I want to welcome you to the research symposium that we have today. I want to just highlight a couple of things that they've asked me to comment on. First of all, those individuals who are here are obviously committed to research. So please don't forget, yes, you people, thank you. Don't forget to apply for the AOFAS research grants because those are available and they are fantastic. They're available funding all the way up to, I believe, $20,000. So look on the website and try and apply for that. Also you'll get some email communications for AOFAS members in regards to this. So it's really my extreme pleasure to welcome you to this research forum where we have four outstanding orthopedic surgeons and PhDs that are going to educate us about foot and ankle topics. And we're going to start today. And I want to mention thanks to John Katz who actually did the organization from this, one of our partners back at the University of Rochester, and he did all the hard work. So I want to thank him. We're going to start today by hearing from Dr. Tim Daniels from Toronto about high quality evidence in arthritis. I think they're still looking for our slides. I'd like to echo the welcome. Research is essential. I mean, you know, in the legal system they have the Supreme Court. The Supreme Court makes their ruling, and if anyone wants to deviate from that ruling they have to have a very specific reason, and they actually have to challenge the Supreme Court. In medicine we've got level one studies, but unfortunately they don't carry the same weight. We all know how level one studies can be trashed, both privately and publicly. But in my view, our Supreme Court is science, and the only way we understand science is through research. Still waiting for my slides. So we're entering the Supreme Court of science. It's technically downloading the slides. They said they just need a minute or two, and then they'll be available. Let me announce the other thing I'm supposed to announce then. Remember that the foundation updates and awards are going to be following this. And the AOFAS presidential updates and awards, additionally, will start at 310 after this symposium. And then that'll be followed by the Diversity Equity Inclusion Symposium, which you cannot miss. It's going to be awesome. So remember to put that into your schedule and highlight it on the app. All right. Does anybody have a good joke out there? Just kidding. I hear Sheldon. Sheldon, you got a good joke out there? Here we go. Yay. Another great researcher. Whatever, rips and rips the girl's head off. Bartender goes, and I don't serve drug addicts either. What? Yeah, that was a barbiturate. Oh. That kind of was clean, maybe not. You need to go to the diversity equity inclusion symposium that's going to be coming up later, so that you can figure out whether or not that was appropriate. Yep, Mark, you're up. All right, so I hope I get this right, because I'm bad at telling jokes. But my wife does a great job of it. So there's a lady driving down the highway, and she's speeding. She gets pulled over by the police. The police officer comes to the door and says, ma'am, you've been speeding. She says, yep, I know I've been speeding. And I don't own this car. He goes, well, let me see your registration. She says, I don't own the car. I stole it. You stole it? She says, yes, I stole it from my neighbor, and he's in the trunk. I killed him. I cut him up and killed him. He's in the trunk. And he goes, well, ma'am. He steps back, and he says, I'm calling for backup, calling for backup. So the chief of police shows up, and he says to the lady, OK, lady, stay in your car. Let's see your hands. He checks the trunk, and he goes, there's no body in there. She goes, yes, I know. Why would there be a body in there? He said, well, he said you killed someone. You did not kill someone. He said, well, let me see your registration. She goes, yes, here's my registration. He said, you stole the car. I didn't steal this car. This is my car. She said, the next thing you know, he's going to tell you I was speeding. All right, that was pretty darn good, though. Thanks, Mark. OK, so now we're going to hear from Tim Daniels, because his slides are available. Thank you for sticking with us. Thanks. All right, so I'd like to thank Ellie Pinsker, who helped me out tremendously. I do have a conflict, ortho evidence. I do have a small investment in that company. I'm going to mention it briefly, because it does serve a purpose for this talk. My other conflicts are in the program, and they're not part of this talk. So high quality research leads to new treatments and therapies, and improved understanding of impact of health conditions and their effects, and the most effective use of existing treatments. Research fills gaps of our knowledge, and it changes the way we provide treatment. It has numerous ways to approach the same research subject, which provide complementary insights. And no matter what the study design, it should be methodical and rigorous. The thing that I find somewhat frustrating with research is that it takes years and years for a new concept to actually catch any traction. And I think I've heard it can take up to 17 years for a new idea to become mainstream. But it is, so if you do have a good idea, and you want to make it a popular item, the most important thing to understand is not only do you have to do the research, but you need to be patient and persistent. And for goodness sakes, don't take things personal. Select a good research question. It should be focused, clear, important, and answerable. There are research questions that we are unable to answer. Look at the difficulty they've had with fatal pulmonary embolus, and trying to figure out what is the correct anticoagulation for patients undergoing orthopedic surgery. This is a very important research question, but it is so hard to answer, because in order to get a powered study, you need tens of thousands of patients, and you need to follow them appropriately. You can, fortunately nowadays, you can check the research that has been done. There's no point in picking a topic where somebody has already gone through it in detail, and come up with an appropriate conclusion, and these aren't always obvious. There's resources such as PubMed, the Cochrane Database. Prospero is an international prospective registry for systematic reviews. This is a screenshot of the Cochrane review section. You can note here, start for free. Most