BoneChat #68 - Leveraging additive design and manufacturing for patient-specific implants.

Foreign. Okay. So welcome to bone Chat discussion 68 with Matt Chomper. Matt is an expert in design and additive manufacturing. He's a fractional CTO for a couple startups. One is Illuminate, which a lot of people know. And he's also a guest lecture at mit. So I'm handing it over to Matt. Welcome aboard. This is going to be technical and fascinating, I promise you.

And I'll, I'll try to go through the slides quickly and go through some discussion points. I'm an engineer by, by trade, but you know, I've been pretty deeply involved in a lot of the clinical side of the product design and development. And then seven years ago, time flies, but about seven years ago joined a company doing 3D printed titanium manufacturing and a couple years ago split off and decided to do my own thing for, for a little bit. But I've been fortunate enough to connect with Alyssa Huffman over at Illuminate. I think she's a kind of somebody who's in the boneshat group regularly.

Yeah.

And so this is not what we're, you know, I'm working on with her, but this is something that has started to see a little bit more of a push nowadays. I, as far as I'm aware, there's probably a few companies, at least within the spine space that are doing Patient specific, or I guess you call Patient Match. I know that Conformis was a big one that wasn't actually using 3D printing necessarily, but it was essentially Patient Match. Probably one of the companies that's doing it the most successfully at the moment in my opinion would be CarlsMed. And they've got a several breakthrough designations as well as a couple of favorable rulings on, on billing codes for being able to use their technology. And then there's other companies that are kind of trying to bridge the gap between. We call fully patient specific, right. N equals 1 for a single patient and more customizable. So one of the ones that comes to the top of my mind is Aurora Spine with their DEXA cage, which essentially has a few different types of cages depending on patient anatomy. But what I kind of want to talk about is a little bit more of a front end discussion. I mean, I'm involved in a few different endeavors right now, but this is sort of, you know, what I'm going to present and what I'm going to show is really more agnostic. It's how could you leverage the current technology? Not necessarily how is any one individual company leveraging the technology? So first to me, people like Cute kids. I've got five of them. Anastasia is our oldest, she's going to be 11. And then our youngest was born last August. We're a hugely musical family. We're actually excited to have all of our kids. You know, each, each one has chosen an instrument. Guitar, piano, drums. We love coffee. We're a little obsessed over here. My kids do drink coffee. Our morning routine right now is I brew a nice big port of pour over coffee. And at least the three older ones, the, the toddler, the, the almost four year old, he doesn't drink coffee. And then I, I've been really fascinated by biomimicry for a long time and its use in human use products. So whether that's heat transfer or filtration, I've done several projects over the last couple years. Consumer goods, earbuds, wearables, you know, garments, and then obviously orthopedic implants. I would say 3D printing has given us the ability to do some really cool geometries, but I don't think the vast majority of companies in orthopedics are leveraging 3D printing to its fullest extent yet. I think it's a canvas and I think that the, the artists are maybe not necessarily ready to kind of paint in full, full color. Again, my personal opinion, but I think people are getting there. I think that we're starting to leverage the tools a little bit more. You know, a lot of the stuff I work on is this, this idea about computational design. And you know, a lot of engineers in the past are like, you know, I would say parametric design. It's this idea that we build out 80 to 120 different SKUs and you pick the closest size for the patient. The idea behind computational design is basically, you know, putting the hands of certain aspects of the design in, you know, an algorithmic form. So in other words, rather than, you know, an 8 high and a 9 high, algorithmic design is essentially saying, well, based on the patient anatomy, we're going to go with a design that's 8.2 high or 8.25 high. It's allowing for flexibility in the design framework that's based on a set of equations or validated systems that you've put together. And I think this background, this seems like the group that probably is well aware of this more than anybody. So I'm not going to read the slides. I told Tiger I would make them available to anybody who wanted to, to read them after. A lot of people are familiar with Wolf's Law. I, I actually find that Less engineers are familiar with the sort of expanded Wolf law in the 1960s by Harold Frost. And I believe that, that there was a more recent, like maybe 2009, 2011 ish, where, where he actually did a review of a lot of his original assumption in the Frost mechanist theorem back in the 1960s. And he basically reviewed and said, here's what recent literature is saying about his mechanostat theory, which is basically this. It's an expansion on Wolf's Law. It's this idea that you've got a certain amount of microstrain that's necessary to put bone in its key remodeling zone. So it's this idea that we've got a certain amount of movement that we want prior to this overload zone where you actually get a really fibrous formation of bone. But under that, if you don't have enough movement and loading on the segments, then you're not going to really get anything. You're actually going to start losing bone. Bone is really good at resorbing itself when it's not being used. So if it says, hey, I see no stress here, I'm going to utilize the material for somewhere where I'm actually seeing stress. So this is important, you know, and it's been studied in lots of different ways over different products and orthopedic, but physiologic loading of bone is incredibly important for this homeostasis of bone for the remodeling and lower stiffness stems was studied to reduce this proximal bone loss. And obviously again, people are very familiar with this idea that if you have a stiff cage put in, there is a very high incidence of adjacent levels to see issues based on the loading pattern of having too stiff of a cage in. So, you know, how are people reducing or what was the, the idea behind reducing stiffness? Well, we started with, you know, saw the rise of peak and peak was really great because the modulus was close to bone. Titanium was a lot stiffer. And so people started to move towards there. Then we noticed that the, you know, peak last lacked the osteogenic capability of, of the titanium. So, you know, people started spraying it with titanium plasma spray. There was a lot of issues of like, well, the dissociation and the particulate and things like that. So then 3D printing came along. You know, pre2020, you had, you know, fairly basic geometry. Right. A lot of that was done with parametric tools. And then, you know, some of the newer products, some of the ones that I find I think pretty Right. Utilizing additive manufacturing. Well, there's the C spine ortho fix waveform, which I think was a pretty cool technology in the way that they designed it. You've got Si bone and a couple of their systems. I believe that's the bedrock. And you know, a couple other companies I think off the top of my head are, you know, Nano Hive is, is kind of trying to, to do something with really low stiffness. This, this one's wordy. Read this, read this later. But essentially what this is saying is we've got requirements from the fda strength requirements. The problem is that they're not really looking at stiffness or they've never really looked at stiffness. They look at construct stiffness. But the device themselves are actually still really stiff. A lot of the early 3D printed devices have got lattice in the middle and then a titanium frame. And when you look at the whole like people are claiming, well, our lattice has got a really low modulus. But when you look at the device as a whole, they're still really stiff. The problem is the difference between what needs to pass the FDA and what's anatomically relevant. A clinically relevant load is a. There's a huge gap between the two, 20 to 40 times different. So you know, computational design and 3D printing. The idea is that rather than traditional problem, right. Is physics. You increase strength, but you also increase stiffness. Is there a way to increase strength while also maintaining stiffness? And something I often say is that the best implant is one that is ignored or forgotten by the body's systems. It's like it wasn't there. So obviously we talked about beginning. Ideally it's something resorbable or it's the body's own tissue. But in the meantime, until we get there, how do we decrease the footprint of this, of this implant inside the body? So something just kind of interesting and put together old cage versus new. Right. The amount of loading that's clinically relevant is a lot lower than if you look at the devices submitted to the FDA. Now 22,000 Newtons is the 50th percentile force of devices submitted under 510 to the FDA for interbody cages. And 500 Newtons is a clinically relevant load. So that's about a 40 times difference in what devices that are being submitted to the FDA are testing to and what's actually clinically relevant. So the stiffness is just too high. And this. Go ahead and Matt.

