and co-collaborators here. I was actually at a, it's nice to see everyone live again because I was actually at a live meeting with the ACS where we all kind of connected and had this interest in bone loss. So that's where we all met. We kind of discussed this and kind of we've launched some of the projects. So I want to thank everyone. I also want to thank Rebo. So he was at Duke, the first three presenters, by the way, are all Duke affiliated. So it was kind of nice. But he is a good mentor of mine, was there when I was in fellowship. And I liked your slide about the lacrosse players and return to play. Cause now that you've left, I take over lacrosse. So that return to play is going to be a lot better, I think. So this one is just on my answer, shoulder instability. And it's a common problem. We all take care of it as sports physicians. It's got a high instance and we are getting further and further understanding that bone loss, both on the glenoid and humeral side are vital to understand when treatment planning, but also in terms of their outcomes following your treatment. And there's a lot of evidence out there showing that there are bony injuries, even from a first time dislocation. So this study here with the military, looking at first time dislocations, 93% have a heel sac lesion. A followup study showing that the amount of bone loss in that 50% of, over 50% of the patient will have at least 5% bone loss after a single even subluxation or dislocation. And that the mean bone loss at first time dislocation was 6.8% and 22.8 after recurrent. And so, and then 13.5 is a big number, we'll go into another side of that, but 18% of first time dislocators have greater than 13.5%. So bone loss is there, whether you're noticing or not, it's there. Now the history of bone loss, this started off a lot with Burkhardt's work on the inverted pair and surgical outcomes and its effect on surgical outcomes. And so, obviously it shows that the more bone loss and some of the activity related, the worse your outcomes are gonna be. And some of the work that was previously listed here from these esteemed authors, showing the normal value of about 20, 25% was when people would say that a soft tissue alone would not be suffice. But there are some recent studies here, and just two of them were showing that maybe that number is actually too high. Maybe it should be 13.5%. And so there's a couple of studies here showing that, looking at a series of 72 patients in the first one and 50 patients in the second one, and showing that that 13.5% may be that critical number that we should be redefining and reconsidering. And then there are more recent developments is this bipolar concept, first brought on by Yamamoto in the bipolar, and then the on-track concept brought on by Burkhardt's group. And so even for, you know, open bladder jays, which some would argue would be the gold standard for bone loss, if the patient has off-track lesions, you have worse outcomes. So the bipolar nature of these injuries have to be considered and have to be accounted for when you're determining your treatment plan. And so our group here, you know, after that meeting, we had a decision cycle. So what is it that drives people to do a bony procedure? And so this is, we use a novel technique called a conjoint analysis, which is very commonly used in marketing, actually, to determine what factors matter the most. So when you're sitting at a table, sitting at a panel, people may say one thing, but what is it that they're really thinking? And so this conjoint analysis is like if you're selling a car, it would adjust the colors of the car, the price, the gas efficiency, and it would do it in an adaptive way where the person taking the survey may not actually know what you're looking for. So it really hones in on what is that you're really considering. So this was 32 cases that we put together. It's adaptive, so every question is varying, trying to hone in on what is it that you're helping you make your decision. And we did surveys, 72 surgeons around the country. And as you can see, glenoid bone loss there, amongst all those other factors, was a clear and predominant decision factor of when you would do a bony procedure, more so than age, activity levels, and you can see it by orders of magnitude. And then further analysis, we can see where is it at that point? What number is it of glenoid bone loss that matters that makes them do a bony procedure? And that number ended up being 21%. So maybe more of the historical data, and maybe people are still waiting for some more data about that 13.5%. But that was what we found here using that really novel technique as a group. So bottom line is bone loss matters. Everyone has their algorithms, and algorithms are changing, and can we do it better though? But how do we quantify bone loss? So there's many studies that show CT scans are the gold standard for it, preferably 3D CT scans, give you a good picture of everything. You can spin it, you can turn it, you can really get a picture of what you're looking at. But there is a cost of radiation. It does require multiple studies. Do I get it at every recurrent instability? Do I have to look at the MRI first and then get a CT scan? So it can take time for patients to go get it, and it also costs money. Now MRI is a gold standard for soft tissue evaluation, obtained in the vast majority of patients. I found at this point there's some Canadians who don't get any MRIs, they just get CT scans on all their instability patients. But I think that most people are getting MRIs on most of their patients. However, the quantification of bone loss can be misleading. I think that CT scans is the gold standard in terms of measuring that. And now the previous studies listing 13.5% were all MRI studies. So is it really 13.5, or maybe it's 15%, we don't know. Even looking at bipolar on-track, off-track, just one or the other, there's a study that showed that the sensitivity of determining on-track, off-track compared to intraoperative and CT scans was only 72.2%. So a lot of people are saying off-track, I'm doing a REM plasage, or I'm doing something. But preoperatively, if you're just using