of these websites, there's not any cost, and you can go into them and educate yourself. This one actually does cost a bit of money, but the only reason I'm bringing it up is this is the first website of its kind where you can actually go in, ask a research question, and do a meta-analysis based on this website alone, because of the way that they enter the data from all the orthopedic musculoskeletal papers that are being presented, or published, I should say. Assemble the right team, and I can't emphasize this enough, and it's not easy, and it takes time. I started practice in 1993. I published a number of papers on my own. It was a lot of work. It took me approximately 10 years to get my first research assistant. I now have a research team around me. It grew very slowly, but I can tell you right now, without any doubt, I could not do the research that I'm doing without that team that works around me. Treat them well, pay them well, support them. They are invaluable. Seek methodological input from epidemiologists, methodologists, and statisticians. Can't emphasize this enough. We have a PhD statistician epidemiologist that's employed at our center, and her work is invaluable. I actually hired her now for the group, and everyone is using her for all their research studies, and the quality of the studies that we design have improved substantially, simply because we have this one person. Now, at other institutions, and before we hired this individual, we would go to the hospital-based or the university-based epidemiologists. They are very intelligent people. They work well, but they're never as devoted as that person that works for your group. So if you are in an orthopedic group interested in establishing a research platform, by all means, look for that person, and if you find the right one, hire them. Discuss your research question, hypothesis, proposed research design, variables to collect, sample size, and appropriate data analysis with these individuals. It's essential. Design your studies so that you are assured that it's gonna answer the question that you're asking effectively. Identify and minimize source of bias. Garbage in is garbage out, and the most important thing is to figure out ahead of time what you're gonna do with incomplete data and patients that withdraw from the study. It is one of the most common errors made in research, and there's some research studies where just adding five more patients to the study changes your hypothesis or your results completely. And Mo Banderi has an excellent presentation on this, and he gives the example of the randomized trial that they did in Canada. The COTS, which is Canadian Orthopedic Trauma Association, did a trial looking at reamed and unreamed tibial nails, and he shows very elegantly how if they would have stopped the study before a certain number of enrollment, they would have come up with an entirely different conclusion than they did when they continued the study, got it properly powered. The conclusions were polar opposite, and you can imagine how confusing that would be if we were coming up with conclusions that were actually not true. And this leads into ensuring that the sample size is adequate. I do a lot of reviews for a lot of journals, and it's probably one of the more common reasons that papers rejected. Nobody did a power analysis. Nobody did just the minimal amount of work that's necessary to make sure that study is powered, and particularly if it's a comparative study. If you're trying to show that one treatment method is better than the other, that study better be well-powered or it probably won't get published. And select outcome measures that are clinically relevant. There's no agreed set standardized outcome measure across the board in foot and ankle trials, disorders of the foot and ankle. Outcomes measures vary considerably among studies, and they do limit the ability to compare findings with other studies. So you need to, first of all, ask the question, work with your epidemiologist in terms of exactly what you need for power analysis and adequate data time. And then you need to make sure that the outcome score that you're using is the outcome score that is specific to your study. So for example, a sports outcome score probably isn't all that great when you're doing an analysis or a study on the diabetic patient population. I only learned about this. OMERACT is a working group that will actually help you determine which outcome scores are the best for your study. When I started our database on ankle arthritis, there were two validated scores for foot and ankle. There was the SF-36 and there was the AO outcome, foot and ankle outcome score, only two. Now there are numerous validated outcome scores for foot and ankle. This is the website for OMERACT, and as far as I'm aware, you can go to this website, interact with these individuals, and figure out which one of the, I wouldn't say hundreds, but there's gotta be at least 60 different outcome scores for foot and ankle, most of them validated, is the best outcome score for your study. Outcome measure selection, consider what you want to measure, for what purpose, prediction, evaluation, description, discrimination, and for which population, and look at patient-reported outcome measures to support the evidence, both in validity, reliability, responsiveness, and interpretability. I'm fortunate enough to have an individual like our PhD working for us, who basically did her PhD on outcome scores, and it is very easy to just basically throw that into her camp, and say, what do we need to collect in order to answer this question? That website is something that you can use if you don't have an individual like that in your institution. So there's now checklists when you're performing a study, and if you're gonna present it to the journal, and Foot & Ankle International is just starting this, it's a study checklist. If you don't check off the various boxes that are required, you're not gonna get that study published. And I'm very jealous looking at this list right now, because again, 20 years ago, hardly any of this stuff was available. But now, I'm not gonna go through the list, but there is a whole number of websites that will help determine, help you determine whether or not all the parameters of the study check the boxes, and if you're going to get the appropriate answer from the way that you have proposed your methodology. I'm almost done here, Judy. Here's an example of a concert