• LecturehallDual-Component Metallic Implant vs K-Wire for PIPJ Fusion: A Randomized Controlled Clinical Trial
  • Lecture Transcript
  • Harold: Topics to cover so our first speaker is a dear friend of mine. He reminded me that he was a student of mine, 1979 or something, right? I hate to think that I’m that old. Scot has got a tremendous job in the area foot and ankle surgery. As you know, he’s the editor for the Journal of the American College of Foot and Ankle Surgeons. That is an incredibly difficult job; one with tremendous respect. He’s also an excellent surgeon, involves got a master’s degree in addition to all these other degrees that he holds involved with research. He definitely has done an outstanding job in our field of foot and ankle surgery. So I’ve given Scot the opportunity to share some of his thoughts on arthrodesis associated with hammertoe repair. Considering we have enough of those procedures with complications, I thought Scott would be perfect to eliminate all of our problems. So please welcome, Scot Malay.

    Scot Malay: Thank you, Harold. Yup, it goes back to 1979, Harold just a few years ago. So I work at Penn with Harold and going to talk a little about hammertoes. I want to disclose that my travel and honorarium is covered by Nextremity Solutions. The manufacturer and the owner of the device that we undertook a randomized controlled trial to access and to compare to K-Wire fixation for proximal interphalangeal arthrodesis. So this focuses on this dual-component metallic implant versus standard therapy, K-Wire fixation for proximal interphalangeal fusion. It’s important when you compare treatments especially a new treatment or an alternative treatment, you want to compare it to the standard. And everyone knows that the standard for fusion is the Kirschner wire and there are lots of other ways. You can loop a wire intraosseously. There are a lot of other implantable metallic devices that’ll do it. You can suture the collateral ligaments tightly together. You can do it in a lot of different ways and as surgeons, we’ve done it in many different ways. But we’re always looking for a better way that’s going to be better for our patients subjectively where they like it better and for us where it may be easier for us and takes away some of the potential headaches that we deal with everyday taking care of these patients. So standard therapy is K-Wire and we’re all good at that and we can use it to transfixate the proximal interphalangeal joint. You’re also traversing the distal interphalangeal joint going out of the tip of that toe. You’re also able to stabilize the metatarsophalangeal joint which is a difficult joint to maintain especially if you’re dealing with like a crossover or dislocated MP joint, which I find personally challenging every time I deal with these toes. But the K-Wire depends on that coefficient of friction between the metal pin and the bone. And it’s primarily trabecular bone at the interface where we’re trying to get the fusion, but also some subchondral cortical bone proximally and distally as you go through these joints – as you go through these phalanges and out the tip. And it sticks out of the tip of the toe, no kidding and that can be a little alarming to some patients; problematic. And anytime we have extension of or pull of our extensor tendons or flexor tendons or contact with the substrate at the tip of the toe – the pulp of the toe, we affect some three point bending or flexural force on the pin which disrupts that coefficient or it may overcome that coefficient friction. We get movement at our fusion interface and that’s why I think we get a lot of effective functional pseudoarthrosis with K-Wire fixation. But from time to time, we also get a fat stiff toe and a reoccurrence of deformity or some new angular deformity that can be problematic. And this has been discussed in the literature, a recent article in Foot and Ankle International in 2012 talked about this, the protrusion of it. Patients have to keep it clean and dry. There’s a certain anxiety and discomfort associated with that pin. The flexural loosening, migration, pin tract infection, bending; making it difficult for us to take it out when it’s time to take it out.


