“Many things have been discredited in our field (bone loss around failed total joints is due to ‘cement disease’), ” says David Lewallen. As a result “blind acceptance of modular metal-backed tibia must be reconsidered.” Mike Berend retorts, “We’ve had excellent survivorship with non-modular implants (98% survivorship at 20 years). The 10-year survivorship of the same implant in an all-poly nonmodular system was catastrophic (68%).”
This week’s Orthopaedic Crossfire® debate is “The All-Poly Tibia: Cheaper and Better.” For the proposition is David G. Lewallen, M.D. of Mayo Clinic in Rochester, Minnesota; against the proposition is Michael E. Berend, M.D. from the Center for Hip and Knee Surgery in Mooresville, Indiana. Moderating is Thomas S. Thornhill, M.D. from Harvard Medical School in Boston.
Dr. Lewallen: “In the work we do there are few things more dangerous than the things we think we know for certain. These are things that I was taught during the course of my training which were proven to be false or harmful: suction irrigation tubes are good for infected joints; bone loss around failed total joints is due to ‘cement disease, ’ etc. Blind acceptance of modular metal-backed tibias as the current gold standard for total knee is something you should reconsider.”
“In my practice, with a lot of revision work, one of the biggest challenges we face is this manmade plague—which we did not see in the early years of arthroplasty—and that is tiny, particulate-driven masses of osteolysis in implants that are still relatively well fixed and not very symptomatic.”
“Some of these cases come in quite silently. There are a lot of different types of poly wear and deformation, but I think that it’s multifactorial. But one factor in some of these cases can be what’s happening on the backside of components. It’s not just top side wear. Take the example of a case 28 years postop with lots of poly damage and not that great of a design. I challenge you to find me an all-poly case with massive lysis.”
“In the early ‘80s there was broad adoption of metal-backed implants for a variety of reasons, some of them related to cementless…some allowed insert exchange. It turns out that the advantage of insert exchange is illusory and almost always a bad idea. This was demonstrated in three Journal of Bone and Joint Surgery articles and one Journal of Arthroplasty article, which showed either no benefit or bad results in revisions where only isolated poly exchange was undertaken. It doesn’t address the other pathology such as malrotation and malposition.”
“Locking mechanism instability is no myth. It’s common and cuts across a variety of designs. We’ve all seen this with abrasion and top side damage to titanium trays, which were all adopted in the ‘80s as well across the industry. You can have extrusion of poly as well; this is a rare problem with a modular tray where it’s up in the pouch. Locking mechanism and backside wear are very real, and have been documented numerous times in the literature. ”
“If you look at what we do in our practice it’s a bait and switch. We quote results of one type to our patients, guaranteeing results with a very different device than we use. So we talk about long term results of total knees, but many cases are with monoblock designs…that’s not what you folks are using. You’re using modular trays.”
“Looking at the modular data you see that the longer follow-up is not quite as good. In a scattergram, as you get further out into the second decade the monoblock and modular survival drops off pretty quickly from wear and particulate-related problems. Is it technique related? Maybe. But there is a series by Dr. Chitranjan Ranawat, a master surgeon using his own cases, where he looked at all polys and metal-backed modular and found a large difference (75% and 96%, respectively).”
“We examined the long term survival of tibial implants at our institution over a 20 year period, comparing 17 different all-poly designs. We compared them to metal-backed components in the same implant systems. We corrected for age and gender, and found that all-poly tibias won all the time across all designs. All-poly tibias were at much less risk of revision versus metal-backed modular versions. We saw something interesting with posterior-stabilized (PS) and cruciate-retaining (CR) where it looked like our PS results weren’t quite as good…and when we took one implant system out that was disproportionately represented and a locking mechanism that was not ideal, that association went away.”
“But the message is important: the more constraint you have, the more you need to worry about these metal-backed modular connections. So the etiology of osteolysis is multifactorial…I’m not here to tell you that it’s all metal-backed tray driven, but this is one important part of it. Remember…osteolysis involves small particles and it’s those particles that drive this process. So going forward we can look for answers in cross-linked poly and perhaps in rotating platforms. But if you’re interested in expense and you care about bone loss and stress shielding consider all-polys…at least for some of your patients.”
Dr. Berend: “I think you’ll find that I have a similar line of thinking with a different conclusion. Mayo had an overpowered study (11, 606 total knees) looking at both patient and implant factors and had a 98% survivorship at 10 years. The best implant in their series was a non-modular, metal-backed tibial component…cemented fixation, all-poly patella, and retention of the PCL. This is what’s been proven in our data as well.”
“Wear is a problem, and is something we’ve not seen with molded, metal-backed implants in our series. There are many ways to improve on the wear issue, such as improved locking mechanisms. We can change the femoral material or the counter surface. You can add more movement like with a mobile bearing device, but that has not been shown to decrease it. We can go into the cross-linked poly era.”