website, and I can assure you that Dr. Altman is far more approachable than he looks here. So internet-based collection is paramount. The paper trail, it just doesn't work. Our database, our ankle arthritis database went from about 35% participation to over 70% the minute we switched to a web-based format. So I'll end with my personal tips. Don't, do not get overwhelmed with the potential pitfalls. I mean, there is so, research is complex, and if you stall because you're worried that if you do this, something bad might happen on the other end, you're never gonna get started. So do your proper due diligence, but then just get started. Do your own investigations on the validity of scoring systems. This is not all that hard to do, and for example, we're looking at the forgotten ankle score, and we can actually take scores from the hip and knee and validate it ourselves to apply to foot and ankle. Online is the way to go. Don't be too robust in your outcome measures. Anything that takes the patient over 20 minutes to complete, the compliance and participation levels drop substantially, and there's no secret. Research is first of all asking the right question, and it's work. If you're willing to put the work into it, it is extremely rewarding. Thank you. Thanks a lot, Tim. Thanks for giving us the data elements, and the guidelines. So now, Mark, talk to us a little bit about the sponsored and investigator-initiated studies and how we can tie this together in the ankle. So Mark Glazebrook, Halifax. Thank you, Judy, and thank you all for coming. This is gonna be a lot easier than telling jokes because I was really nervous telling my joke. This is simple stuff here. Uh-oh, looks like they're slow. I hope I don't have to tell another joke. Talking with slides is much easier than trying to remember the intricacies of the joke. So what I'm gonna talk to you about is kind of my experience with two types of research I do. I do industry-driven research, or third-party-driven research, and I do investigator-driven research. And you probably can imagine that investigator-driven research, what I wanna do and the questions I wanna answer is the most important. But what I'm gonna try to show you in this talk is I'm gonna try to show you why they both are very important. And because, you know, so many times my residents say, well, this is an industry study, so it's no good. Well, that's just wrong, right? And I'm gonna tell you why that's wrong in many cases. And I'm, you know, there certainly is bias and something we have to be aware of, but bias is something that can actually be measured. And I'm gonna try to help you understand how we measure it if we do science properly, I hope, given I get some slides here in a moment. So what happened was, just so you know why there's this kerfuffle about the slides is, apparently the committee had us in a different room and put all of our slide decks into a different room. And they almost asked us to change rooms, but then we thought, how would you guys know to change rooms? So they're now shuffling the slides to get things back, and here we go, great. Judy, do I still get, oh, my clock hasn't started yet. Thank you. Okay, so here's my disclosures. We had a long discussion about disclosures. It's very important to make your disclosures. I'm gonna bring your attention to the fact that I am a consultant for Stryker, and I will be talking about a Stryker device in this talk, but I think you'll see in a very generic way. I wanna talk to you about the principles of research, which I'm sure you all know, but we gotta remember these, okay? I'm gonna give you a good example from my practice of investigator-initiated study, a good example of sponsor-initiated study, and then I'm gonna remind you to do the highest study, highest quality possible. So please put your hand up for number one or number two. I'm gonna ask you this question. Which study is best? You have a choice. It's an industry-sponsored study. It's a level one RCT. Now you have an investigator-initiated study. It's a level two prospective comparison study. So hands for number one being the best. Please put your hand up for one of them. So some people think number one is the best, and some people, number two, the rest of you, hands up. Okay, all right, so we got something to teach people here, I think, all right? Let's talk about the principles of research. This man is one of my heroes. He's from the University of Toronto. He was the guy that introduced the level of evidence to the Journal of Bone and Joint Surgery, which almost every journal in medicine now is starting to adopt. He didn't set this systematic way of measuring quality of studies, but he popularized it. There were other people at his university in epidemiology that did the original work. But here it is, and I think we all know this, a level one, two, three, or four, or five study. And these are the definitions. I keep this slide on my desktop because I'm constantly looking at, is this study a really level one or level two? And this is a better way of thinking about it, okay? This is the pyramid, the hierarchy of evidence. I think we all know that the best trial is a randomized control trial. That is the highest quality study. But it's only the best available evidence if it's done well, and Tim touched on that with assessing studies. So how do we know it's done well? We have to measure the quality of the RCT, and there's many ways to measure this. This is a simple way. This is the Cochran Risk of Bias tool. These are six items that we all think about when we're reading papers, you know? Did they do the sample size right? And, you know, things that make a paper good. But if you really wanna drill into it, you'll use the CONSORT statement. A CONSORT statement is a 25-item checklist. So if you're a good reviewer and you're looking at an RCT for one of your journals, you gotta get this out and do a really good job. Most of us don't, let's be honest. We kinda know what this is, and this is very arduous. Look at the detail here. However, remember, this is how you judge a RCT. And if it meets all these criteria, if they do a good job in all these criteria, they get to be a level one. If they don't, and it's an RCT, sometimes an RCT is a level two, maybe even a level three. This is the job of the journal and the reviewers to make the quality of the study appropriately. So remember, a level one isn't an RCT. A level one is an RCT done well, and this is how you measure