    And that reabsorption of bone around the pin at the fusion interface, leading to that fibrous pseudoarthrosis and god forbid, you get pinned breakage. And you look at that radiograph there, this was fixated with a K-Wire. So was the third toe, right next to it. So why did the third toe proximal interphalangeal joint fused nicely and not fat, swollen and indurated with angular deformity whereas the second one was deformed as you see here and required additional work. So K-Wires work but they don’t work all the time and there’s certain problem with them and nothing works all the time. You can adjust or adapt to this by building up, put them in cast avoiding any contact with the tip of that pin and we do all of those things. The two-piece titanium implant that I’ve been using since the summer of 2012 is a nice device that I have come to appreciate after a not very long learning curve, that this device is very precise. It has excellent thread purchase of both the middle and the proximal phalanges. It can go together and if I don’t like what I have, I can uncouple it and I can do that as many times as I need to. I can change the angulation of it and get it so that the toe is sitting where I want it. I can compress it so I have bone to bone apposition tight because it locks into place or I can have gap healing depending up on the length pattern of my digits and phalanges as I want them to ultimately end up. So this device, I have come to really like and we went to the company and asked to do a trial and compare it to standard therapy which we’ll talk about after I show you a little bit more about this device. So here it is, it’s two components. Really a third because the ratchening component on the proximal phalangeal segment is a second – is another separate piece if it’s in there, but there’s a lot of engineering here. It’s made out of titanium, a medical grade type titanium. It got its 510K approval in July of 2012 and the proximal phalangeal component ranges from its most proximal side and from 2.2 millimeters to a 3.2 millimeter diameter. It’s 12.5 millimeters in length. The distal component comes either straight in the sagittal plane or with a 10 degree down angle to create a more natural appearance of the fused toe, which is nice because you can’t really bend a Kirschner wire and then plan to take it out too easily. And a lot of us and we’ve all had the experience where we have a perfectly rectus fused toe and it’s floating off the substrate because of the lack of that down angle which we like to achieve and have come to appreciate. The middle phalangeal component is 11.5 millimeters in length. The distal threaded component ranges from 3.5 to 4.5 or 5.5 millimeters in outer thread diameter depending upon the quality of the bone and the size of that middle phalanx. That’s really the judgment you have to make when you’re looking at that phalanx, which sides you’re going to open up. It has this proximal stem component with these cams that lock into the proximal segment. And all you have to do is compress them together and you can feel it ratchening in as each clip goes by the cam and you can get it as tight as you want, as short as you want or as I said earlier you can affect some gap healing because it’s going to be very stable. If you need to, you can twist the distal segment - the middle phalangeal segment about 45 degrees either counterclockwise or clockwise and that ratchet will unlock and you can uncouple the device. You can then take one implant out and size up if you need to which I’ve done on from time to time. So there’s a lot of control and convenience to the surgeon with it and a lot of precision in where you put the arthrodesis interface and how ultimately the toe lines up. That’s why I’ve come to like it. It comes packaged like you see here. This particular package says it’s the 3.5 middle and 3.2 proximal segment, that’s the standard one but you can size up as I said if you know you have soft bone or you may want to use the bigger middle phalangeal component or if you’re doing revision work, sometimes I’ll go to the bigger device. I tend on younger patients with good strong healthy bone. I tend to prepare and I’ll show you how we prepare the channels for implantation.


    I tend to prepare by not over reaming it as much and counting on the elasticity and strong bone but then I don’t want as wide and implants particularly in the middle phalangeal segment. It comes with its own sterile instrumentation. The device is in the package as well, you have this reversible driver and handle. It’s reversible because one side fits the middle phalangeal component. The other side fits the proximal phalangeal component. You have these cone and cup reamers for each reciprocal surface on the middle phalanx and the proximal phalanx. And the step drill which is really the area where as a surgeon and when I’m teaching my residents now, this is where the learning curve is in. Just how much of that to use and I’ll try to show you some photographs here in a moment, not to over ream it and to count on the strength of the bone to allow for good stability. So this is nice instrumentation right there and here you see the step drill. And I’m drilling the proximal segment and if you look at this step drill, you have – supposed to have a – sorry. Anyway, so I’m using the very narrow, it’s 1.6 millimeter component on the step drill and then that 2.8 millimeter middle segment, I will enter just a little bit with that and then I will feel like I have prepared the proximal segment enough to insert the proximal phalangeal component. Rarely do I take it up to the shoulder of the step drill which is 3.2 millimeter diameter. Understanding that the thread outer diameter is 3.2 millimeters. And then I insert the proximal segment into the proximal phalanx. And then I do the same thing; ream and insert the middle phalangeal component and this is a 4.5 millimeter middle phalangeal component with that 10 degree down angle. So the ultimate stem that’s sticking out is cam shaft stem is – once I couple it with the proximal segment, I have that 10 degree sagittal plane flexion; plantar flexion which makes for a nice anatomic alignment of the toe. So at this point, I’m now going to take