Rethinking Aneurysm Treatment: Why Flow Reduction Isn’t Enough and Pressure Still Matters

In this edition of Cardiovascular Pioneers, we drew inspiration from a recent publication by Prof. Tim Becker’s laboratory at Northern Arizona University on fundamental neurovascular research and had the opportunity to follow up with an interview to explore this fascinating topic in greater depth.

Karim Mouneimne Tim, thank you for joining us. I read your article on 3D printed rupture prone giant intracranial aneurysm. Super interesting. What was the most interesting finding from all this research you did?

Tim Becker Thanks for the opportunity to talk to you. This is a really impactful research because it’s very clinically relevant. The neurosurgeons that we work with in this field really came to us and tried to understand what is going on with aneurysms before and after they’re treated. And really the after was what they were looking for.

They noticed that putting a coil or a stent to try to treat an aneurysm does stop the blood flow. But they keep seeing these aneurysms regrow in a lot of patients, especially the big ones, these VARs, these larger aneurysms. And the most impactful thing we found was when you place approved devices, stents, coils, even though you might be packing this aneurysm completely, you’re not changing the pressure inside of the aneurysm. The pressure of the blood flow, 120 over 80, that’s not being changed when you place a device. So the pressure is always pulsing on the aneurysm, which can cause it to regrow. They never really had real quantifiable data on this, because in patients it’s hard to get that. But in the in-vitro models that we’ve set up that act like a patient, have the same blood flow, same blood properties, same pressures, we can measure that directly. And from an engineering perspective, we were able to give the clinicians of some quantifiable numbers that they had never seen before. So that was really exciting.

Karim Mouneimne That’s very interesting. If you don’t change the pressure in the aneurysm with different treatment modalities, and I believe that you had two or three treatment modalities in your paper, then what impact do you have clinically that you were able to see with this model?

Tim Becker So on the impact side, stopping blood flow is obviously the most important component. And then the part that they didn’t quite understand is, blood clots form inside this aneurysm, which are not mechanically stable. So, it really helps kind of figure out what’s the best way to approach this as with newer devices or better devices. Because coils do stop flow, but they cause blood clots to form. Stents really minimize flow, but they definitely don’t completely stop it. What we ended with the paper was, what would be the new type of device that might do a better job of treating these aneurysms?

It looked like initial data showed like a gel material that you could inject into the aneurysm and seal off the entrance with a gel material that will stop the pressure pulsatile component. And if you reduce that pressure, as well as obviously reducing flow, then you can stabilize the aneurysm and theoretically, you have much better long-term results and even coiling and stenting that is used right now. So there does need to be a paradigm shift, a new type of treatment, small incremental changes can help. But again, if you’re not changing that pressure, which they are not doing, that pressure is not going down, you’re not ultimately going to help the patient.

So I think this will lead to some really cool advances in new device designs. And as engineers, that’s what we’re most interested in anyway. So we’re excited about that.

Karim Mouneimne What about on the physician side? Is the tool you have, especially with the 3D printed aneurysm, something you can use to print custom patient specific models and for physicians to do pre-surgical planning and learn different treatment modalities and determine which one would be efficient? Is this an angle that would be interesting for physicians?

Tim Becker Yeah, for sure. That’s gotten quite popular is AI related settings as well as software analysis takes images of, say, CT scans and MRIs. You can isolate those vessels of an actual patient. You can 3D print those actual models and you can do that whole turnaround within about six to seven hours after you scan the patient. And that’s something we do in the lab as well. We’ll take scans, we’ll convert their vessels to SolidWorks files and actually 3D print those in these soft materials. So that is definitely an area where the simulated surgical suite that we have here can really help utilize that. And the clinicians can use kind of software simulations to pick the right device, but then they can also hook it up to a model such as the ViVitro system, simulate the actual blood pressures, and do a surgical procedure in a simulated environment just like I have behind me the day before, right, before they actually treat the patient. Make sure everything works, that they get the result that they want initially. And that’s going to be very impactful. And that’s becoming, it’s not a standard yet, but I can definitely see that coming.