MRI, are you having the right data? Are you doing the right thing? So the question remains, do we have the true representation of amount of critical bone loss, both on the humeral side, as well as the glenoid side? So would it be nice if we had one scan, one image done at one time point on all your instability patients, that you could identify a soft tissue pathology, as well as the osteoesthetics? So there are several early studies looking at cadaver, studies, small subjects, study looking at validating some of this data. And our group wanted to kind of use it in our own hands and see what we could do. So we had 18 patients, 3D MRI and CT scans. And we measured glenoid defect size percentage, humeral defect size percentage, the clonative origin, as well as the glenoid track. And this is a VIBE sequence, and this is the bottom left here. It's actually obtained in the axial plane. That's a unique design to look at bone. And then you can reformat it, similar just as you would for 3D CT scan. And so these are kind of representative models of what that looks like, and how you might make your measurements. This is the humeral side. So some examples here for the humeral defects, both for the CT and the MRI, looking at that. So these are our preliminary data. So it shows that they were similar. There's no statistical difference between this 3D CT and the 3D MRI. I had these asterisks there because we presented this at one of our multidisciplinary research conferences the other day. And actually one of the limitations we thought we had with the 3D MRI was that it came up as pixels when you measure it, and you actually draw across. But we found out from some of our radiologists that actually we can create a conversion rate. And so right now we're working with one of our MR physicists to determine that conversion rate from pixels to millimeters. So shortly we'll be able to have that data. But it is possible, and that is pending. We did show that glenoid tract was not significantly different between 3D CT and 3D MRI. The intra-rater reliability was between 0.83 and 0.97. An ongoing study also. So you're great, all the measurements are similar, but does that really change? How is the surgical treatment similar between 3D MRI and 3D CT? So this is an ongoing study where we're using 10 of those cases. And you might ask why we're not doing 18, was when we trialed it, when we've been through 18 cases, going through all the images and all the physical exams, it was taking people over an hour sometimes to go through it, which is a lot to ask people to go through. So that's why we've pared it down to 10 cases. The first review is they review the cases, standard MRI and a 3D MRI. They can make all the measurements, ask whatever things they want to measure and what they plan on doing. And they give their treatment plan. We had them re-review the case, but this time with the standard MRI and a 3D CT, at least a week around. So far for the three people who've done it, it's been about a week and a half since. These other people, seven other people will be even further out. And we're still waiting for some people to do the first round. But this is the preliminary data we have. But comparing that, we're looking at 86.67% concordance on their treatment plan between a 3D CT and a 3D MRI. Again, with the same group of surgeons. So the overall objective is to improve our ability to quantify the glenoid and humeral bone loss preoperatively using a 3D MRI. So what are the steps to get there? So the first step is determine the validity and reproducibility across institutions using a 3D MRI measurements of glenoid and humeral bone loss. So the development of an MRI phantom of known glenoid and humeral bone loss. And we want to do two, because we want to make sure that whatever the measurements are is not just one, but comparable if you have a separate amount of bone loss on the glenoid and humeral side. So what is an MRI phantom? Well, an MRI phantom is something that you can create and it has known criteria. You can put it, an MRI has a soft tissue component, which mimics the soft tissue, has a bony component, which mimics the bone. And why do we do that? Well, for looking at this review and why we need to use system phantoms is that you want to have the quality control, system consistency, particularly in a multi-institutional study that can be variability between institutions, between manufacturers, between different hardware and software. So it is an important step that you can't skip if you're going to try to use this new technology. So how much does that cost? Well, there are commercial institutions that create these things. And there's a range here because if you look at a non-customized, just a standard phantom, it costs about 1738. But if you want to customize it, so you want to be able to create the amount of bone loss that you know is going to be there on the humeral glenoid side, it's going to obviously cost you more. And that range is up to close to $4,000. These are estimates. I talked to these institutions, these companies, and they wouldn't give me the exact until we gave them everything, but this is what their estimates were. So then after that, what do we do? Well, we got to send it around the country to all these different institutions to test it. So we send them out there for MRI acquisitions. They'll measure the glenoid and humeral bone loss measurements and then we're going to compare it across institutions to make sure that we're looking and speaking the same language. So how much is that all going to cost? Well, research MRIs, at least in our institution, some of the institutions that we spoke with is around $600. So there's been some variability between institutions. There's 15 separate institutions, 17 surgeons, about 15 institutions. Multiply that out, you can see that number there. The phantom development is listed there. And we also, as I mentioned, ideally would have two. So there's two different measurements. And then how much is it going to cost to ship around the country? We're using FedEx grounds, so the most simple costs, so not overnight, two nights, two days. For 100 pounds, it's about $52. You gotta