That's like £500 and £100, something like that. I've Forgotten my conversion but I think.

I think it's like a 1.6 times or something like that. So. So there's a difference between like consistent loading and then peak loading. 500 Newtons is like a day to day consistent loading of a few hundred pounds.

Got it.

Okay. Now I don't, I don't know. I don't, you know, if you, if you've got, you know, certain people that are day to day in, in and out of the, the biomechanics, you know, I'm friends with several who are really deep in the weeds on that and I would say about 1500 Newtons is like what you typically want a safety factor design a device for because of peak loading. Right. The person's not walking statically, they're going to jump or they're going to bend over still the idea is that what is clinically relevant biomechanically is typically a lot less than.

Yeah.

Devices that are being submitted. So I love show and tell. I. It's never bitten me yet to show stuff live, but I did want to show like what I think a workflow could look like based on my experience. This is something that I made internally actually a couple years ago before I think a lot of these companies really started displaying this idea about patient specific. But people are starting to look at this now, right. You've got, you know, this is just something I downloaded that's, that's kind of agnostic but you've got let's say patient segment that you have uploaded into your software. And I want to show you how, how easy it is with the current design tools to do this. So right now we've got a nameplate here at this, this top level. What's this? L 5, 4, 3, 2. So 2, 3. If I want to change the target, you know, which I have shown here by kind of a cylinder and a sphere. Not only can I change the location and basically instantly get here, but I can also change the trajectory and the orientation. So change the pitch of the trajectory. If I, you know, if the surgeon said hey, rather than this particular trajectory, let's say with, with this consideration I really would like to go lateral, then it's as simple as sort of dragging and dropping and then changing the footprint. Same thing. Let's say hey, I can go lateral so maybe I can increase my depth by 6 millimeters. I change that to 6 and there you go. And if I want to place, you know, anterior, posterior, I just sort of drag and drop it in the same spot and then everything auto updates. So Once you know, if you do it this way, what's cool. And I, I'm not entirely certain I totally agree with full end plate matching because the surgeons are going to go in and they're going to decorticate. However, you know that that's neither here nor there. But what I think is, is even cooler is once you get this shape, then the ability to go from the shape to something that's got a structure in it is again that this is all auto generated. So you've got full conformal matching. All of your structure touches the end plate. And then you've got a spot here for your, you know, your instrumentation, you know, added threads if you want to, or whatever adjacent features. But now it's not super trivial. I've got a lot of experience that leads me to build this kind of workflow. But you can see how a clinician or a user or somebody, you know, design engineer could take, you know, unique patient anatomy and very quickly say here's the design for that patient. And then added loading considerations, right? This person is 300 pounds, this person's 250 pounds. And now you, you can start to, you know, design the lattice in particular for that particular application. So if you want to go with a different structure type, then you drag and drop a different structure in and, and there's a new load type for what you're currently doing. So this is why I think, you know, again, absent of the regulatory hurdles from a design standpoint, there really is sort of unparalleled design freedom that I think that's being offered and is going to be offered to, to engineers and the ability to be able to do something like this. And you know, again, if you, if you're curious, look, look at what CarlsMed is doing with some of their prevost like software. They've got like a whole software side that the surgeon sort of plans out. They plan corrections and then they take it into an internal workflow and do something, you know, I would say fairly similar to this to be able to sort of design that patient specific implant. Now imagine this with patient specific loading considerations so that the stiffness that goes into a small, you know, 150 pound woman is entirely different than a large, you know, 350 pound man. And it's, it's designed for that segment for that patient with their anatomy and you know, weight considerations in mind.