the distal segment and insert it into the proximal segment as I’m doing right here and just simply push. You don’t have to twist in any direction. It just locks in and you can feel it snapping as it goes passed that cam on the RevLock section and you can put it as tight as you want. You can have direct apposition or a little bit of gap healing depending upon what is optimal for that particular patient’s deformity. Here’s a patient you see with a second toe that I’m going to straighten out with this device. And here it is after implantation, proximal and distal segments. Here’s another patient and of course, we have to always address the metatarsal phalangeal joint on the corresponding digit to whatever degree the deformity requires. And I do my sequential release all the way up to transfixation which obviously this implant can’t transfixate the metatarsal phalangeal joint. And so you may need to use Kirschner wire stabilization for that or just a Kirschner wire for the whole thing. It depends on what’s going to be best for your patient. So once it’s all released here, you see the proximal phalangeal and the middle phalangeal segments just prepared now for me to insert. Another patient here, doing three toes on him and I have this ready now for insertion. And here, we’re going to snap the thing together. Pushing it axially and now it’s set. And you can’t really pull it apart unless you twist it. We can yank on that toe now and it’s not going to come apart unless you use the RevLock design where you stabilize the proximal segment and rotate the middle phalangeal segment in either direction about 40 to 45 degrees and then it’ll uncouple which is really nice. So here’s one, three toes across the board with an arthroplasty on the fifth. I also did a while on that third metatarsal. You can see the gap between the phalanx base and the metatarsal head as I kind of unloaded that somewhat in regard to it.


    The length is set at whatever I feel it should be and the device is locked even though you have some gapping at different points there. You snap it in as far as you need to, to get the correct amount of – the desired amount of correction. I will say too and I’ve done this before where I’ve over reamed the middle phalanx and entered the distal interphalangeal joint which typically, you don’t want to do. In some cases, you may want to stabilize that but I just caution you against that. And here you see it with a titanium screw – snap off screw in the – while not trying to get dissimilar metals into this patient but releasing the metatarsal phalangeal joint as well. So if you needed to transfixate the metatarsal phalangeal joint, I use K-Wire. I understand the dissimilar metals titanium versus stainless steel. I don’t want them contacting one another and the K-Wire’s only going to be in there for four to six weeks temporary. So I don’t think I’m going to have much galvanic degradation of my devices as you see here. And there’s less subjective – there’s less flexure on that type of K-Wire than there is with the axial K-Wire going out the tip of the toe. It’s pretty easy to remove. But obviously, if we don’t need to have a Kirschner wire in this, we’re not going to. So here is our randomized controlled trial. Our aim was to compare this two-piece titanium implant to K-Wire fixation. Our hypotheses were that patients would find the implant more favorable early in the post-op phase because that’s when the K-Wire’s there, early in the post-op phase. And we’d have earlier and more complete fusion mass consolidation with this implant versus the common pseudoarthrosis, fibro pesudoarthrosis we get with the K-Wire. And then we’d have fewer post-operative complications with the implant because we wouldn’t have this pin sticking out of the toe which can lead to certain complications. The randomized controlled trial is a clinical experiment. You heard a lot about it earlier today. The beauty of it is that there’s no selection bias. Coin flip decides who gets the K-Wire and who gets the titanium two-piece implant. So there’s no bias and me saying, “You get it,” or, “You get it.” But we let chance do that and this was done with a random number generator, sealed envelopes and as patients enrolled, they got one method or the other. Non-participating surgeons, surgeons that didn’t do the surgery measured our X-rays. We used valid health measurements for quality of life related to the foot as subjective outcome measurements. And I did all the stats and I didn’t do – and none of my patients – surgical patients were in this cohort of patients. This was sponsored by Nextremity Solutions. They paid the primary investigators, Adam Landsman, myself, and Rick Jay. Rick Jay was our principle investigator. He got me involved with this project. And they supplied the implants. The sponsor, Nextremity Solutions, had absolutely no role in the design of the study. I designed it. I laid it out in a way that I learned how to do a randomized controlled trial and to limit bias so that hopefully we have some valid information that we can use in treating our patients. They had no say in the design, the sample size determination, who got what, who was in the study, the intervention obviously, that was chance and we decided to compare it to a K-Wire and not some other implant. Outcomes assessment, data procurement or storage and statistical analysis or preparation of the report. If the report said it was a bad device, we were still going to publish it and they consented to that. We had a contract. They allowed us to maintain clinical equipoise which means we don’t know which device is better. We want to try and find out so that we can then either use it or not and that’s just a really good thing if you have a company that’s going to sponsor and allow you to run an investigation like this. To me, that’s very important. A lot of co-investigators, this was a multicenter study. As I said, Rick Jay was our principle investigator and you can see some of the participants in the investigation who were not actually doing the surgery. Most of these people were providing patients. And the inclusion criteria where it had to be 18 to 75 years of age, isolated painful intermediate, second, third or fourth hammertoe.


