Trousdale v. Mullaji: Knee Navigation: Lost in Space

This week’s Orthopaedic Crossfire® debate was part of the 33rd Annual Current Concepts in Joint Replacement® (CCJR®), Winter meeting, which took place in Orlando. This week’s topic is “Trousdale v. Mullaji: Knee Navigation: Lost in Space.” For is Robert T. Trousdale, M.D., Mayo Clinic, Rochester, Minnesota. Opposing is Arun Mullaji, F.R.C.S.(Ed), M.S., The Arthritis Clinic, Mumbai, India. Moderating is Thomas S. Thornhill, M.D., Harvard Medical School, Boston, Massachusetts.

Dr. Trousdale: I’m going to argue against the role of routine use of navigation in our simple, primary total knee replacements. And I would argue that it remains a somewhat cumbersome, time-consuming, relatively expensive tool with some, but very limited proven clinical benefit.

Certainly, we all share a common goal that we want our knee replacements to be reliable, and I think we’ve got that. We want them to be durable. I think we’ve achieved that. We want it to be safe. I think we’ve achieved that as well. We want to alleviate pain. We’re pretty good at that for the majority of our patients. But I think the unknown is improving function in our patients. And I think that’s what we are striving for now and talk about a lot at meetings like this.

Arun is going to tell you, and I agree completely with him, with manual, routine techniques alone, we are not very accurate or precise with our component position. That’s inarguable.

Today, computer-assisted surgery (CAS) has limited proven clinical benefits. That might be okay if this was a brand new technology, but we’re now well into our second decade of computer-assisted surgery and there is limited data that supports using a computer with your total knees – that you’re going to have better range of motion, better function, better durability, although there is some new data that supports that, and better WOMAC SF-36 scores.

In terms of Level 1 studies that have been published, a meta-analysis (Bauwens, et al., JBJS-Am 2007) looked at 11 randomized controlled trials and the conclusion was “Navigated knee replacements provide few advantages over conventional surgery and its clinical benefits are unclear and remain to be defined.”

I think that’s still true today.

Another series, Level 1 study, randomized controlled trial, CAS versus standard total knee replacement (Harvie, et al., JOA 2012). At 5 years, CAS had better alignment and fewer outliers.

But no difference in Knee Society scores, patient-reported outcomes, and patient satisfaction in the CAS group versus the conventional group. At 5 years the computer appeared to be no different than the conventional techniques with these outcome measures.

What has been demonstrated with the computer is that it’s a better tool to hit a specific target than manual instruments. Undeniable. But the problem is we don’t know what the right target is for individual patients. I think there are factors other than alignment that may be more important for durability than the sagittal and axial alignment of the total knee—soft tissue balance, patient factors, etc.

CAS has increased cost for the surgery. And there is an increased prevalence of specific complications such as periprosthetic femur fractures from a computer navigation pin site.

The fundamental premise of the computer is if you use a computer, your alignment is going to be better, and that’s going to give better function and survivorship of the total knee replacement.

But that’s not really been proven.

Mark Pagnano at our place, I think, was one of the first people to question the role of the mechanical axis as far as durability and function with total knee replacements. If you look at a series of Mayo Clinic patients at 15-year follow-up, 275 patients, 399 knees, the mechanical axis was measured in 5 zones (Parratte, et al., JBJS Am 2010). Most of the patients ended up in zone 3, that’s the middle. But there were some outliers—a little bit of varus; a little bit of valgus.

And the authors also looked at the overall durability of the knee. Revision for any reason was no different at 15 years whether the knees were in the well-aligned group, or if they were in the outlier group. They looked at survival free revision for aseptic loosening or wear/osteolysis by alignment. Again, no difference.

This study doesn’t mean that alignment isn’t important. All of us would agree, I think, that alignment is one of the factors that is important in total knee surgery. What this study does tell us is that there are other factors that are more important for 15-year durability. Alignment is important, but if you have a little bit of varus in your knee I’m not sure the durability is going to be badly affected if other factors aren’t taken into consideration.

CAS is more complex. There are tracking devices. You’ve got a camera in the operating room. The work is more complex for your assistants.

I do think we should want to harness the power of the computer for our knee replacements, if accuracy, efficiency, decreased cost (potentially) and fewer complications can be achieved.

I think, however, it may be better to use the computer outside the operating room. There are a lot of techniques that are coming into play with advances in 3D reconstruction. I think in the future it’s going to be patient-specific solutions, not generic average solutions—putting the mechanical axis down the middle—that’s going to really shift the bar for our patients.

I argue to you that, today, routine use of a computer in the operating room remains a cumbersome, relatively time consuming, relatively expensive tool, with very limited proven clinical benefit for your patients.

As a research tool it will help us, I think in the future, to find the right targets and once we find that for the individual patient, I think then the computer is going to help us with that individual patient—put that alignment exactly where that patient needs to be.

Dr. Mullaji: Houston, we have a problem. Some surgeons have developed the ostrich syndrome and are ignoring facts. The first one is that alignment is superior with navigation—I think Robert has accepted that.

But he quoted a meta-analysis from 2007.