it. Okay, so it's a principle. If you've got a level one study and it's done well, it meets the Cochran checklist or the consort checklist, it's science, and if you don't believe it, then we might as well just stop now. Let's just do lousy research and let's see who can promote it the best. All right, let's see who gets the mic and can say about it. Scientific principles must be followed if we're doing research. If not, please don't do research, okay? Now, you don't have to do RCTs. You can do whatever you want, but it's gotta be done properly and it's gotta be assessed properly. All right, quick examples, and we won't get in on the details here. Here's a bunch of rascals that I don't know anyone should trust. All these young guys here from Canada, they decided to do a study way back on ankle arthritis, and we got a little traction with this. In the end, there was four centers, six surgeons. Here are the mugshots of the rascals that participated. Study went pretty well. We did a good job. This was a question that we thought was important about 20 years ago, which is best, fusion versus ankle replacement? We know it's more complex than which is best, but we started out to tackle this. This was a prospective level two study, not a randomized control trial. It started in 2004. You had to fail non-operative treatment. Then the surgeon and the patient decided whether you got a fusion or a replacement. The inclusion and exclusion criteria were appropriate. Let's just take that for granted. We thought a little bit about this, and we think we got the right patients. As Tim said, we used appropriate clinical outcome scores at that time. SF36 AOS were two validated scores. We also looked at complications, and we became a little bit well-known for our complication system because we did a good job of that. The data's been collected since 2004, and we now in our base have over 1,500 patients. This is true investigator-driven. This was a bunch of Canadian orthopedic foot and ankle surgeons that wanted to answer this question. No funding. This paper went well. The original paper went well, was published in the JBJS, and we won an award for it, the Roger Mann Award. Our five-year intermediate study, published under Tim's primary authorship, showed that the outcomes were the same. Both operations worked well. First time that we heard this on pretty good data. We also showed that there was more complications with total ankle. I don't care about the message. We can let the arthritis symposium talk about whether we should be doing these or not. The point is, it was a level two study, done by people who wanted to answer a very honest question. Now let's talk about PDGF, bone graft. This is another study we did. This was an RCT level one. This did well as well. This was another paper that was published in the Journal of Bone and Joint Surgery. In this paper it was a prospective randomized control trial, two-to-one, PDGF versus autograft. Here are the rascals that did this study. I'm sure you know a lot of them. There was a lot of good surgeons involved in this, bar Tim and I. And they did this study and recruited the patients and followed good principles. If you look at the consort statement, most if not all of them were ticked. And it was a true level one study. This is a box with the design, 414 patients. It was properly sample sized. The indication was appropriate. We had a hypothesis with a primary endpoints and secondary endpoints. This study was done really well. And I'm going to tell you another reason I know it was done well, because the FDA eye in the sky was watching this study. So when you're doing a pivotal trial for the FDA, the FDA is coming to your sites, inspecting your sites on the ground, making sure your data collection is reliable and precise. Because the FDA doesn't want to make a mistake. So a pivotal trial is a RCT level one pivotal trial is really getting up there. It's hard to deny that that's not a real well done study. Here's the results. This was just PDGF was shown to be as good as autologous bone graft. So, you know, good message, good level one study. The results were persisted at 52 weeks. Again, the message of whether we're doing PDGF or autologous bone graft is not what we're talking about today. It's an example of an RCT done well, a level one study. More equivalent results from this study. So which study is best? I now go back and challenge you. Is it a level one randomized control trial, industry driven, or does a level two prospective study, investigator driven? I hope we all agree a level one study, whether it's funded by industry or my mom and dad, is a quality study if it meets the concert statement. So I challenge you now to do good science. I'm not saying you have to do level one studies, but you have to do the best available, you have to follow the best available evidence. A randomized control trial is the best available evidence if it's a level one study, okay? And why is that? Because it has less bias. Bias is a systematic deviation from the truth, and true science is void of bias if it's done right, and the consort statement will guide you to that. So, in conclusion, all studies are good. Well, most studies are good. They all have a different quality, and the levels of evidence, if followed appropriately, will tell you about their quality. The scientific principles are solid. You have to follow those, and you can't argue with them, or we shouldn't be doing science. Studies can be investigative initiated, or they can be third-party initiated. It doesn't matter whose idea it is, long as they do the science well. Regardless of who's paying for the study, you must ask important clinical questions, and do the highest quality research possible. I challenge you all to do that, and follow the principles of science. Thank you very much for your attention. Thanks, Mark. So now we're going to hear from Sam Flemister from University Rochester, and he's going to talk about the utility of weight-bearing CT and ankle arthritis. Thanks, Sam. All right, well, my slides are up, so no time to tell jokes. So I have nothing to disclose. So I'm going to talk about weight-bearing CT scans, and what we can do with these. Here's just some pictures of different types of weight-bearing CT scanners, and these use cone beam technology, and if you've been to the dentist lately, you know, they'll put your chin in this thing, and then have something rotate around it, and that's the detector. So this, the