    You could have metatarsal phalangeal joint contraction but it couldn’t be dislocation crossover deformity or requiring transfixation with a K-Wire. Unilateral deformity, metatarsal phalangeal joint could not require transfixation which limits some of the generalizability of this because we often treat hammertoes that need to transfixate the metatarsal phalangeal joint. The participant had to be able to volunteer to participate and say after informed consent, “Yeah, flip a coin. Decide what I get.” That takes a lot of explanation and a lot of time and effort on an investigators parts and a lot of willingness on the part of the participant. Previous surgery, infection, injury to the toe was an exclusion, crossover or dislocated MPJs. Any of these diseases that could affect healing; sensation because we wanted subjective assessments of pain and how the patient was doing or anything that could impede bone metabolism was an exclusion. If they needed for normal walking an assistive device like crutches or a cane. Narcotic dependence, anti-seizure requirements - medication because it affects sensation. And then certainly, titanium allergy. So this one where we eliminate the crossover or dislocated metatarsal phalangeal joint, that’s what we call bias toward no difference between the effects because I feel like when you transfixate that metatarsal phalangeal joint, your risk of complications increases and that would also affect how we measure the radiographs in the post-operative phase. So the intervention was to use an extra hammertoe implants system or K-Wire and chance decided who got it. We used the Bristol Foot Score which is a 15 item subjective patient centered outcome measurement for foot related quality of life. The lower the score, the better. If you’re having a lot more pain and a lot more disability, then you’re up toward the higher end. We also used a well-established foot function index. Similarly, lower score’s better. 23 is the minimum, 230 is the maximum and that has domains of pain, disability and activity level. We also looked radiographic fusion determining whether there was a gap, contact and ultimately when union occurred and none of the surgeons assessed these radiographs. None of the operating surgeons. They were done by my research fellows and other individuals who weren’t participating in the surgery. And we just measured when a complication occurred. Define complications. So we had 95 toes and 91 patients enrolled between June 2012 and October of 2013, 4.5% of these patients underwent surgeries on two separate toes. We had a mean age of almost 59 years of age, ranging from 18 to 84 years. About an 80% female to male ratio and that’s common with elective reconstructive hammertoe and bunion forefoot surgical proportions that we see. Random allocation to treatments. You’d expect about a 50:50 distribution if you had 30 or more samples and we certainly did. One patient was allocated by randomization to the Nextra device but got a K-Wire instead. In this type of clinical experiment, if you randomize someone to a treatment and they get the other treatment, you analyze them as if they got their treatment they were randomized to. That’s called intention to treat. It eliminates selection bias and a bias toward no difference. So if you still see a clinical or a statistically significant difference, if you hadn’t done that, it’d probably be an even bigger difference. And excluding the transfixation on metatarsal phalangeal joint in this particular patient here, what we measured in reality is probably bigger if we hadn’t of done these two things. We found that there were no pin tract or post-op infections in any of these toes – any of these patients. We had no symptomatic non-unions or complications that required return to the OR in either group. And here are some – I’ll go through some of these results tables just to show you the actual results and then summarize it for us in the next couple of minutes. So on the Bristol Foot Score, you’d expect a subjective quality of life measurement to get worse immediately after surgery and then as they heal for it to get better. And that’s exactly what happened. And at each stage, comparing baseline to one week, three weeks, six weeks, three months and six months post-op, that’s exactly what happened and each difference was statistically significant.