“There are some advantages of all-poly implants; the non-modular feature is important. It is cost-sensitive, and perhaps in some markets or with bundled payments there may be an issue going forward. It does perhaps give us less resection for a given polyethylene thickness, however we’ve learned from the AGC experience that it may be design sensitive. And it hasn’t been adopted worldwide, nor have non-modular metal-backed implants, representing just under 4% of the knees in the British registry.”
“Looking at our data, we’ve had excellent survivorship with the nonmodular implants; using the Himont 1900 molded implants we had 99% survivorship at 15 years, and we now have 98% survivorship at 20 years. Unfortunately, the 10 year survivorship of the same implant in an all-poly nonmodular system had rather catastrophic results of 68%. These failures were all due to the same mechanism: the medial tibial bone could not support the implant.”
“The all-polyethylene implant has given us an excellent surrogate for studying the effects of alignment on the bone. In the long term failures we’ve observed the change in the medial tibial plateau from an overload from lack of transfer through the medial tibial plateau with metal backing. We’ve quantified this is in our lab, both on the anterior and posterior aspect of the medial tibial metaphysis. We had somewhere between a 40% and 500% increase in strain in the proximal tibia in the absence of metal backing. So for those patients you think are at risk for overload metal-backing is an important thing to consider. ”
“We’ve repeated these experiments with finite element analysis and the data is the same—increased load transfer from metal-backing in the proximal tibia. So we think there is a clinical correlation with this medial tibial overload, bone resorption and remodeling, and eventual loosening in all-poly implants that is much more unlikely in metal-backed devices.”
“We’ve looked at it in an RSA [radiostereometric analysis] model and developed an estimated tibial stress model looking at the size of the tibial plateau in the patient’s mass. We’ve correlated that with our ability to produce tibial component alignment. In looking at the failure rates we found that in smaller tibial components with larger patients there was a 20% failure rate at just 1.2 year follow up.”
“This same study was correlated with a radiostereometric analysis (RSA) study from Hyldahl out of England with 41 metal-backed implants and 40 all-poly implants. We found the same thing: smaller implants in bigger patients had higher rotation in the anterior-posterior direction, and a higher rotation in varus-valgus. There is some registry data which reflects selection bias on behalf of the surgeon to show equal survivorship between all-poly and metal-backed implants. And a meta-analysis by Forster of over 5, 000 knees (Level 1 evidence) shows no difference in appropriate patients with all-poly implants.”
Moderator Thornhill: “David, what percentage of your total knees gets all-poly implants?”
Dr. Lewallen: “Probably 40%.”
Moderator Thornhill: “Mike, what percentage of your practice is nonmodular, metal-backed components?”
Dr. Berend: “99%.”
Moderator Thornhill: “So if you agree that most of the implants available today have done things to substantially improve backside wear, and with material and mechanical things, etc., you still feel that we’re not there yet with the modularity?”
Dr. Berend: “We don’t have long term data. We change the design every 5/8/10 years before we have long term data. And 75% survivorship in metal-backed modular implants…we can do better than that.”
Moderator Thornhill: “David, is it possible that these people do well because you’re usually choosing lower demand patients?”
Dr. Lewallen: “When we corrected for the things that people usually list (age, etc.) the all-polys still outperformed metal-backed designs. I think a problem in this industry is that when we have a problem we don’t always address it; then when we start to recognize it, we stay one design iteration ahead of what we know for sure. We sometimes need modular trays. But many designs in the ’80s and ’90s didn’t work and created numerous problems.”
Moderator Thornhill: “If you are using a modular system you could always change it out; not so with an all-poly tibia or a nonmodular system.”
Dr. Lewallen: “So you put the trials in, and if you have well-designed trials that fit well you can go through the exact exercise right then and decide if you want a 10 or a 12, etc.”
Moderator Thornhill: “With your volume you change the geometric properties of these trials with repeated autoclaving. Another issue: what specifically was it with the AGC that made their all-poly tibia so much riskier?”
Dr. Berend: “It was a combination of the coronal plane geometry (flat on flat); we didn’t understand the importance of depth of tibial resection, and maybe the post geometry is too flexible and allowed medial-tibial loading. To your prior point about deciding during the operation, we haven’t had to deal with that in over 16, 000 nonmodular implants; so there’s education needed about when to make the decision, how to make the exposure to put in a one piece implant.”
Dr. Lewallen: “Tom, you’re right. When I’m doing all-polys or monoblocks I will err a bit towards the next thicker.”
Moderator Thornhill: “Good discussion. Thank you both.”
Please visit www.CCJR.com to register for the 2013 CCJR Winter Meeting, December 11–14 in Orlando, Florida.
“You may now view CCJR meeting content on your mobile device on the CCJR MobileTM App. Please scan the QR code to download from iTunes.”