detector rotates around the patient, and although we live by plane radiographs, the plane radiographs can't always give us the 3D images we need, such as finding this occult fracture at the base of the first metatarsal on the lateral. So when you look at weight-bearing CT scans for standard multi-detector CTs, we think that there's radiation exposure is 10 to 66 percent less, depending on what study you want to read. There's decreased acquisition time. I think our techs tell us that they can get this done in about 15 minutes. There's less space requirement. It fits in a simple x-ray bay in our office. The study by Richter recently looked at over 11,000 weight-bearing CT scans, and found a 10 percent decrease in radiation dose, and a 77 percent decrease in acquisition time, and this all led to profit per patient. But when we look at the negatives of weight-bearing CT scan versus weight-bearing x-rays, while there's increased radiation, there's certainly increased upfront costs and increased cost per study. Overall use is somewhat limited. You know, in our system, our foot and ankle division and the hand division mostly use this weight-bearing CT scanner, and weight-bearing x-rays can be used to diagnose and treat many, many foot and ankle pathology. So when we look at what is a weight-bearing CT scan being used in research most commonly now, well, flat foot deformity, overall looking at hind foot, hallux valgus, especially first-rate pronation. There's some studies on syndesmotic injury, the cavus foot, and ankle instability, and notice I didn't mention ankle arthritis much in there, because there's not much public literature on ankle arthritis in the use of a weight-bearing CT scanner. So this study goes back to 2011, and this was a load-bearing CT scan versus a non-weight-bearing CT scan, and they looked at the flat foot deformity in the position of the talus and the vicula and the calcaneus, and they were significantly different when loaded. This was a simulated weight-bearing, so it was probably only about 70 percent of full weight-bearing. This is another study done a few years later. This looked at controls in 23 patients with various stages of acquired adult flat foot deformity. Again, the weight-bearing CT scan showed differences in all those angular measurements we usually use for flat foot deformity, and much better than plane radiographs. This is a study that compared weight-bearing CT with non-weight-bearing CT, again looked at those angular measurements. They had a high inter- and intra-observer reliability, and again found weight-bearing CT scan better demonstrated the severity of the deformity. And impingement, we have these patients, they all come in with valgus, they have lateral foot pain, so where's the pain coming from? Well, this study looked at the subfibular impingement and found about 35 percent was calcaneal fibular, 38 percent was more of that talus-calcaneal-sinus-tarsae-type impingement, and the remainder was soft tissue impingement. Again, looking at alignment, weight-bearing CT scan used to compare stage 2 flat foot patients to controls. They found that the stage 2 patients had at least 10 degrees more of subtalar valgus, and another study looked at subtalar valgus in stage 2 flat foot, and noted it was underestimated on non-weight-bearing CTs when compared with weight-bearing CTs. And so these are other studies that are more recent, even better studies, and they all indicate that there's better evaluation of flat foot deformity, both preoperatively and postoperatively, with the use of a weight-bearing CT scanner. And this led this consensus group studying the progressive collapsing foot deformity to recommend use of a weight-bearing CT scanner, if available, in your preoperative evaluation of flat foot deformity. Look at what other things have been used for hallux valgus deformity, especially first-rate pronation, but there's been some studies on mid-foot stability and the role of sesamoid position. The first-rate pronation's really been studied. They've used some reconstruction 3D studies, and then looking at the alpha angle, or the angle of the sesamoids in the metatarsal. And these two studies that are, again, recent, both found better correlation with weight-bearing CT scan versus plain x-rays. And I was just looking through the posters, and there's studies looking at pronation, more and more pronation, all using weight-bearing CT scans. So what I didn't talk about was use of weight-bearing CT scan in ankle arthritis, because there's really not much published literature. And I say published, because, again, I was just out there looking at the posters and looking through some of the talks, and there's a lot of people using, now, weight-bearing CT scanners to look at ankle arthritis and evaluate the ankle more. And I think we're going to have a lot more publications coming out on that. So, basically, what I'm going to talk about is my experience with the weight-bearing CT scanner. So, in reality, for end-stage osteoarthritis, I don't actually use a weight-bearing CT scanner very much, because the total ankle system I use has a special CT protocol that's a non-weight-bearing protocol, so I use that. I'm not going to get two CT scans on patients. However, I think for complex deformities, it can be very helpful. This is a 53-year-old who presented with fibular hemimelia, so she's still at her leg at age 53, but she was having worse and worse pain over the last couple of years. You can see she has a tremendous valgus deformity, but if you look at the lateral, her foot looks plantigrade. You know, a weight-bearing CT scan told me a lot about this. You can see that the hind foot is all the way over, but it's relatively congruent and not much arthritis. You can see where she's loading this, and so I'm like, well, what do I do with this? Do I try to rearrange her ankle and fuse it, or do I just fuse it? You know, I think the CT scanner helped me a lot. I ended up fusing that valgus ankle in situ, but the only thing I felt I really—deformity I really had to correct was the hind foot, and I moved the heel over, and the weight-bearing CT scan helped a lot with that decision-making. I think it's great for complex deformities around the ankle. You can see this various subfibular impingement and how much arthritis there, but what we can do, we can look at that ankle. We know how much deformity is in that ankle, and it's starting to get a little bit valgus, right, and that's going to be important when we go to do our hind foot fusion and make sure we don't make it go into any further valgus. So I think you can get a lot of information. So this is a patient who I fused her subtalar joint after a calcaneal fracture 15 years ago, and she was doing very well. I had to remove one screw about two years after I did it, but she did fine. Now she came with this anterior lateral ankle pain. So I got a weight-bearing CT scan, and I thought this might be ankle arthritis, but in reality, she has some valgus. She's got a little bit of ankle arthritis, but what's bothering her is that one osteophyte right there. Now I did mention that maybe we could move her hind foot over at the same time. She said no to that, and we just took off her bone spurs, and she was fine. This is a patient where I'm showing a lot of my complications because that's where I initially get a lot of these CT scans. This is a patient I did an opening wedge distal tibial osteotomy for a varus ankle, but what does this weight-bearing CT scan tell me? Well, it tells me that she's got a non-union, which is what I suspected, but overall I think her ankle alignment is pretty good, and there's not much arthritis there. So all I had to do is go back in and revise her non-union, and she did fine. I think you can get a lot of information about osteochondral lesions of the talar dome. This is a failed allograft, a failed bulk allograft, and you can still see there's a lot of ankle left. Now I'm not sure what I do with this, but there's still a whole lot of good ankle left, and certainly doesn't make me want to go in there and do a fusion. If you do look at some literature, there is literature on the syndesmosis, and this actually says non-weight-bearing CT scans are better at identifying syndesmosis injury versus a weight-bearing CT, and these were all patients who were confirmed by ankle arthroscopy. So I think, again, you can get your degree of deformity. I look at this. I know how much valgus there is. I know how much arthritis this is, and this happens to be after a triple arthrodesis. I think these are great for evaluating moderate osteoarthritis and deciding what you want to do to save this ankle, because now you know how much arthritis there is. You know the alignment. You know where the arthritis is. I think it can be helpful when you're looking at painful total ankles, especially those that are early. This tarsal coalition, I think I just threw this in there because I thought it was a pretty neat picture, but that's the kind of things you can get. If you want to look at hardware position, this is a patient who came to me with her painful ankle after a fracture. You know, this tells me, no, that screw's not in the joint, although on the plain x-rays there was concern. It says that she doesn't have that much arthritis. Maybe she's getting bothered by the fibula. You can look at subtle osteoarthritis. If you look at this one on the right here, you know, there's a non-union of the medial mal, the syndesmosis isn't reduced, the ankle's going into valgus, and, you know, we know this is weight-bearing. This is really, really helpful, I think, when you're trying to decide what to do with this. So there's not a lot on actual ankle arthritis research, but, you know, when you look at what has been done with the flat foot and what's being done with Hallux valgus deformities, there's multiple research opportunities, so you should jump on them quick. It helps you evaluate deformity understand the deformity, and you have to pair what you're doing with outcome measures. So I think, clinically, the weight-bearing CT scan is better for moderate ankle versus severe osteoarthritis. I think it helps you identify surrounding joint problems. It opens up your treatment options and helps you evaluate outcomes. And 47 seconds left, Judy. Thanks, Sam. So our next speaker doesn't need any introduction, since we just heard an outstanding talk from him previously, but Dr. Anderson's going to talk to us about pathomechanics of ankle OA. Thank you very much. I want to apologize in advance. There's a little overlap between this and my last presentation, but I seldom come to the podium so well teed up by my prior speaker, so we're going to talk a little bit about pathomechanics of ankle OA, work we've done at the University of Iowa, and I'm happy to be here to be able to present that work. Here are our disclosures. Maybe what I should start a talk like this with is telling you, as a disclosure, I'm a PhD engineer, which means I like to measure things. So you'll see a lot of information here about measurements. I guess this audience, I don't need to spend much time on ankle OA, and specifically the prominence of post-traumatic osteoarthritis. We study at the University of Iowa both biological sequelae and the mechanical implications of surgical results in these ankles, but today I'm going to speak mostly just about the pathomechanics or the altered mechanics associated in the joint that relate to post-traumatic osteoarthritis, specifically focusing on the acute fracture severity and the chronic contact stress elevation. So surgeons have long understood that there's a whole spectrum of severity, and they often talk about low energy and high energy fractures, and often they're talking about how high somebody fell from or how fast the car was going, but it turns out fracture mechanics is able to put a number on this. It's called fracture energy. It's a quantity that can be measured objectively, especially in the case of brittle solids like bone. It's actually been known since 1867 that the energy absorption is proportional to the fracture liberated surface area, and I think you can appreciate that with care that can be measured on preoperative CT scans. We spent 20 years developing the capability to do this to where now, and when I say routinely in this slide, we're talking about a three or four hour process, but we can routinely quantify pathomechanics here in terms of the fracture severity. If you have a pre-op CT scan, of course that's not always a standard of care, but weight-bearing CT can also be an opportunity for this. What we were able to show is in a series of 20 patients here colored as columns. The height of these columns are the fracture severity, and the color indicates whether two years after surgical treatment of their tibial pylone fracture, whether they had developed PTOA or not, according to the KL grade. You can see it's very good at