So here Andrew's got a question.

Yeah, I'm wondering, so you said that you have, so is this something, is this platform software algorithm something that you have developed and you're saying that it's easy to develop or have you downloaded it?

No, this is, this is mine. You know. Yeah, I've often been told, you know, personal note, like, hey Matt, you stop posting stuff on LinkedIn. You really should patent it.

Well, so that, that's my question, like as so much today, right? Because sort of the software and open, open software and everything is available and someone, not, not everyone, I couldn't do it. But someone with like, you don't have to be a super expert to be able to create something, right? It doesn't mean that you can get a product out of it because other people can have patents on it already. Like there is a fine line or it's just out there and so you can do it, but so can a hundred other people that have the expertise and the drive and the time to do it. So yeah, I guess my question is how much of it is out there, how hard it is to do it and have you looked into seeing if anyone has protection on, on things?

So yeah, that's a really good question for you know, I've shown similar workflows for, you know, probably as back as far as 2020, maybe 2019. So I would say that some of this stuff is probably public domain. I do think that there have been some recent protections on like not necessarily something similar to this, but really more on like how the correction is performed. I know that on the orthopedic side you've obviously got Carl's Med. On the extremity side, I think there was a company called Redpoint which was acquired by Treece and they had, they were doing patient specific cutting guides actually for bunion surgery. So there's, you know, I, I think right now there, it's just one of those unexplored frontiers and there's a difference between, here's the kind of final design and then there's this idea about like methods like what, you know, this, what I'm showing here is essentially an off the shelf software. But you could program your workflow sort of in the software. So the steps to get there, anybody could do. But the know how and the methodology to get there is probably still a fairly narrow group of people that could build such a thing. But you could essentially package this up and say use it in the spine, using the hip, use it in the knee, use it in the foot and sort of adapt it accordingly. So you know, I think that's a really good question and definitely not my expertise, but certainly something to consider that hey, if you're going to in this realm, I think it's less crowded than it's going to be in a few years when this stuff becomes ubiquitous.

Well, I, I actually argue that it is already from a patent perspective. Right, right, patent perspective. It's already crowded and I think companies that are trying to get any kind of patent protection have a, most likely a hard time because it goes back decades like the, like you said, especially for guides, cutting guides, materializing cutting guides have been around at least in theory, I don't know in practice in the market, but in theory have been around for a very long time. And yeah, I mean I wonder too, like each, it seems like every company is going for its own platform that is going to be on their, you know, with their implants, with their instruments, with the robots.

Right.

There's probably going to be like at some point I would assume there's going to have to be some kind of a coming together to, to create some kind of common platform that, that then will allow companies to have more, more individualized, you know, on top of it kind of ideas and products.

But yeah, I think you know, patient specific geometry right now is, is something that is fairly, I wouldn't say trivial, but getting, getting, you know, this like bone into a 3D design software is a little bit more common. One of the things I've seen recently in some new studies is people now trying to take density of bone, like the Hounsfield's scores and integrating those into software as like a field, like a density field which you then can control the density of your structures in accordance with those. So that may be you know, kind of a newer area. I, you know, this idea about like a platform that multiple people can, can use. I mean maybe, maybe there's a company that wants to, to try to commercialize something like that, but it seems right now that it's very domain specific and if the company does not have the clinical ability to do so, then they probably don't really understand how to tune or create a software that does that. So it does, it does seem like these companies are taking it upon themselves to build their own internal tools rather than a company that's trying to provide a tool set that multiple people can use. Plus software is expensive. You know, people are like why I don't want to add another software package to my 16 that I'm already currently using.

Yeah, not sure if you're familiar with Tech Fit, but they, they probably have one of the best surgeon directed design tools on the markets called Disrupt.

Okay.

And this surgeon can plan like if you're going to rebuild a jaw.

Right.

Or cranial plate for trauma or oncology surgeon can go on tablet or desktop and literally design the case. I call it poor man's cad. They can do it five minutes and then it's, it's delivered in, you know, 10 days, two weeks. It's incredible.

Yeah, yeah. And somebody mentioned Materialize. I mean, Materialize has been doing something similar. You know, I've, I've spoken a couple places with some folks from the Cleveland Clinic and they're doing a lot of stuff with like patient specific as far as in the orthopedics, mostly cut guides at this point. Not a lot of, you know, sort of on site metal additive at clinics, you know, unless you talk about hospital for special surgery. But yeah, that's fascinating too, for an entirely different reason, which would be sort of regulatory. But I think that's the push and the pull. It's like if you can design a software and validate it accordingly, a similar tool, then now you bridge the gap between the engineers who are specialists in the design and the clinicians who are specialists in the application. Right. What needs to go in here? What's the size? And so providing a simple interface that just says, I want to target this segment. Here's the way I'm going to be going in. Here's the patient metrics, weight and otherwise, like, I think I could really revolutionize the paradigm of care.

