develops advanced arthritis at a much earlier age.

Most of these can be treated with NSAIDs, with aspirin, with injections, or varying sorts of physical therapy before you progress on to any sort of arthroplastic operation. But at some point, when you have a patient with advanced arthritis in their knee who is young, you have to think of some sort of arthroplasty, and it is usually unicompartment or tricompartmental total knees.

The major problem with both of those is polyethylene wear and loosening, and these can lead to recurrent pain, instability, and the need for revision. So why not choose a uni for these patients? Well, as you saw before in 1978, I published this paper in JBJS showing a 33% failure rate. In my first experience ever meeting Dr. Insall, I presented in a meeting at Albert Einstein, and he was presenting, and I said, "I am doing this study on uni's," and he looked at me, and he raised his eyebrows, and he said "Interesting." And I had no idea what he meant by that, but I found out very soon thereafter why it was interesting -- they didn't work. They really didn’t work at all, and when I performed this study, I inset the implants in the bone and I felt there weren’t many contraindications. I thought that it was great for most patients. I thought you really didn’t need any instruments; you just burred the bone out and stuck it in. And over the years, people have said, "Well, it’s a lousy study because you use thin poly, you burred into the bone, you didn’t use any instruments." Having said that, it is now 2002. Open up all the advertisers in the journal and they are replete with people using thin polys, insetting the implant in the bone, using few - if any - instruments other than dental bur, and I must tell you if you look at these implants they are the Marmor total knee from 1972. It is the same prosthesis. So there are some problems coming around, and we all know that if you don’t learn from your mistakes you make them again.

A couple of reasons why I wouldn't pick a uni:  Most of these patients that I see will have patello-femoral involvement, and this is a contraindication to my doing a uni. None of them would have this involvement if I did a total knee replacement.

How about the absence of an ACL? I have heard some people who do uni's say, "Well, it's no problem, you can do a uni in the absence of an ACL." But patients without ACL's in general don’t have normal knee kinematics. They tend to get earlier medial compartment osteoarthritis, and when you put a uni in, all you are doing is resurfacing the surface, so tell me why the uni wont wear out the same way that the normal cartilage wore out.

How about a patient with a fixed varus deformity? Some people say you can do uni's in these people - young patients - and that it’s not a problem. But you are really not correcting the fixed deformity. As Russ Windsor said, you are not doing ligament releases, you are being told to leave them a little bit of varus if you believe the Oxford people, and again, all you are doing is putting on a new surface. So won't that wear out the same way?

John Insall did a real landmark paper in the 70's that showed that if you want your arthroplasty to fail, don’t put it in the right position in the knee. Yet, we are being told in uni's that we can leave them in varus, and that somehow or another, it's going to be okay.

How about patients with inflammatory arthropathy? Well, everyone knows you don’t put it in someone’s rheumatoid arthritis. But how about patients with chondral calcinosis? That is inflammatory arthropathy on the other side of the knee. People say that's no problem. But why is it no problem? It's not a problem with total knee.

It is said that uni's need less blood, they get better faster, and their satisfaction is greater. That may be true if you compared uni's with traditional total knees with these foot long incisions. But more recently, since I learned how to do this operation from Steve Haas through a much smaller incision and not flipping the patella, that may not be true -- I will show you that at the end of the talk.

How about patient satisfaction? You see people get up and say "people are happier with their uni's than with their total knees". This is based upon three uni's that they did and four total knees. Let’s go to the Swedish Joint Registry. Over 30,000 operations with this satisfaction rate: Uni's 83%, Total Knees 82%. Patients are happy with both; there is really no difference.

Dick Scott has listed these as the indications for a uni, and Dick says that back in the 70's, when he tried to convince people to do uni's, he was a voice in the wilderness. Now, he is trying to convince people not to do uni's. How many of my patients in the past three years fulfilled the Scott criteria? 12%. Russ said 10%, same number. Steve Stern had a paper in 1989. He has 6%, even less. So it is not a huge number of patients.

Ten-year survivorship of uni's is very good, no question about that in any age group. This is one I did, which lasted 16 years, but look at post-10 years. If you look at the second decade, these are Dick Scott's numbers; look what happens after the first decade -- they really drop off. Look at TKR survivorship after the first decade from about 4 articles I put together that compare the two:  The survivorship into the second decade is better with the total knee.

So in patients who are younger, say patients in their 50s or early 60s, there is a very good chance that this will not be the definitive operation for them. In fact, for patients in their 70s, there is a good chance that it wont be the definitive operation. People say that if you do a uni, it can be easily revised if it fails. I've had experience with these revisions; there is no problem, but there are some that can not be easily revised. How about one which requires stems, augments, and everything else?

So I did another analysis of uni revisions in the journal for the last 10 years, and 40% of the cases described in the papers, when they needed revision, needed augments, bone grafting, or stems. Saying that revising a uni is simple is misleading; if you pick your simple failures, it's simple, but most of the time, it is not.

Now, the uni spacer: you don’t even need the polyethylene anymore, just put this thing into the knee, which, by the way, costs 4,500 dollars. Okay, we did this 20 or 30 years ago and it worked somehow, but it didn't work in a lot of patients, and I have no reason to think that it is going to work now. I think that this is really marketing hype over any sort of clinical sanity.

Patients are being told that if it doesn't work, "we will do another operation in two years." I don’t think that is a reasonable thing to say to patients -- we have already revised a couple of patients that have failed early, and this is the answer in a lot of places. On the other hand, there are lots of 15 to 20 year follow-ups of TKRs in patients under 55, and I just put three of them up with good long-term follow-up results. So if you decide that an arthroplasty is the procedure of choice, there is only a small subset of these younger patients who fulfill all these criteria to do a uni. We have to realize that it does indeed have a limited life expectancy. TKR, on the other hand, is predictable and has long term survivorship, and with newer materials, better accuracy, maybe surgical navigation, it should be even better.

And now that we can do them through small incisions and without disrupting the suprapatellar pouch, maybe we are getting better, and that is what the uni people have taught us to do. They taught us that that concept is the way to do total knees, and again, I learned this all from Steve Haas who is sitting in the back smiling, I hope; small incisions, three days postop, this mini-midvastus which Steve came up with, less blood loss, realistic surgical times - not two hours before they cut you off, rapid regaining of flexion, and decreased need for pain medicines postop. This is the typical kind of patient for this.

Now, the statement that uni's are better because they need less blood, they bend faster, and you don’t violate their quadriceps, all of those can be extrapolated to the total knee replacement patient. So I think that that is going to make us want to go towards total knee replacement in some of these patients, even bilaterally.

To sum up, I believe that we should consider total knee replacement for unicompartmental disease to be the operation of choice in our younger patients.

Reviewed by Stephen B. Haas, MD, MPH

^ Back to Top

Back in the Game Patient Stories: