Evan F. Ekman, MD: Well, next it's my pleasure to introduce to you Dr. Kirby Hitt; he is the Head of Adult Reconstruction and Joint Replacement Surgery at Scott and White Memorial Hospital in Temple, Texas. He's also Assistant Professor at Texas A&M and he's going to talk to us a little bit about design and the title of his lecture is, "Improving Patient Satisfaction Through Design: Enhancing Knee Motion."

Kirby D. Hitt, MD: Thank you; I appreciate it. First, Peter alluded to patient expectations of total knee replacement are highly variable and they depend on the age, the diagnosis, and the activity level of your patient. These preoperative expectations strongly influence the patient's perceived outcome after total knee replacement.

We all can agree that, I guess, the expectations of our patients have changed over the years. No longer is it acceptable for a patient to have less pain than they had before their knee and Peter alluded to that. And what we can see is we're not doing a very good job and in reality, over 50% of patients do express the dissatisfaction as Peter talked about.

Range of motion is important for activities of daily living and if we don't provide these patients with this amount of motion after our knee replacement, they're going to be dissatisfied.

Well, can we restore knee function? This article by Noble addressed that and 52% of patients reported limitations in functional activities, and he compared those to age, gender-matched peers and found that 22% couldn't do those activities either. But 50% of patients are not happy with that procedure. And there are all kinds of different functional limitations. Much of this difference is a reflection of the biomechanical deficiencies of contemporary implant designs, which leads us to our talk as improvements in the procedure and the designs are really necessary.

There are a lot of factors that go into maximizing our results; you have patient factors, you have surgical technique factors, and implant design.

And I would suggest to you we cannot let the implant be the reason that those patients aren't realizing their expectations. Design must approach expectation, but we have to be realistic and we have to sit down with our patients preoperatively and let them know what our expectations are and understand their expectations; we can't have this, this is not possible.

But the problem is current implant designs are not meeting those expectations and there's a lot of kinematic abnormalities that we've seen in the literature -- reduced rollback; we have paradoxical translation occurring; reverse axial rotations; condylar liftoff. And only when we understand those kinematic variances can we insist in designing better total knee implants.

Can we improve range of motion by changing design? What I'd like to share with you is just a couple of slides on a design that is relatively new, the Triathlon knee, and comparing it to a good knee -- a good knee, the Duracon; it's been out for a long time. We look at the primary diagnoses of these two groups, which are very similar.

But what wasn't similar and is probably fairly significant is the range of motion we've seen, even at 6 months.

But probably what is more encouraging is the results of the range of motion that we've seen at 6 weeks; these patients are recovering quicker, they're getting their function back. The only variable that was changed in these two groups who were almost identical is the design.

So I'd like to look at the design a little bit; what makes this knee different? It's an anatomic radius design. Kinematic and biomechanical studies by Encabo, Churchill, Hollister, and others have shown us that the natural knee motion occurs around a constant radius and that constant radius is centered through the transepicondylar axis. By having a constant center of rotation, what we can do is improve on ligament isometry; we can also replicate the kinematics and the natural motion of the knee.

This challenges the previous idea of the instantaneous center of the knee changing, as we go from flexion to extension, the so-called multiradius design. The problem with changing the center of rotation is we've all seen problems with midflexion instability and some implications as far as patellar femoral mechanism, and we'll talk about it in just a minute.

An anatomic radius design puts the posterior flexion-extension axis back where we can lengthen the extensor moment arm that decreases the quadriceps muscle force and reduces the joint reaction force unlike a multiradius design. As the knee center changes as we go from flexion and extension, we actually decrease the extensor moment arm, which increases quadriceps force and that little old lady who's trying to get up from a chair, stoop or bend, very difficult with a multiradius design.

Mahoney had an excellent article looking at total knee design and extensor mechanism function, looking at these two design philosophies. His conclusions were that the single radius knee improved the chair rise, decreased anterior knee pain, had a more rapid flexion gain, and that's very similar to the results we're seeing with this new Triathlon component.

Banks and Hodge evaluated five different designs, using fluoroscopic imaging in both gait and stair climbing activities, and they evaluated different design philosophies performed in vivo. And what they came up with, the conclusions very interesting, is the single radius design or anatomic design is less sensitive to patient and surgical variables and may provide more reproducible kinematic results than the multiradius; again, what we're seeing clinically and in the lab.

Well, what about these reproducible kinematics? What about the tension in the ligament? Well, that's a very interesting lab finding. If we look at the multiradius in blue and the anatomic radius in red, what we see evaluating knee collateral ligament forces as we go from flexion extension, we see with the multiradius design, we see peaks and ligament force that's unacceptable. That creates two scenarios: Either one, over time, that ligament is going to stretch out and the patient's pain may decrease, or that ligament stress stays there and we have a pain--a patient has persistent pain. I think we've all had patients who come back at 6 months to a year and you're feeling their knee and they're a lot more lax than you remember when you were in the operating room. And I would suggest to you this is what's happening; we're asking the ligaments to do something they do not want to do and it could be here -- it could be a design problem.