I think the software interface is going to, I think that's the most important thing and it's going to be winner take off or two or three companies that can figure out the real easy to use software planner for surgeons. Go ahead, Mark.

So, Matt, it seems to me that the patient specific design increases importance to the youth of the patient because it seems like a patient specific design would just last longer, especially if you're taking into account, you know, modulus of elasticity, bone density, patient weight, that sort of thing. So where would you recommend here and today a younger patient going to get, say a hip replacement or a spirit, a spine surgery to take advantage of this kind of technology in today's market? Yes, you mentioned hss.

Yeah. So I don't know that it's a particular. Maybe take a step back, you know, because motion preservation, at least in the spine, is becoming a more common thing now with additional companies entering the market and decades and decades worth of data that they're drawing from. So I think that we shouldn't minimize the fact that if you are a younger patient that does need work on your spine I would say the first thing you should look at is what are the available motion preservation options available to you to fully preserve the motion and your, and your functional range of motion and loading, if you know what I'm talking about here is the razor edge between a fusion device, because you do want, you know, in this case we're talking about fusion devices, things fusing together, but while basically the thing that's left over, the metal that's left over or whatever the material is, that it becomes the most unobtrusive piece as possible. So it's literally only there to incorporate that original fusion and it should be functionally no stiffer than the adjacent stuff. Now obviously in most cases you've got supplemental fixation or you got posterior fixation that's going to all increase this idea of the construct stiffness. So I wouldn't say it's necessarily a company or it's not necessarily like hospital. I think it's more of a push from my like, side to the device companies that, that's just like, hey guys, we're, everything's still too stiff. You know, we felt like we were getting there with Peak and then we doubled back because of Peak's, you know, lack of osseointegration. So like, where's the right point? Is it 3D printed peak? Is it what the curativa guys are doing with 3D printed peak and super low modulus that still allows for integration because. Or is it, you know, continuing to reduce the stiffness with titanium? Or is it patient specific? Dr. Ali Kiapur is doing some, some interesting studies on the importance of end plates touching. So even if you, even if you don't have everything conformal, but you get a really like, consistent rasp, like a box broach, and then you, you touch as much surface area as possible, you can actually have a fairly stiff cage. But if you get really good surface area match, you can actually decrease your risks of not only subsidence, but those at that adjacent level stuff down the line as well. So for me, it's just really more a challenge. Like I, I'm an independent entity right now. I don't, I don't work for a company. You know, I'm not necessarily tied to an oem. So I've just, you know, for the last five, six years just, just tried to speak my mind on where I think the standard of care could be with the tools we now have available to us rather than the status quo, what everybody else has been doing for the last 20 years.

Yeah, that brings me to actually Kind of another observation. I see more and more big institutional hospital networks making their conversations only with what I affectionately call the cartel, the bigger companies. And just from a standpoint of economy of effort and time and money, you know, ignoring smaller startup companies unless they somehow go through one of the big companies to be, to be heard. Which makes someone like you talking probably to more innovative companies in a more challenging position.

Yeah, I mean, there are several companies, I would say, that are on the smaller side that, that are trying to do something like that. So Agada is one of them with like, like, you know, this kind of like planning algorithm. And I mean, I think most of this group knows it. Like you've got large companies that just aren't great at innovating. You know, one of my, when I was back when I was doing, you know, contract manufacturing, you know, customers approach us that are smaller that would want to do cool stuff and I would point to some of the bigger guys and be like, just don't do what they're doing because their stuff is, you know, I mean, if you look at like, like Stryker Spine, they, they acquired K2M and the K2M stuff was one of the, some of the very first 3D printed cages on the market. The problem is they're like, they've not changed in a decade and they're still the same thing. So they don't really want to innovate. There's no reason for them to innovate if they're, you know, their contracts are still being fulfilled. Whereas the small companies see the need. You know, that's why I love Alyssa. Like she, she's trying to commercialize a 3D printed pedicle screw. Not because it's easy to do, but because she thinks there's a very relevant clinical need in a orthopedic product that has an incredibly high revision rate. And this idea of 3D printed device that can sort of deliver targeted therapeutics to a site for people that need it is like, this is why we use the technology. But I think that's a really astute observation, is like most people will look at this and if they're not a mid to small sized company, they would say, cool. I mean, I had a conversation with somebody recently, they're like, okay, so I can do patient specific, but like, you know, does that make me extra money? Exactly. What they said like is, does that increase my bottom line? And nothing about is it better for the patient? It's like, does, does it bring more revenue on the door for me?

Most companies, most of the cartel think that Customs is just a lost leader. You know, it's just a service they provide. Yeah.

And probably one of them. Probably one of them will buy Alyssa's company one day, and then they'll run in place for 10 years.

Yeah, but, you know, somebody at the start was talking about, you know, the printing of tissue as, like, the holy Grail. And, and my. My thought is, I mean, take. Take something like magnesium. Magnesium, if it's. If it's super thin, resorbs fairly well. But the relationship, as you start to add volume is actually kind of logarithmic. You get a big slowdown in the body's ability to resorb magnesium in larger quantities. But so if you. If you took this cage and you tuned it with a bioresorbable polymer as a scaffold, there's some new research that I'm working on where we're basically attempting to model the resorption rate of the polymer and the integration of bone. So basically, over time, you get resorption, but the resorption rate has to match the osseointegration rate so that at the end of the day, you have nothing that was there in bone, which was there. So less so for orthopedic, like, interbody cages and more so for, like, large bone defects, for, like, long bones, where if you've got, like, resection because of a tumor and you need to fill that, putting a really chunk, stiff chunk of metal is not great. And so this idea about a resorbable scaffold would be really cool. But technologically, it is tough. Regulatory, also tough because of some of the challenges with the biocompatibility. But there's definitely people trying it. You know, I can say that.

Yeah, keep going. This is interesting.

Yeah. I mean, obviously, I don't know how much time we've got, but questions are welcome. You know, I don't want to. I feel like I've stayed up my soapbox long enough.

So let me throw out the first question, kind of. So I totally agree with that. The big companies aren't really using additive design and manufacturing the best ways, and sometimes they make mistakes. Like, I remember Stryker was printing cages, and then they started having fatigue problems because they. They forgot to heat. Heat the transfer process to take the stress out. So what. What are the mistakes that people can make just because they can go on a screen and scaffold with sharp edges and any material and maximum porosity? What are, what are the. What are the red flags?

Yeah. So, you know, I actually. I like poking fun at myself when I'm designing this. So we'll do that. So if you took this and you printed it, you know you typically like to stand these up, right? Print, print them this way. Most AM providers would look at some of this internal geometry and say yuck. And that's because an edge like this, that's kind of down facing in the process, that starts from nowhere, it's sharp, it's not going to print too well.

Steps in there.

Well, functionally speaking, that down facing edge, if you don't put a support in there, is going to have, you know, kind of a bad melt pool. It's going to look fuzzy, maybe you'll have a little bit of porosity in there. And so it's become, it's become a fatigue sensitivity point. The FDA is, is painfully aware and I can say this confidently because I'm helping support several submissions all right now of additive devices. They're painfully aware of the fatigue challenges of orthopedic devices. Part of that has been because of some findings of products. So you mentioned the stryker, the striker had caged. The titanium cages originally had some issues with breakages because of impact. So now the FDA is also investigating new impact standards for additive for impact testing because of the propensity of fractures in that case. But from a fatigue standpoint, you know, there's typically a 30 to 60% depending on the process reduction in fatigue performance for 3D printed cages, for raw titanium cages. And part of that is because the material set, you mentioned it, you got to heat treat it and do special stuff. Part of it's because you're introducing really complicated geometry that has stress risers all over the place.

Yeah.

So whereas an old design, I remember designing a cervical plate, you know, more than a decade ago, like trying to keep everything round. Well, you maybe had four or five corners to be like, I really got to put a fillet in here. And then you had to worry about the cutting tool. Well, with AM you may have 3,000 sharp points because of the way that your lattice is built. And so from a material standpoint like that's really challenging. And so people are starting to like this really randomized geometry is kind of cool looking people like random because it's like, well this looks like trabecular bone. But from a physics standpoint, an ordered structure is way more predictable. So I give people all the time, like companies that are like, they show a side by side of their structure and bone, they're like, ours is mimics trabecular. But I'm like, it's made out of titanium, so it's not mimicking it. Your modulus difference is, you know, 100 times different. So why do you want to mimic it? Like, you should rather be using a structure that is purpose tuned for your application. So I find that that's a big problem with people, too. They're choosing structures based on looks because it looks like trabecular bone, not based on the physics or the loading scenarios. That's a big irk for me. Somebody who puts up a random lattice and stochastic and says, this is trabecular bone. It's not.

Interesting. It sounds like, I mean, going back to Andrea's question about ip, you could have secret sauce in your algorithm that developed a lattice that was. That reduced stress risers. You could. I don't know. I don't know what that would look like, but there definitely could be secret sauce. And in this, in the software that builds something that increases strength with the same amount of flexibility, seems like you could. That would be. It seems like code would be the ip.

Yeah, it's possible. I mean, I know that's, I know several companies that, you know, there, there's the design aspect, what is, what does it physically look like? And there's people trying to patent, you know, specific structures that do a certain thing. And then there's this idea of like, well, I'm going to patent a method or a workflow that takes in any number of inputs or lattices. And then, yeah, do what does what you're saying, you know, either targets a stiffness within a certain modulus range, depending on the application, or, you know, any number of things. You know, it's one of the, the fun tools that I worked on. This is, this is, it's kind of free. Fun stuff is, is we worked on a little site called Lattice Robot because we noticed that engineers were like, they're like, I don't know which lattice to use for a particular application, so we'll use a common gyro. So what we did was we performed a lot of functional testing because most people are interested in modulus, right? The stiffness. So if we go to the Z modulus, this idea is we can look at high and low stiffness structures and get basically simulated results in an area that we like, and then get the parameters basically of what gets us in the range that we want to be. So whether that's the thickness of this sheet structure, the bias, which is basically the offset size, we know all these variables and we can just say, here's the type of lattice that the engineer could use in this particular application. Here's some data about it. I, I mean functionally, if you just look at like the number of, of particular structure types, I mean it's a, it's an unexplored area. Absolutely.

Like looking at two variables or three.

Yeah, this is just a, this is just a couple variables for all of these different data points. And we've got a, a few, one here that, that I particularly like. It's a little slower, but it's this spinoidal nose. There's actually been some recent math papers on it. And what's really cool about this is although it is random, you can actually control the anisotropy. So basically you can make it, you can sort of force it to order in a particular direction while still keeping kind of randomness to the structure. So it really is cool for tuning for like compressive moduluses or not. So I, I, you know, one of my, one of the sort of cool current things is, is talking to a couple companies about using sort of different, different structures that are similar to this in their application. So I think metamaterials as a whole is an entirely unexplored area. And even though there is, I mean, Andrew is probably aware there's some of the lattice for orthopedic applications is kind of crowded, right? You've got gyroid structures which are patented across a certain range. And you have, I mentioned like Nano Hive. Nano Hive has some fairly broad patents. There's some broad patents in the SI space for, for lattices. But there's probably still hundreds and hundreds of, of unique structures out there that, you know, you could, you could apply and either skirt current IP or develop your own new class of metamaterials for the application. I could nern out on that all day.

Elizabeth got a hand up and then Mark.

Yeah, I just want to circle back to. Did I understand you to say in the beginning of the presentation that the FDA does not consider stiffness? So they, if you look at the, the submission framework stiffness of your device, they. You actually like, you do your stiffness as a subsidence essentially. So you subside your device into two representative bone pieces, bone foam. And as long as you have good surface area contact, your reported stiffness number can actually be really high. I mean, I remember, you know, we cleared an Aleph cage years ago that had like 40,000 or 40 kilonewtons was its stiffness, which is just absurdly stiff, you know, 80 times the clinically relevant load. And this is a, you Know, this is a product that's still out there in use. It's integrated cage. So I wouldn't say the FDA doesn't consider it. I would just say it's, it's the lowest number on the totem pole. So stiffness is not. If your device is really stiff, it, it's not the reason why it won't be considered substantially equivalent. I got it. Okay. It just seems kind of seemed odd, you know, just on the face of it. Okay, so. So yeah, subsidence is based on surface area contact and subsidence is absolutely looked at. But if the stuff in the middle is a solid chunk of metal and your device is just super stiff, as long as your subsidence number isn't bad, you know, there's, there's not an issue there. Okay, that makes sense. Yeah. Thank you, Mark. I think Mark had a question. What do.

Yeah, what are you doing? What are you doing in your additive manufacturing with materials that are more biologic, like you know, different types of bone or bone mimicking type products other than metals? Or are you.

So, so right now the biggest problem is, is structural. You know, if you intend to produce a fusion device, the device must adhere to the F2267 or 2077 requirements. So I think the problem right now is again, if you intend to produce an orthopedic device that's a fusion product, most non metal parts, while biologically might be okay, are not going to be able to withstand the loading required to pass muster at the fda. So what I'm starting to see is people taking metal or polymers like peak or pec and impregnating or coating them or surface treating them in a way that makes them more osteogenic. So there is a. The nano, the nanov guys, I think that they just had a win with somebody licensing a big company licensing, or maybe it was Medtronic, but, but they do a surface treatment on, on metal that allows the part to be more osteogenic. I think the big limitation is trying to find a polymer that holds up structurally that's similar to Peek, but that also maintains osteogenic properties. So what's the name of the company that just, just popped up on my radar and I want to mention it here because I think it's, it's super cool. The company is Vivolta V I V O L T A and they actually have a technology that uses it's. It's like electrostatic weaving of fibers, but they can essentially use almost any polymer and the polymer can be bonded, basically produced. You Know, as a standalone device, either in strips or it can be bonded to a substrate almost like on a mandrel. And what's really neat is they can work with. With just about any polymer and you know, something like a pla. PLA is interesting because PLA is a. A lot of people that have home 3D printing use PLA and like, you know, I've got several printers over here that use pla. It's super cheap, it's easy to use. But PLA is actually bioresorbable, especially with like really tiny fibers. And so there is already some really cool applications from the Volta for like hurt valves because they can print certain polymers with like, really good lubricity. And you can bond directly to like a nitinol stent versus current products. They have to be like hand sewn. So I know that this is an orthopedic. This is entirely different. But I think if somebody could. Yeah. Figure out how to.

Soft tissue repair in the future will be printing the patient's own cells. It's not going to get there in one step. It's probably going to be five steps. But you're right. I think PLA or some resorbil first.

Yeah. I had the opportunity to go about a year and a half ago to a research institution that's doing 3D scaffold printings, volumetric cellular printing. And although it's tiny at the moment, the technology will scale, I think, to be commercially viable, you know, in the next 10 years or so. And then. And then it's the regulatory side. It's. It's what the. You know, how. Well the company can argue and how much money they're willing to pay to convince the FDA that.

The favorable thing with regulatory is if you're using, if you're taking human tissue out of the person and giving it back to them, it's a much lower hurdle. You know, I mean, just think of stem cells, you know, or.

Right.

You know, you can do that all day long. Interesting. Andrea's got a question then. I know Kyle's gotta have a question. I've never heard him this quiet.

Yes. I don't know if it's because I worked more in spine in the past two decades than in other orthopedic areas. But am I correct in saying that there's been more 3D printing activity within spine and if so, or patient, you know, custom patient, like implants versus like hip and. And shoulder, where you would think that the surface matching would. Would be even more important.

It's a. It's a lower barrier of entry because of the mechanical test requirements, because of the temporary fixation until stabilization. You know, as the vast majority of these implants versus in the hips and knees. I mean, I think the lowest test for fatigue is, is at least two times that of a spine cage. I think it's 10 instead of 5 million. And I believe it's even more so when you start talking about sort of maybe like fretting testing or particulate analysis. So there are people that are doing 3D printed hips, hip stems, and like tibial trays. The other thing is that the technology is, it's more expensive than traditional technology and like plasma spray. So it's less commercially viable in a commoditized market like hip and knee. Now I know some spine products are going a little more commoditized, but right now there's still a little bit more leeway in my experience to charge a premium or you know, work with hospital systems to maybe get a different type of billing because of the value of the product that just hasn't existed in a commoditized sort of. You know, you mentioned that the cartel. Well, the cartel essentially controls big joint stuff. There's not really been a small company that's elbowed their way into big joint space with additive. I know several companies have tried and then their tech got acquired. But I, I've not seen somebody really make waves in that space yet.

I mean, the other thing, Andrea, is the, the margins are higher in spine and the costs are less. And so the business model just makes more sense than hips and knees and shoulders.

Kyle, you mentioned the user interface. And to Tiger's point, I would think that there's a small portion of the small demographic of the surgeons who would be comfortable even designing their own implants or have tags that would design it for them and then use it for each patient individually.

Yeah, I mean, essentially right now, patient specific is a prescription. At least from the regulatory framework. It's a one off device. You know, the FDA has added some restrictions for how you can essentially use the patient specific designation because, you know, it's not cleared under a typical regulatory framework. Now if you look at what Carlsmed's done and other companies, they're essentially qualifying or even conformists. They're qualifying the size and shape that the implant could be. And then they sort of base their worst case on that. They test that worst case and then like, let's say you were designing this. Well, if this implant grew beyond the bounds of what was qualified, then that would be an Issue that wouldn't be covered under the kind of class through or the Breakthrough Class 2 designation or whatever. So I still, I think people are going to push the limits because they can. And a surgeon wants to have the opportunity to say, well this is my patient, so you know, obviously I know what's best for them. Um, it'll be interesting to, you know, as the, as people continue to try to push the limits for this patient specific stuff is if the FDA cracks down more or if they maybe, maybe relax some because of the success of companies like Carlsmed in the space.

Well, it'd be also nice to have the data right when, where enough implants have been done and to have sort of long term data on patients are doing.

Yeah, don't get me started on data because I'm, I've written articles about it. But I think everybody says they want clinical data and nobody cares about clinical data. I've taken me 40 years to come to that conclusion. But by the time you get clinical data, you can't have that implant anymore. It's gone. Now you're on version 17. You know, if you want to figure out what Total D was the best one ever, it's not for sale anymore. So I'm, I'm, I think clinical data is something surgeons say what they want, but they don't. They just need to see another surgeon using it and they're fine.

Yeah, the. I think you're right. I think there, there are some, there's a lot of people that I've seen that are like, hey, this 3D printed titanium, titanium is great. It's osteogenic and they, they sort of hang their hat on that. And the companies I've seen that are successful now, usually it's because they get acquired, not necessarily, you know, by a big company, but they actually bother to perform the data. Even if it's just in vitro data cell study, you know, or an animal study. I actually worked kind of ongoing project. I'm not sure when it'll be published with, with Bill Walsh on the struggles with like I'm been in 3D printing titanium for a while and so obviously I believe in the technology, but there is a problem with you know, understanding, looking at like X rays or imaging of like how well the bone actually grows in there. What type of tissue is it because of the scattering of the metal versus something like peak which was a lot easier to see what was in the body. So we decided to do a little study because a lot of people would say we got titanium. It's Great. And Bill would do a study and be like, I don't know if it's as great as you think that it is. But then the company be like, shut up. Looks good to us. And so he's reached out to me and was like, hey, I need some agnostic additives cages so we can actually put stuff in here and then image it and then see, like, what this stuff actually looks like. So if you look here, it's like, okay, I might get a really good visualization. But like, no one takes this view. Like, this isn't the important view they want to see like, like AP or lateral. It's like, well, if you go this way, it's like, well, they're not going to necessarily see a full column in there. So then the question becomes like, how do you tune the structure also to give good visualization so you can actually see clinically what's supposed to happen? And does that even matter to people? Is that. Is that, you know, is that something that they really want to see that. That clinical data or not? So I think you made a really good point there.

Interesting. Joe, you have any questions about. Joe's. Joe's got a. They have a total knee. Total hip company, but you ever thought about printing tibial plates? Titanium?

I think it'd be excellent. What we're targeting first is, and you kind of touched on it, Matt, is, you know, the, the porous coating is more commonly accepted and it fits more with the large joint models. So doing forest coding on. On a knee, on the tibial side or on the. On the femur side is a maybe less barrier to entry. Next step.

Right.

For the benefits of that. But believe me, I do. I. I am a big advocate for additive and what additive can provide. But you guys touched on a lot of the. A lot of the, I guess, struggles with it in the industry to get adopted. And I think Tiger even wrote an article about, or maybe wrote it a while ago and was reposted about how big companies don't like innovation. And I think that that's. That does speak true. It's something that I hadn't really noticed until reading that article and bringing it to my attention.