Karim Mouneimne Fascinating. Without any transition, what are you working on now? What is the most important priority in your research lab?

Tim Becker We’ve been expanding the surgical suite because we want to mimic many types of stroke related research in my lab here at Northern Arizona University. And we’re looking at stroke in two phases. It’s hemorrhagic stroke with aneurysm treatments and ischemic stroke with blood clots that can go to the brain.

We’ve expanded our models from just aneurysms that can predictably rupture. We’ve also created full vessel models of the circle of Willis, which is the vessels that go into the brain and rotate and send blood flow to the front, back, and sides of the brain. That’s where blood clots usually get caught. We can simulate those exactly using you guys as pump to create the pressures within the flow. And we’ll put in medical devices, measure pressure changes while we do that, because it does disrupt the flow in the vessels of the brain while you’re treating them. So, we could measure all that in real time. And we could also calculate forces, shear stresses, things like that. And it gives a really good feel for how these devices that remove blood clots can do that. So we simulate the circle of Willis through 3D printed resin models, and we simulate patient specific blood clots. They actually have the exact same mechanical properties. And then we put in these industry devices; aspiration catheters in particular, and they will basically suck the clot out of the patient. So we can do that exact procedure here. Industry loves it because they’re like, hey, we came out with a new aspiration catheter. We’d like you to compare it to the competitive device. Before that, the surgeon was like, I like this one because it kind of feels good when I use it, but they could never really quantify why it was better. But now we have all these pressure readings, these flow readings, these times that it takes to create aspirations. We’re getting a lot of really cool data that says, your device is a 20% more effective than the competitor, it has, 15% more force. So those are numbers they’ve never seen before. And we can get all that stuff for them. So that’s been really exciting to standardize everything, turn it into an engineering problem, and get some actual data that the clinicians can use in their daily routine. So we’re excited about that.

Karim Mouneimne This is great. You actually shared almost my third question. You give us a bit of a sneak preview on your findings, and you did actually very well. It looks like the ViVitro system is a component of your mock flow loop. Is it a critical component of your system?

Tim Becker Yes, most definitely. I’d say it’s the key component because we are trying to simulate the actual pulsatile flow in the brain. So we have the compliance chamber that you guys provide as well. We were able to modify that with our flow system. We have a reservoir that has simulated blood in it. So it has the actual viscosity and shear thinning effect, so that when you have the flow go through the smaller vessels of the brain, the blood flow responds appropriately. And as we’re measuring pressures, you know, proximal and distal to our model, we can, we use LabVIEW data acquisition, you know, kind of based on a lot of the data acquisition you guys provide as well.
We can measure pressure drops across any of our branches, and we know the resistances of our model, so we can actually determine flow rates in real time. So we have real-time pressure that the pump gives us, and we have real-time flow rates coming from that data. So yeah, the fact that we can do that pretty complicated, 70 beats per minute. We can do an older patient, younger patients, you know, female, male, that’s all very different. We can program your pump to have those flow cycles. And that’s going to be really relevant because we hope to make multiple models that represent 99% of the population. And we just plug them into the ViVitro pump, turn it on, set up all the pressures, and we can run full simulations of surgeries. It’s been a very critical part to it. For sure.

Karim Mouneimne Indeed, bringing, importing and generating the right pressure and flow waveform for patient-specific cases, is very important in having a successful research.

There are different levels of complexity of Mock Flow Loop (MCL) systems. I saw that in your publication you’re using a divergent branch. I assume that branch basically takes most of the flow out, like the ascending and descending aorta, to branch off into the upper cerebral circulation. Is that the case? I read that, but I just want to make sure that I read it correctly.

Tim Becker Yeah, we can run your pump system both ways. Most of the time, you know, if you wanted to fully simulate, you would have, say, 5 liters a minute of blood flow going through, right? And then you would only divert 1/2 a liter to a liter to the brain and the rest would go back to the body, right? In the neurovascular space, I’m not terribly interested in that branch, right? So a lot of times we will actually slow down the pump to really more mimic the flow in the brain. And then we’ve seen that we were able to create the waveform. And the waveform is a very specific shape. Once it’s gone through the vessels and heads to the brain, versus when it comes out of the heart.