ship it to 16 times because you gotta ship it back. So that's the cost there. That's just the total cost of what this part of the study would cost. And Duke has agreed that if we do obtain external funding, that they would help support some of this, whatever, there is a gap there. So next phase, which is not for this current funding proposal, but where do we want this to go? Where do we want to do with this next? So we want to determine the role of glenohumeral bone loss and outcomes following treatment of shoulder instability. So the development of registry where we have this preoperative image and measurements, both on the glenohumeral side, we have intraoperative findings and treatments, the PROs, recurrence and complication rates. And so I know your question, you're asking is, what makes this registry different? There's a lot of registries out there. Well, we have a 3D MRI where we can measure on all consecutive patients, both primary and recurrent. You're not dependent on whether or not a surgeon decided to get a CT scan. The 3D MRI will be evaluated across institutions. So we know we're looking and measuring and speaking the same language. There's a mix of academic and private practice surgeons. There's a mix of master surgeons, mid-career and early surgeons. So final thoughts here. So bone loss is critical to outcomes of following shoulder instability. I think there's growing, growing literature to demonstrate that. The amount of critical bone loss is changing. We don't know what exact number yet, but there is some data showing that it may be getting less and less. The role of bipolar bone loss and the optimal treatment remains unknown. We don't know what is the size, the depth, the width, the glenoid on and off track. There's a peripheral track concept now out there as well. So there's still a lot to be known of how to manage bipolar bone loss, but we have to make sure we're measuring it the same way. MRI is obtained as a standard of care, but I think it underestimates bone loss. In a 3D MRI, multi-center study, utilizing a 3D MRI on shoulder instability provide a greater understanding of true incidence of glenoid humeral bone loss and its role in treatment outcomes. So I want to thank my co-investigators and I'm open for any questions. Hey, Brian, thank you for your presentation. I have a personal interest in this area because the paper you cited by Volpatt et al, Brian Volpatt was one of our fellows when he worked with our radiology group to work on the 3D MRI technology and when Matt Provencial was at Mass General. So currently we have, we've taken to the next level and they've created a button that we can press and it'll give us a percent bone loss on the 3D MRI. So there's some parallel work here, but wanted to ask you, once you make that determination of the percentage that this can give you, how you propose to determine the on-track, off-track determination? So there's one issue is calculation, but once you have that number, how does it ultimately affect clinical decision-making and what's the middle step? How are you going to assess the dynamic nature of on-track, off-track? Yeah, that's a good point. So for this, I think that's for the second part. For the first part, I was just trying to make sure that we're making the same preoperative measurements. I think we were talking about the registry. Yeah, so that's why it's important to make sure you have the intraoperative data looking at dynamically what you're seeing at the same time. That'd be a factor that you want to make sure you include it into the registry. In terms of having a button where you can push, that'd be great. And I think one of the benefits of having a phantom that's available that we can use is that if other people want to join on to later and be involved, you have this available. So let's say MGH wanted to be involved, so you have this already available. They can say that in order to come here, we just make sure your MRIs look in the same way we are, we can send this to you and you can join on. And we could definitely use, if you have that technology and willing to share it, that'd be great to include and we could all be using it. Happy to share the button. So Brian, this is great. And I think I also have similar interests as well. And I think one question is what imaging platform is this on? So this is on Siemens and I knew that question was gonna come. So this is on a Siemens MRI and there are data out there. So there are comparable sequences on other scanners. Now, there hasn't been any study to compare them all, but that would be something that this phantom study would be able to show whether or not all these different sequences for these different ones would be. And are all of your 16 institutions, are they all Siemens institutions? They're not all Siemens institutions. We don't have the exact number. We actually tried to look that up, but you'd be surprised that a lot of surgeons actually don't know what scanners they have. And we tried looking up online as well. And it's hard to find that data. So that's something that we'd have to tease out before to figure out. But- And then, do you know, so I imagine that you're looking at this on a 3T scanner. Yeah. And are all of these institutions like high field strength or a lot of community scanners are 1.5 or less? Yeah, that's a really good question. I think the majority do have 3T scanners. There's the NOVA is the one private practice group in there and it's pretty big group. I think that he normally uses 3T. And then one more question is your processing, is it all automatic or is it- It's not automatic, but it's the same. If you get a 3D CT scan, it's the same process to do that. So it's, there is a same person who does, who should help me develop this, who does our 3D CT scans. She does the same. So it's a couple, a little bit of segmentation. So it's manual, but it takes her about 10 minutes to do that. The scan time for the vibe sequence is four minutes. So thank you. Excellent presentation. We need to move on to the next one, but- I didn't get any music. No, we're ahead of the-

sports physician

bone loss

shoulder instability

3D MRI technology

treatment planning