Yeah, they don't. They don't like risk. And anytime you're playing around with new technology and trying to solve clinical problems with something, it's too risky. They will wait till somebody's selling $50 million worth and then they'll buy the company.

Yeah, and I see you're using N topology. That's something that we've. We used for some of our. We're 3D printing instrumentation and wanted to lightweight it as best as possible. And we came across this software. We're really impressed with it. It is talking about how. Or Tiger talking about how you can or there's merit to selling code or an algorithm and topology makes it very doable to sell that algorithm of. I put together this formula for which lattice structure meets all the requirements and I can. Here's this formula. Apply it to your geometry, apply it to your use case. And the formula itself has been tested through multiple different lenses.

Right.

Interesting. Have you ever, Matt, ever thought about selling your software or licensing it like, like a Unigraphics or SolidWorks, you know.

So I'm actually friends with the founder of ntop. I, I told him I was going to wear his shirt today, but I don't work for the company. They've, they've never paid me. I'm, you know, I'm happy to say that because I actually use lots of different softwares in my day to day, but I just, I really like this software because it provides a pretty blank canvas for you to build whatever you want to in it. And I've actually been using this. I was one of the very first alpha users of the software. They actually handed the original alpha version to me on a flash drive back in I believe 2018. So you know, you could talk directly to the founder of this company and he'll tell you I'm, you know, I'm, I'm the medical power user guy. He, he tosses a lot of, a lot of folks over to me. So we talked before, right? Yeah, so we, we've talked before about like packaging a block, like what you said, where you've got inputs and outputs, packaging a block and then like selling it to somebody who would use. There's definitely, there's, it's definitely viable. I've found in my practice as a consultant, most of that like in consulting work is owned by my client anyways. And so it would just be extra to be like I've got this special secret thing and then I have to go like figure out who's gonna, who's gonna buy it, OEM or otherwise, which would require me to do my own testing and vetting and sort of paying out of pocket. So is it viable? Absolutely. I think a cus. I think a company could specialize in creating packages that would go into software, but I think it's still fairly narrow because a lot of companies, they, they, they don't care about which lattice and I Don't mean that like disparagingly. I mean just people are like just put something in there. It just bone needs to grow in it. I don't really care what it is. So people are okay with the off the shelf stuff and you need to prove to them why your bespoke custom thing that you did, that's super cool. They're like why did I use that? What's so cool about it? You know, I think that that's so, so it's not been worth my effort. I just like posting and showing this stuff to show people what's possible and they make their own decision on whether they use it or not.

Well, maybe one day when they realize stiffness is, is probably the number one most important variable, they may go there. Interesting.

I mean you taught me that today that whenever. I don't know if there's a difference there on the large joint side, but all you ever do is you just try to meet the, the different specifications for stiffness and, or for, you know, for the various specs. And those are likely there several factors of safety above what's necessary for an. On the actual clinical side. And this is the first time that I've seen the idea of hey, we're, we're putting, we're putting product into the body or you know, materials that are into the body that sure it's strong and it's not going to fail, but how is that actually impacting the day to day life of the patient?

So reduce.

A very interesting concept to me.

I called this the pile of spaghetti because it just looks like a bunch of spaghetti, but it's actually connected from the top to the bottom. This was an example I did recently to show that. A fairly common lattice that you'd see and something that's more bespoke. These are both the same volume fraction, they contain similar amounts of porosity. But the interconnectedness from top to bottom means when you load this, you displace it to a certain amount and you want it to move, the structure is going to fail. This has virtually no stress when you displace it a millimeter. And this is, this is in titanium. So this idea is that, well, what's so special about a structure? It's like, well, if you create a purpose built for the application, you could really create, you know, which I've actually got some, some intellectual property that I'm working on. You could create something akin to a motion preservation device. And the motion all comes from the tunability of the structure itself where you've essentially Got no moving pieces. Nothing is rubbing on one more like.

A spring than a, than a block. That's interesting.

So it's. I just wanted to. Third, I posted on LinkedIn a while back, but I. Not sure if everybody follows what I do. I'm not, I don't have that big of a head. So I just think that that's. When you talk about this, the structures, it's like, should people be interested in cool structures? In my opinion, absolutely. You know, zoom into any natural structure on the earth and you'll find something fascinating that's purpose built for its application. And so why, why wouldn't engineers, you know, use that to mimic their, for their applications?

Yeah, nature, nature has been tested over centuries.

Right. I always tell people, you know, I don't, I don't really get too spiritual about it. So regardless of whether you think stuff was created or whether it evolved over billions of years, either way, those structures are still incredibly cool because they're purpose tuned for their application. So regardless of what you believe, I think we could all come to the conclusion that if you look at the cross section of a tree or even bone, it's, it's incredibly purpose to not just structural, but for, you know, mass transport, fluid transport, for a heat transfer. I mean, it's just, it's like, I don't know how people can look at that and not get super excited.

It's perfect. Well, that's, we kind of went a little long. That was fascinating. It's a good one to end on before we get into religion, I think. Well, thanks Matt. This is, we're probably going to have to do part B in the future. This is fascinating.