This week’s Orthopaedic Crossfire® debate was part of the 16th Annual Current Concepts in Joint Replacement® (CCJR) – Spring meeting, which took place in Las Vegas this past May. This week’s topic is “Medial Compartment Arthritis: TKA the Only Tried and True Solution.” For the proposition is Thomas P. Sculco, M.D., from the Hospital for Special Surgery. Fares S. Haddad, M.B., F.R.C.S., University College Hospital, London, United Kingdom, is in opposition. Moderating is Steven J. MacDonald, M.D., F.R.C.S.(C) from the University of Western Ontario.
Dr. Sculco: The title is probably a little extreme here. The only true and tried solution. I think certainly there is a place for unicompartmental knee replacement, but I’ll present an argument in the opposite direction. There are many, many advantages of total knee replacement [TKR]. You all know these. Technique is very reproducible. Long-term results have been outstanding and revision techniques and implants have certainly gotten much better. For some patients total knee replacement is, in fact, the only option. So if you have a patient with severe malalignments and deformity after a fracture, medial compartment arthritis, there is no way you can do a unicompartmental knee in that situation. It lends itself to augmentation and use of stems.
In fact, if you look at medial compartment osteoarthritic knees, about 85-95% of arthritic knees in general are really indicated only for total knee arthroplasty, in my experience. We looked at 250 osteoarthritic knees and followed the criteria that were outlined by Scott and Kozin. We found that about 8% of the knees were really good candidates for unicompartmental knee. That means, in the United States, the average orthopedic surgeon does 25 total knees per year, he’s really seeing only 3, maybe 4 patients at most that are really ideal candidates for unicompartmental. That can be a problem in that, as David Hungerford has said, the real disadvantage of unicompartmental arthroplasty is the technical difficulty of the procedure.
The technique pitfalls are in the area of soft tissue balancing; how much you do, how much you not do. Instrumentation can be a problem as well. Visualization as you put the device in could also be a problem. Technique in the hands of neophytes, or those who are not doing these on a large volume, and the un-replaced lateral compartment, are problems in terms of failure as well.
There are some real disadvantages to total knee replacement. They do have some functional limitations, probably will not flex as much as a unicompartmental. The revision can be more complex, and the recovery is longer in the short-term.
But the rehabilitation is not always slower with total knee replacement patients and if you have a fit, younger patient with good motion, in fact, he/she will recover very quickly. Probably not dissimilar to a unicompartmental.
Now what about results. If you go to the literature, you’ll find results all over the place. They show excellent long-term results with unicompartmental as well as tri-compartmental. But overall today the chance of a total knee replacement lasting greater than 20 years is better than both osteotomy and unicompartmental. I think most surgeons who even do unicompartmentals would agree with that.
Rodriguez-Merchan (Arch Bone Jt Surg, 2014) performed a meta-analysis that looked at medial OA, unicompartmental versus total knee replacement, and what the paper showed is higher revision, as you would probably expect, at 5, 10, and 15 years in the unicompartmental group. But complications were greater in the total knee group, who are often sick or older and have co-morbidities. Parratte, et al. (JBJS-BR, 2009) wrote a paper where the medial compartmental group under age 50, 35 knees followed out to almost 10 years, had 6 revisions with an 80% survivorship at 12 years.
And revisions of unicompartmentals are not as straightforward as often they may look. Bone loss can be significant on the tibial side and require augmentation and stems.
In summary, I think there are limited indications for unicompartmental knee replacements, but in the hands of Professor Haddad the results will be good and surgeons who do high volumes of them, the results will be good. Technique for the average orthopedic surgeon is more difficult. Revision rate is higher as we’ve demonstrated, and most people agree, and revision can be more complex.
Mr. Haddad: You just had a very nice exposition for total knee from Tom, but the realities in my practice is that over 30% of my patients have unicompartmental disease. There is no doubt that my best joint replacements are unis. They’ve got better kinematics, better proprioception, better range of motion, and higher activity levels with fewer complications.
And it’s even recognized at HSS [Hospital for Special Surgery]. I remember first visiting there and you hear a pin drop when uni was mentioned. Nowadays if you use Doctor Google, you’ll find unis are going in very regularly at HSS and patients recover pretty quickly and do well.
Now we know that medial OA [osteoarthritis] is ideal for unicompartmental knee replacement. It’s been well studied. The indications are clear. The patient needs to have medial osteoarthritis, full thickness cartilage loss, an intact ACL, and correctible deformity with preservation of lateral cartilage.