predicting who is going to get the PTOA. I'll just emphasize this is based on a preoperative CT scan. So PTOA classification, when we rely on the Kellgren-Lawrence grade, it's difficult in ankles and other smaller joints. It really was developed for the knee back in 1957, but one of the primary features is, of course, joint space width narrowing. Other features include osteophytes and cysts and other things that can be seen well on imaging like weight-bearing CT, but when you rely on plain radiographs and the Kellgren- Lawrence grading, this is subjective, strictly categorical, and it's 2D in nature. That means altogether it poorly captures the early changes, and it's not as useful in these joints other than the knee. So we've turned to weight-bearing CT as a way to look at ankle OA in particular. It's an emergently standardized 3D joint structure and a loaded apposition, and we can measure the 3D joint space width off of this by identifying the subchondral bone surfaces and measuring the distances between those bones. Here's a series of 20 patients. This is 40 patients, excuse me. We're looking at following these patients at 18 months after their treatment. At six month spans, we acquire weight-bearing CT scans. These are from the University of Iowa, Washington, and Utah. Top row, you can see difficult to appreciate on a plain radiograph, a clear posterior malreduction or step-off here, seen on the CT slice. And then we can use modern registration methods to use an intact contralateral ankle to compare how these changes in the joint space width occur over time, and that allows us to take these joint space width maps and turn them into measurements of a change in joint space width, so we really can define joint narrowing in a patient-specific manner. So here's a series of these patients. The intact contralateral, which is often scanned at the same time, allows you to establish a reasonable indicator of the standard joint space width, and then you can see there's a variation across these patients on how much joint space width narrowing you see, going from the least here to the most severe. And here's a study showing how that fracture severity metric I mentioned earlier correlates with this change in joint space narrowing over the course of 12 months. I also mentioned chronic context stress elevation. Patient-specific computational modeling is an area of expertise for me and our group, and we're able to take these post-op CT scans and be able to generate models that allow us to predict context stress. And then we look at that in a reasonable habitual loading. In this case, we modeled a stance phase of gait. We all walk anywhere between one and a half and two million steps a year, perhaps. Accumulating these over time, on your right, you can see an intact contralateral ankle. There's relatively uniform distribution of the context stress, whereas on the fractured side you can see relatively elevated areas. And then if we consider the fact that the context stress is normally healthy on the joint, it actually maintains the health of the articular cartilage, we might start to define a damage threshold over which the cartilage is potentially damaged, and see how that accumulates over time. Because it's not just a matter of how high the context stress is, but how often the joint is exposed to it. So we've developed this context stress time overexposure metric that tries to put that into a single unit, the megapascal second per cycle. And when we take that, the overexposed area, which is the derivative measure of that, and look at how it correlates with KL grade, you can see there's a high correlation and an indication of a threshold value below which you may well be able to reduce the likelihood of PTOA. We've taken what was initially a finite element analysis methodology, which is kind of a standard engineering methodology, relatively bulky for this kind of application, and developed a more expedited discrete element analysis method, which just treats the cartilage as a bed of springs. And then that means we can take a post-op CT scan when we have it, or if we don't have it, a weight bearing CT scan, to be able to generate these predictions of context stress and study how they may correlate with joint space narrowing over time. And in fact that's what these data show. This is an early small sample size. We're beginning to look at how these overexposed areas influence the joint space narrowing in the joint over time. And again, I'd like to add my words of invitation to consider writing a grant proposal to the Arthritis Foundation AOFAS Paragon 28 sponsored ankle OA think tank. The RFA will be coming out soon and we look forward to receiving a number of high quality applications. Thank you. I think we have time for a couple questions. If anybody has any questions that they'd like to bring up. Thank you for that amazing session. It's a true rubric about how to think about approaching research and then for specific diagnoses. Mark, one of the things I've always struggled with with research, obviously level one is what we aim for, but we're often doing retrospective, and this idea of length of follow-up. We always think more is better, but do you ever find that research can be biased and reversed that the people don't generally stop by to see me just because, hey, I just was in the neighborhood and wanted to let you know I'm doing great? Do you think we enrich our patients with poorer outcomes sometimes? Yeah, the first part of that question, I kind of want to unpack that a bit. The first part of the question is you don't have to do level one studies. That's not the message, right? What you have to do is the best available evidence. You have to look at ... One of the key factors to do in research is your literature review. When I get my fellows and residents to do research, their introduction has to have a systematic review of the literature so we know where they are. I don't want to be doing a level three study when there's two level one studies out there. I'm putting in a scientifically inferior study to answer a question, but if there's only level five studies on a certain topic, a level four study is grand. It's the best available evidence. The message isn't do high quality research. It's what you're doing, do it right and get it right. Do what is needed based on a literature review, best available evidence. The second part of that question was about follow up. Can you rephrase that? Yeah, I feel like when