The normal knee kinematics of the functional active arc of motion on the medial side behaves almost like a ball and socket joint; it's very stationary. The lateral side assumes a more anterior and posterior position to accommodate internal rotation, and it's important that we have internal rotation of the tibia in order to maximize flexion. This all happens, this active flexion up to about 120 degrees; once we go from 120 degrees, extending up to 145 or 150, it involves a more passive flexion. Now that involves a very complex set of kinematics that we cannot reproduce in a total knee. But what we can do is learn from those kinematics and apply those kinematics to design to try to approach those expectations that our patients want. We can go on.

One of the design features of this new knee in trying to address this, we have to design a knee that focuses on the backside of the knee in order to gain function and motion in that backside. One idea is to flare the posterior femoral condyles.

What this does is you can see by extending that articular surface, we can increase contact area in deep flexion and decrease contact stresses. But probably more importantly, by increasing that contact area, we can allow for increasing tibial rotation, internal rotation that is required for maximum flexion.

Now if we look at this in a deep flexion rig, we can see the older design up top, a more contemporary Triathlon knee below. We can see in deep flexion the increased contact area and you can just feel the stability that will occur when we have increased contact area and clinically we're seeing that as well.

If we look at it here, again, it's a good knee system but what we saw at deep flexion, it didn't have design parameters to address deep flexion. We had condylar liftoff, high peak stresses on the medial side as opposed to the newer designs.

They actually have some designs that are actually directed towards deep flexion. We don't see liftoff; we're seeing good contact area; we're seeing peak stresses being decreased, so we can have an implant that hopefully will provide long-term durability.

This is combined with a patented machining process of the tibial insert, the so-called rotary arc. And what this does is provide a smoother, more consistent articulation and also again facilitates that natural rotation.

When we have movement, flexion and extension, the femur moves forward and backward, and what we need to try to produce is a rotary arc to allow that femur to naturally smoothly flow in that internal rotation, tibial flow internal rotation, as the femur externally rotates. And this type of design seems to help.

If we combine that with data, known data that single radius knees can decrease the rotational constraint, now we've got a knee that can attempt to approach the kinematics that we're trying to accomplish.

We can't forget what happens in the front of the knee, either. If we're going to ask these patients to bend beyond 120-130, 140 degrees, we need to decrease patellar tendon implant impingement. We have to modify the implant on the tibial-based plate and insert to allow this flexion to occur. When patients kneel, they don't kneel on their kneecap; they kneel on this tubercle and on their patellar tendon and we have to design an implant that allows these patients to kneel.

Modifying implants to increase range of motion can put the materials at increased stress and we have to assure ourselves that we're not creating a problem trying to treat a problem. And we've done wear studies and wear evaluations, and what we see with this particular design is decreased tibiofemoral articular forces in deep flexion, 120 degrees of flexion, 6 degrees of internal rotation, the peak stresses are very similar.

As we increase range of motion, we also have to have a locking mechanism, which will prevent backside wear, which I would suggest to you we can and we should eliminate backside wear in total knee replacement. It's about time this is not an issue and I think we can get it accomplished.

The days of having these implants where we have the backside wear and we see 5-year failure rates such as this, where we have poly problems where we have osteolysis that requires us to go back should be eliminated. This is not just a poly problem; it's a metal problem.

When you combine metal particles from that base plate with the poly, you have a very aggressive inflammatory reaction and osteolytic response.

I suggest to you we should eliminate this by providing locking mechanisms and micro motion, decreasing micro motion and have a locking mechanism that exceeds industry standards and it can be done, and that's the role of the implant industries.

Appropriately sized implants are also necessarily to maximize inflexion. Some prosthetic dimensions differ from actual knee morphology; we've all seen this, when an implant just doesn't fit. We've done a good job; we're getting ready to head out and play a little golf or go home to our families and we've got an implant that looks like this.

And we're interested in the same thing and back in 2003, JBJS published this article looking at the measurements of the human knee and correlating to implant systems. We measured the critical dimensions of the femur, the tibia, and the patella after the bone cuts were made; we analyzed the data to try to define average sizes for these patient groups and separate them out by gender.

And we did this because there are really no studies evaluating morphologic dimensions after bone cuts -- if you ask the companies where they got these dimensions, they really can't give you an idea; they're not real sure. And this was used to provide some guideline, maybe directing future prosthetic dimensions.