The problem is the old contraindications. You’ve heard about the Kozinn and Scott criteria already. The reality is they no longer apply. There is now good data to suggest that they don’t really change outcome. An Oxford series compared the idealized Kozinn and Scott criteria with cases where that didn’t apply. In over two-thirds of the cases, those criteria were not met. Yet, when we look at the outcomes in terms of scores, there’s no difference in score, no difference in failure rate. If we look at survivorship, good survivorship at 96% at 12 years. No difference between whether you follow those criteria or not.
Therein lies the rub, because if you follow those criteria as you heard, you’ll do unis in less than 10% of your cases. But if don’t, and let’s still stick within the rules I gave you, that percentage goes up dramatically. Up to 50% for the enthusiasts and you extend the indications beyond where they were traditionally.
A study from Sweden showed that if you do that, you have a 20-year survivorship of over 92%. Very impressive, rivaling a lot of total knee studies. And there are multiple studies out there that show you similar data where you can have high survivorship with good looking unis.
Why do this? Well, the reality is, it’s better. Patients retain the cruciate so they have better function. They have better range of motion, better gait and better pain relief. They recover more quickly with fewer complications.
Let’s look at function. That’s what it’s really all about. For our patients we’ve looked at function a number of ways. I’ll just focus for a moment on gait analysis and when we put them on a force plate treadmill. The reality is if you make them do something demanding, like walking downhill, UKAs [unicompartmental knees] can do that, but less than two-thirds of TKAs can. If you get them to maximum speed that’s much better for the UKAs than it is for the TKAs. They’re much closer to normal on most gait perimeters.
So in a matched group of patients, we’ve seen a dramatic difference in functional outcome, that doesn’t seem to be age or sex dependent. The really interesting thing is if you look at complications, across the board, fewer complications, fewer DVTs, fewer infections in the UNIs than the totals. And that’s something we need to take seriously for our patients.
Look at mortality. Significantly decreased for UNIs in the first year, and that continues out to eight years. That can have a huge impact. Now the elephant in the room, of course, is revision. You’ve already heard that. There is a high revision rate for UNIs. Sometimes that’s because they’re put in badly. The reality, however, is that seems to relate to low volume use. You’ve got to be doing quite a few of these if you’re going to do them well and succeed. If we look at just those who do 20-50% UNIs, we see no difference in reoperation rates.
So what we should do is treat medial OA as unicompartmental disease because sometimes less if more.
Moderator MacDonald: Thank you gentlemen for a well-balanced debate. So Tom, do you do any UNIs yourself?
Dr. Sculco: I do not.
Moderator MacDonald: Have you in your practice over your career? Have you at some point in the past in your career?
Dr. Sculco: Yeah, I did UNIs when they originally became available in the early ‘80s. The technique was not as good. The implants were not as good. We had a lot of problems with them and we abandoned it. We have 5-6 guys now who are doing UNIs regularly. And if I get a patient who comes in to me who’s a good candidate and absolutely insists on a UNI, I’ll refer to one of the people who do it.
Moderator MacDonald: Fares, who’s an ideal UNI patient?
Mr. Haddad: I think we get messed up in this. Most people are ideal UNI candidates, as you know Steve.
It’s not age or sex, it’s not about obesity. I think we got hung up on Kozinn and Scott way too long. If you’ve got bone-on-bone medially, and you have a preserved ACL and your deformity is correctible, you’re probably a good candidate.
Moderator MacDonald: Tom, who do you think should not get a UNI?
Dr. Sculco: I think if somebody has tricompartmental or significant bicompartmental disease, those patients in the end will not do well with a UNI. I only found 8-9% of people who are ideal candidates for it. Now you can expand that to maybe 15%. There may be more. There’s some people who would expand it more, but I think roughly 10-15% of patients are ideal candidates.
Moderator MacDonald: Fares, anything to add to that. If someone walks in the door and you say, “No, you’re not a good candidate.”
Mr. Haddad: If your lateral compartment has significant disease with cartilage loss, you should absolutely have a total. And I think it’s worth stressing…I’m not a mad enthusiast. If you look at the UK data, there is that little group who do a fair proportion of UNIs very well. We’ve got a few guys who are doing far too many UNIs; when your UNI proportion goes up beyond 50% your failure rate goes up. You can definitely push the envelope too far. I think we’ve gone from 2%, 5%, 6%–we’ve cracked up to beyond 20% up to about 30% and I think we’ll find a middle point where everyone’s comfortable.
Moderator MacDonald: Thank you gentlemen for the wonderful debate.
Please visit www.CCJR.com to register for the 2015 CCJR Winter Meeting, December 9 – 12 in Orlando.