    What’s that mean? It means that the difference was not likely due to a chance observation, due to our intervention. With regards to the foot function index? Same thing. We saw it get worse in the first week and then get better. Lower scores each phase between each time period, statistically significantly better. I’ll say it again, “What’s statistically significant mean?” Not likely a chance observation but within our willingness to be wrong, 1 out of 25 that that 5% - 1 out of 20 observations that that 5% level or probability 0.005 or less – 0.05 or less, it was better. So it’s likely due to the surgery that they’re getting better, that that score is better. In regard to Bristol Foot Score by treatment group. Now I’m comparing the Nextremity Solutions implant to the K-Wire. At baseline, you don’t want to see any difference in these patients and randomization should equally balance this as you see in the first – second column there. No statistically significant or clinically significant difference and then as you go through the healing phase, one week, three, six weeks up to three and six months. You see that the only statistically significant difference was that the patients were better in regard to Bristol Foot Score at that first week post-op. They didn’t have a K-Wire sticking out of their toe. And then down the road at three months, there was a clinically significant improvement difference between the two treatments. In regards to the composite foot function index, the only difference is there were no statistically significant differences at any time period. However, there were substantial clinically significant. We defined clinical significance as a 10 to 15% improvement or difference between one treatment and the other because that in general is enough to make physicians and surgeons choose one treatment over another. And you can see the red highlighted differences there, at one and three weeks and then again at three and six months. I can’t explain why it wasn’t that way at six weeks but that’s just what we observed. It could be related to our sample size. As far as pain early on and then again at six weeks, statistically significantly less pain with the implantable devices compared to the K-Wire. I can understand that early on because they don’t have a K-Wire sticking out of their toe. At six weeks, most of them had the K-Wire out by that time. And then as far as disability, again, clinically significantly better with the implantable devices compared to the K-Wire at one and three weeks and then again at six months post-op. And then in regard to the activity domain of the foot function index, very strong clinically significant improvements with the implantable device as compared to a K-Wire at all time frames except for the six week post-operative time period. When we looked at fusion and had our non-participating surgeon investigators assess the radiographs for contact and actual bone bridging, the Nextremity device did better at all phases compared to the K-Wire at every period from one week up to six months post-operative. So closer and earlier arthrodesis or bone healing with the implantable device as compared to the K-Wire. I think we all know that because we see these fibrous pseudoarthrosis all the time. We have a patient that’s okay with that who may have a little bit of a fatter toe, a little bit more of a deviated toe in the transverse plane, but you’re not seeing that with the Nextremity device. So to summarize these results, in both groups the Bristol Foot Score and the foot function index improved after initial worsening over the healing phase. Thereafter, the implantable device displayed statistically and clinically significantly better Bristol Foot Scores at one and three weeks. Clinically significantly better composite foot function index at one and three weeks and then at three and six months. Clinically significantly better foot function index pain domain at one and six weeks, clinically significantly better foot function index disability domain at one and three weeks and six months post-operative and at every phase, better radiographic evidence of healing and apposition.


    So this was a randomized controlled trial where we had full reign of the data and how we wanted to do this. There are some limitations to it. The biggest one is blinding – lack of blinding on the investigators as to what treatment was done and it was hard to do when you have a K-Wire versus an implantable device. And that’s why we used people looking at the radiographs who weren’t participating surgeons but the generalizability is restricted somewhat because we looked at isolated toe and a lot of times we’re doing – in the cases I showed you my own cases, two, three, four toes on the same foot or combined. We also eliminated the metatarsal phalangeal joint which we often have to deal with. The assessors were not blind as I said but we used valid subjective health measurements. We asked the patients how they felt about it and there’s nothing more important than that. X-ray may look great, patient may hate it. They feel bad. So it’s important to use valid health measurements like that. And our follow-up duration was six months, long enough to get arthrodesis in the majority of cases. So in summary, there were no infections, symptomatic, non-unions or need to return to the OR in either group. The two-piece titanium implant was favored by patients statistically and clinically significantly more than the K-Wire early three months in the post-operative phase and clinically significantly better up to six months after hammertoe surgery. I grade this as a level two clinical evidence. A randomized controlled trial can be level one but because of the lack of blinding on outcomes assessment, I grade it as two and certainly, more clinical research is necessary. Thank you.