you're doing a retrospective study, there's some turning point at which if someone's still following up beyond a year after an ankle fracture, they're probably not doing that well. When we set these thresholds for minimum follow up, I always feel like there's a catch 22 to them. Right. Nothing ruins good outcomes like follow up, but too much follow up can enrich bad outcomes. I was curious how you think about that. Yeah. Okay, so I hear you. The point of that is, and I'll give you a great example, I just presented a paper yesterday on the long term follow up of PDGF and ankle fusions. And guess what? The joints that were fused at two years, they were all fused at 10 years. So that was very unexciting and it maybe enriches that topic, which is unnecessary. But you get to the essence of the quality of research. When you describe a retrospective review at a set time point, you get to choose that time point. And that's why a retrospective review isn't a high quality study, because you can choose the time point, sometimes tailored to the question you want to answer, which may be a bit biased. Remember, you're introducing bias because you're trying to get the answer you want. So a level four study in some cases is exactly what you would suggest. So back to the quality of research. I'll put a plug in for what Tim said, which was everything should be collected electronically. So if anybody knows me, they know that I would be an advocate for pushing up patient reported outcomes so that you could get your information on patients that are even doing well, so that you're not biasing your information based on the patients that aren't doing so well. But it is a really challenging problem in getting follow up for patients. And how much follow up is necessary in order to answer the question is an interesting question, especially when we have materials that we're putting in people that may not last forever like an ankle replacement. Yeah, Tim. It's an excellent point. And the trouble we're, you know, our database has been now 20 years, 21 years. And the trouble we're running into is that patients get diabetes, they start getting hip arthritis, they, you know, some of them run into kidney failure. And the outcome questionnaires really aren't designed to capture those comorbidities. And Charlie's already published on, you know, I mean, ankle arthritis is as bad as diabetes or rheumatoid, I forget what he compared it to. End stage renal disease. End stage renal disease. So you know, it's hard to uncouple everything. And so let's say a patient's doing worse. Is that because they develop renal failure, and then subsequently had a transplant and on and on? Or are they actually doing worse because your surgery is deteriorating? It's not, it's very difficult to tease those points out if you follow up people long enough. I'll just say as somebody who works very closely with clinicians and we're following patients, I think if you have any kind of research partner who's a scientist and you may have some questions, you might be amazed how easy it is to collect data that later could be useful to them. So if you have some potential idea about a mechanistic process, even in the face of doing those clinical studies, if you can just keep some of that information available, we have long-term DDH, developmental dysplasia, the HIP data, that we were able to analyze years later using a mechanical analysis approach just because somebody had the foresight to keep all the x-rays. So I think sometimes you need to also, it's helpful to consider ways in which you might be able to partner with scientists who can take that data and do something in kind of an ancillary study. And that's one of the reasons why I mentioned just at some point you just have to do it. Pre-op data is golden. Even if it isn't great, it's golden. If you've got pre-op data on your patients, man, there's so much you can do with that. So we'll just go through the crowd here, right here on the table, and give us the one tip that you'd like people to walk away from today in regards to doing research. Tim, I think I just heard yours, but why don't you just reiterate what you just said? Well, I mean, first of all, Judy, have fun. Like, I mean, it's got to be fun. It really does. Like, I've had so much fun with our group, and we've had so much fun at these meetings. So, you know, I mean, first of all, don't look at it as work. Have fun with it. And then secondly, get the data. Once you have the data, and you're finding this out with promise, aren't you? Once you have the data, I mean, 20 years from now, you'd be amazed at what you can do with that, even if you don't use it initially. Mark, what's your pearl? I'm with Tim. Have fun. We've had a lot of fun doing research together, and we've had a great relationship. But when you do your studies, do them well. Aspire to the highest quality you can, but remember, it's the best available evidence, not an RCT. Sam? I agree. You should have fun. You know, I'm also the guy who says, if you're going to do the research, you got to be willing to put in the work. You know, it does take time. It does take effort, and you can't just expect somebody else to do it. And I think I've been able to partner with PhDs to get NIH grants, and it's been the best thing in the world. But it's been a – I couldn't expect them to do it all. So, you know, I had to put the work and the good news is most of them pushed me pretty hard. Don? I'll turn it around this way, and I'll just say, I agree. Do the research. Don't let the work of doing the research get in your way, but don't rule out partnering with a scientist when it makes sense. And I'll just add, look to your right and look to your left, because these are the people that are interested in research, and these are the people that would perhaps be willing to partner with you, as I've partnered with the vast majority on this table here. And it is a lot of fun when you get a group together that's willing to work hard and be successful in answering a very important clinical problem for your patients. So having said all that, I think we have to wrap it up, but don't go too far, because we've got the awards, the Foundation Awards next, and it's going to be exciting. Thank you very much. Thank you.

conducting research

orthopedics

ankle-related issues

high-quality research

randomized controlled trials

study design

data collection

weight-bearing CT scans

collaboration between clinicians and scientists