We had a perceived problem; we've all seen this, especially in the female population where the medial lateral width was too wide for a given AP and overhang would occur. We had eight centers that collected data (we excluded those patients with severe deformities), 337 knees. As you see, most of those were females as a general population, 69 years was the mean age. We looked at six different knee prosthetic systems back in 2003 and defined an aspect ratio and looked at them statistically.

I'm going to just show you some of the highlights. If we look at the femur on the Y axis, the ML dimension of that component, and the AP dimension on the X axis, we can see these six implants matched pretty well the bone morphology that we were seeing.

If we did separate them out into gender, we look at the men, again we see it's not bad, but we see some problems, that most designs were really smaller in ML for a given AP usually in those smaller sizes.

But the striking thing is when we looked at females, our clinical perception was this is what we would see. But what happens? The medial lateral overhang increased as the implant size increased, and you can see some very unacceptable overhang, and I guess the question is, what do you do with that?

By using an anthropometric-based design, we can better fit the majority of those smaller typical females. We can have the female friendly femur, if you will. We are flexible enough to address those larger anatomies, but we need to accommodate that additional population that was done with this implant.

Women have generally narrow femora than do men for a given AP dimension, so what do you do in a situation clinically if you see 8-10 millimeters of overhang? You've got a couple of options. But what if you're a posterior referencing surgeon? What you can do is you can downsize but at the risk of potentially notching the femur -- probably not a good idea.

There are companies that make these specialty jigs where you can flex the distal femoral cut in order to prevent the notching when you try to downsize, and we can get implants that match the bone a little bit better. If you're an anterior referencing surgeon, we want to downsize this component on the upper left; we have to modify our posterior condylar resection and I would suggest to you that makes for a very challenging surgery, to try to get that balanced.

And also it creates a situation in flexion where we impinge the tibia on the posterior femur earlier and may impede our flexion.

So how much overhang is acceptable? We did kind of a clinical study, kind of a fun clinical study. We took one implant with the same AP dimension and all we did is vary the ML dimension; this is one that fit well. We take it through a range of motion; it looks pretty good. We did take a component that's between 5 and 8 millimeters of overhang and go through that same procedure. The only change is in the femur, the AP dimension was the same.

Very interestingly, that thing didn't come out straight. And what we saw is when you have 5-8 millimeters of overhang, less than 5 really wasn't a big issue, but 5-8 millimeters of overhang, the surgeons who were performing that operation 27% of the time felt like they should reduce the tibial insert by 3 millimeters. The bottom line: Size does matter! And it creates a balancing problem if we have implants that don't match the bone we're trying to replace, and that's up to the implant manufacturers to address this problem.

What about clinically? We looked at overhang greater than 5 millimeters. What we saw, kind of summarizing, the results weren't as good. The range of motion wasn't as good. So again, size does matter.

It's important for the implant design to restore the AP femoral dimension in order to prevent overstuffing.

If we overstuff, as you see here, we're not going to get that knee to flex.

We must also keep in mind we need to try to restore the posterior condylar offset. We know that if we vary that resection, we're going to have problems getting maximum flexion and potential for impingement. Banks has showed us for every 2-millimeter decrease in that posterior condylar offset, we're going to reduce flexion by 12 degrees because we're going to impinge earlier on inflexion. Likewise, in 28% of the knees, this flexion was directly blocked by impingement of that posterior aspect of the tibial insert on the femur, so we need to maintain the posterior condylar offset.

So if we want to restore the AP dimension, we don't want to compromise by taking more posterior condylar bone off; we need another solution. And this is a solution -- a new design of this implant where we address it from the anterior side. Keep the posterior condylar offset intact; direct the anterior femoral cut 7 degrees so that we're in between sizes; we can now downsize without notching. Before, if we were in between sizes, we always went to the higher size, generally preventing a notching situation, which automatically overstuffed that patellar femoral joint.

Can simple changes help range of motion? Here's a good study by our next speaker, looking at in vivo kinematics using computer-assisted navigation. And all that was done using the same implant, they changed the insert, the tibial insert, by relaxing that posterior slope you see in red and that's the only change that was made.

And they found fairly significant improvements and greater overall range of motion, so simple design changes could go a long way of helping our patients. I think computer-assisted navigation is a unique opportunity for us to assess intraoperative kinematics and try to see how these design changes will affect certain parameters in total knee replacement, and I think we can use this as a tool to help design better implants.

When it's all said and done, it's all about the patient and the patient's satisfaction. And as Peter would say, I'll tell you there are changes in design that we can make; we can improve designs in multifactorial approach, but design has to be critical. We want patients who can recover quickly, who have a long-term implant stability. And when those patients are happy, I tell you we're happy; we have a better practice.

Maximizing results in total knee, again, is multifactorial; it depends on the patient, depends on your surgical technique, and it depends on the implant design. And we can't let the implant design be the reason these patients can't function and get back to their expected expectation. With that, I thank you.