So the compliance chamber helps with that, but we’re able to tune your pump to give us that waveform. So most of the time we can do it that way, Karim, but usually we do slow down the pump’s flow rates to around a liter or half a liter a minute, and most of it is going to our brain model. And a lot of it too has to do with resistance and flow effects to be able to pump up the model to 120 over 80 millimeters of mercury, that’s very difficult to do with any pump system at 5 liters a minute. So it’s much easier to do when you’re doing slower flow rates because of all the resistance tubing and everything we have.

It’s a very effective way to do it, and you don’t have to be doing the entire flow of the body to mimic the brain. So in our case, that’s great because doing it at the cardiovascular level would be a little more difficult to get the same pressures that we’re trying to do in the brain and maintain that high flow.  The brain flow is about a fifth to a 10th of that. So it’s much easier to manipulate with your system, which is great.

Karim Mouneimne This is great. You’re already giving some pointers and advice to our readers. Generally speaking, related to the research or designing a mock flow loop system or 3D printed model, what other advice would you give to researchers and people in the industry? Things to watch out for. Things like: we’ve discovered that and it was painful. What advice would you provide?

Tim Becker Simulating blood flow to create models of patients is really very tricky. I think a lot of the system, the pump systems are great to get the flows that we want, but maintaining the waveform, maintaining the pressures, maintaining, you know, getting all the air out of the model, that’s probably the most difficult thing. Calibrating all of our pressure transducers. That’s kind of where we’re at right now. We are working on creating a system that’s more portable, where we could potentially take this with your pump or with maybe a smaller version of something to hospitals and set this up, measure pressures, measure flow rates, you know, and this system allows us to figure all those nuances out. But a lot of it is calibrating, you know, getting everything calibrated at the beginning, getting all the air out of the system, you know, and we’re trying to create a somewhat closed loop system. We’re using blood analog materials that hopefully when they sit around for a while, don’t start growing stuff in it, right?

Tim Becker That’s a problem with these flow models when you have fluids static for a while. So those are the things we’ve learned is finding the right materials, finding the right components that will last. Because the pump runs forever. We love the pump. I think we barely ever have to do maintenance on the pump, but everything else.

Seems to like to fail, right? Valves fail, branches fail, 3D printed models start to leak, you know, because they are really interesting and cool, but they’re made all these soft materials that can fail. So we’ve learned the right types of blood analogs that are compatible with 3D printed models, so they don’t swell up and burst over time, and they can handle the pressures that we’re doing. We like that, though, because they’re very sensitive, kind of like it would be in a patient, right? If we go in and put a catheter up into the brain, into the blood vessel of the brain, and we push too hard or turn around a corner too fast, you could perforate a vessel, which happens, unfortunately, in the actual patients.  We get a good feel for that for our models because they’re fairly delicate too. We don’t really want something too robust. So I think those are the kind of things you got to learn when you’re doing these. What is your model going to do? Is it kind of a one-time use? Is it very critical to match kind of the sensitivity of a patient? Or do you want it to be really robust so you can use it over and over and over again? That’s a real interesting balance that we’ve played with quite a bit on our models. And depending on what we’re doing, that varies quite a bit. And it does change the kind of data that you can get. And you want to be as close to the clinical relevance that you can. So that’s something we’ve always are working on is how do we balance kind of our models, the flow system to match actual patient information that we get from our clinicians.

Karim Mouneimne That is fascinating. I completely relate to you. When I was doing all this work in the lab, I had major explosion with glycerol splashing everywhere in the lab. I’m glad it wasn’t a patient. It was only tubing. But yes, I completely relate to your advice.

So thank you very much, Tim. I really appreciate it. We’re out of time. Thank you for sharing all the experience, your research, and all the good work that you’re doing.

Tim Becker I appreciate it. You know, neurovascular work is important. I know most of your work is in cardiovascular, but neurovascular is a big area too. And your guys’ systems work great in both areas. We really appreciate your help and expertise as well. Thank you.

 

Read more Cardiovascular Pioneer interviews here.