More recent papers show that the risk of malalignment is much higher and all favor navigation. No question about it. If you can restore the alignment within 3 degrees of what you want, the scores are much higher. The International Knee Score, the SF-12, mental and physical scores are much better, if you can get them aligned. And this is particularly valuable in the obese patients, in those where there is an implant and an extraarticular deformity.

Fact #2: you can do much better balancing of soft tissues with navigation.

Consider a severely unstable, malaligned patient. You can assess his/her deformity very accurately with navigation, do a tibia cut and then check your alignment and balance in extension. Do a soft tissue release as required and then again check that you’ve obtained balance and alignment.

Then you can balance your flexion gap exactly to your extension gap and then proceed with the cuts after you have done the planning to get equal flexion and extension gaps. You can check your alignment and stability, not only in extension, but also right through the range of motion to full flexion.

Fact # 3: Functional results are superior. Dr. Trousdale didn’t look at all of the published papers. Gothesen, et al. (Bone Joint J 2014) reported at 3 months and 1 year that the KOOS scores are significantly better with navigation.

At 5 years, a prospective study (Hoffart, et al., JBJS Br 2012), compares the pain scores, the knee scores, the function scores, as well as the Knee Society scores. And all these are superior with navigation.

Fact # 4: Complications are actually fewer and the most dreaded one is of emboli, and a recent paper in JBJS-Am (Malhotra, et al., 2015) is a prospective, randomized study which looked at the number of emboli and calculated the embolic scores. With navigation the scores are much lower when compared to conventional surgery.

Fact # 5: Revisions are reduced. A paper based on the Australian Registry (de Steiger, et al., JBJS-Am 2015) shows that CAS reduces the revision rate for patients under the age of 65 years (6.3% vs 7.8% with conventional). This is significant because patients under the age of 65 years are most likely to have a higher rate of revision. Data from the Australian Registry over 9 years, shows with CAS the revision rates have gone down significantly.

Within our own series, navigation outcomes were much better than conventional surgery. Before CAS where we had done conventional surgery, we had 15 revisions for mechanical causes in 1,000 patients. After navigation we had only 3 revisions in the first 1,000 cases and just 1 revision in the next 4,000 cases with navigation.

Lastly, navigation is cost effective. A recent paper from Norway (Gothesen, et al., BMC Musculoskeletal Disorders 2013) looked at the economic models to evaluate cost effectiveness. In order to be cost effective navigation needs to increase the total knee survival rate by 1% in centers doing more than 25 total knees annually. If the center is doing more than 250 total knees a year, then the survival rate needs to increase by just 0.1% for navigation to be cost effective.

Navigation outscores conventional surgery in terms of achieving the alignment you want; getting the balance that you want. Better function. Reduced emboli, reduced risk of revision and it’s cost effective.

Is navigation lost in space? No! It is out of this world. You’ve got to use it and be ready for a joint space odyssey.

Moderator Thornhill: These were 2 great talks. Rob, you’ve used it before. When would you or when do you use navigation?

Dr. Trousdale: Yes, I used it for a period of time in a consecutive series of knees and I was happy. I had precise tools and accurate tools, relatively quick for me. The problem is a couple of things.

We jumped from no intra-operative tool, skipping the X-ray part, and going straight to this expensive computer. While CAS is worthwhile in complex cases, the problem is for each individual patient when you look at sagittal alignment, axial alignment and balance, we really don’t know what the balance should be for an 85-year-old patient who needs a total knee versus a 60-year-old. Until we know the target, I’m not sure we’re going to realize the benefits of intra-operative computers and navigation.

Moderator Thornhill: Let’s say you’re not very experienced and the CAS learning curve is 40 cases to really become good. Will navigation shorten your learning curve?

Dr. Mullaji: I think so. You know exactly what to do. You can titrate your soft tissue releases. You can check your alignment. We’ve demonstrated this with a lot of our young colleagues.

Moderator Thornhill: Every time I start a knee, I sit there for a moment and I feel the knee to get an understanding of whether it’s tight or loose; whether it’s flexed, whether it’s not and as Rob said, the soft tissues are different. In rheumatoids, osteoarthritics, young people, old people. Are you as good with CAS at individualizing the soft tissue balance of that knee or are you better at just the alignment?

Dr. Mullaji: I think alignment is a given with navigation. The real advantage is balancing your soft tissues and I use a tensioner and we did some studies with transducers and I think that’s going to be future.

When we combine sensors with navigation we get a number to the ‘feel’.

We did that with transducers. We got an accurate sense of balancing pressure medially and laterally. If you can get a number with these sensors then you know if you’re tight medially. Do that maneuver, check with navigation to see if you achieved what you wanted to set out with. Balancing the flexion gap, again, is something that you can flare out on the computer and make sure that it’s balanced as you like before doing your cuts. If you’re not happy with that, you can change that and achieve more or less what you’ve set out to achieve. And that may be tighter in a younger person and more lax in an older person.

Moderator Thornhill: Gentlemen, thank you so much. Very good job, both of you.

Please visit www.CCJR.com to register for the 2018 CCJR Winter Meeting, – December 12 – 15 in Orlando.