.png){kind=small}
.png){kind=small}
BackTable / VI / Article
LAVA Liquid Embolization Technique: Pearls to Practice & Pitfalls to Avoid
Audrey Qian • Updated Jul 22, 2025 • 82 hits
Liquid embolization is a minimally invasive technique used to occlude blood vessels by delivering a liquid embolic agent through a microcatheter to a specific vascular target, such as an aneurysm, arteriovenous malformation, or tumor. Upon injection, the liquid agent solidifies in situ, effectively stopping blood flow without relying on the patient’s coagulation system. This property makes liquid embolics particularly valuable in settings such as trauma or coagulopathy, where achieving rapid hemostasis is critical.
As liquid embolics gain more attention, their effective use requires a high degree of technical precision and consideration. Dr. Gary Siskin, Chair of the Department of Radiology at Albany Medical Center, describes technical strategies and common pitfalls to avoid for more effective and controlled use of liquid embolics.
This article features transcripts for the BackTable Podcast. We’ve provided the highlight reel here, and you can listen to the full podcast below.
The BackTable VI Brief
• Precise control during liquid embolization is essential for safe and effective delivery. Slow injection, appropriate viscosity selection (LAVA-18 vs. LAVA-34), and microcatheter positioning near the target vessel are factors that must be considered to enhance control and minimize reflux.
• Cohesive, non-adhesive liquid embolics like LAVA reduce the risk of the catheter getting stuck in the patient, a common complication with glue-based agents. Additionally, flushing microcatheters for reuse is discouraged due to the risk of retained embolic as a dead volume, leading to unintentional off-target embolization.
• Vessel size and flow dynamics dictate embolic delivery. Liquid embolics should not be used as a primary agent in vessels > 5-6 mm or in high-flow lesions like AV fistulas without distal access and mechanical backstops, such as coils or plugs, to control the flow spread.
• Avoiding common procedural pitfalls requires anticipation of volume needs and reflux behavior. Estimating cavity size with geometry-based approximations, targeting distal branches, and injecting as slowly as possible are key strategies for maintaining control and ensuring complete occlusion.
Table of Contents
(1) Liquid Embolization Technique: Pace, Viscosity & Control
(2) Catheter Handling in Liquid Embolization
(3) Liquid Embolic Use in Large Vessels & High-Flow States
(4) Common Errors to Avoid in Liquid Embolization
Liquid Embolization Technique: Pace, Viscosity & Control
In trauma cases requiring rapid hemostasis, liquid embolics offer a more time-efficient and controllable alternative to coil embolization. Delivering a slow, deliberate injection is critical for controlling the forward progression and minimizing reflux of the liquid embolic agent, while maintaining visual tracking under fluoroscopy. The choice between higher and lower viscosity formulations, such as LAVA-34 or LAVA-18 respectively, should be based on catheter position. Higher viscosity is typically utilized when the catheter tip is positioned near the target, while lower viscosity is often used when the catheter cannot reach the target, allowing the embolic agent to fill the gap via forward flow. However, the best practice is to advance the catheter as close as possible to the target, as predicting the extent of low viscosity flow can be somewhat unreliable.
Often, only minimal volumes – as little as 0.1-0.2 mL – are needed in small vessels, with administration via 1 mL syringes for precise volume control. Dr. Siskin explains that clinicians must also account for about 0.5 mL of dead space in microcatheters as a result of DMSO, a solvent put into the catheter that acts almost like a primer. The DMSO solvent should be injected slowly to avoid vasospasm or patient discomfort before beginning fluoroscopy when the embolic agent nears the catheter tip.
Assessing the endpoint for liquid embolics is most accurate with pre-embolization angiography and seeing whether the final case replicates the vascular architecture. If the result is sup-optimal, re-navigation may be needed for additional embolization. In cases when vessel access is particularly challenging, coil embolization may be preferred to avoid re-navigation.
[Dr. Christopher Beck]
I put out one of the ways in which I thought, like having a liquid embolic as a tool in the tool chest would be helpful. Just curious of how you guys are practicing at Albany. What cases are you using your liquid embolic for?
[Dr. Gary Siskin]
I think first and foremost, we're looking at peripheral hemorrhage. We see a lot of trauma here. It's a busy level one trauma center. I think when we see active bleeding on a trauma angio, so let's say pelvic angio, renal angio, anything like that, we're very quick to grab liquids. I think that the DMSO-based liquids like LAVA, for example, does a really nice job at being able to administer it quickly and to maximally fill the space and stop bleeding quickly.
Now, I think all of us in a quiet moment would probably argue that we're really fast at what we do, but no matter how fast you are, it takes time to put in multiple coils. As people who put coils into vessels, it's pretty rare to have one coil occlude a vessel. It's normally a couple of coils, and that just takes time. Grabbing a liquid and putting it through a catheter that's appropriately positioned, even if you're injecting it very slow. For LAVA, we do advocate that it be used very slowly. The reality is that you're still going to be faster than putting in multiple coils. Therefore, in a patient who needs us to work quickly, this is probably one of the fastest ways we can accomplish that goal.
[Dr. Christopher Beck]
You also mentioned injecting slowly, like how slowly, is it just a feel? What is slow?
[Dr. Gary Siskin]
The way I was taught was inject it as slow as you possibly can and then go slower. The reason for that is, it's not that crazy. There's actually a reason. The reason is that you want the control. If you administer it that slowly, then what you're doing is you're having it almost imagine dripping out of the catheter, but staying together. The agent is coming out as one growing mass of embolic. It has a leading, almost like a leading membrane or a leading shell that comes out.
As you continue putting that slow forward pressure on the embolic, you start to crack that shell and that allows it to sludge forward. You have this very controlled injection of the material. You can see it well, you can see exactly where it's going, but you can also appreciate when it starts to reflux. If you inject it too quickly, it'll run away from you a little bit. You'll either push it too far forward or you'll cause it to reflux too quickly. That's when you run the risk of non-target embolization. The slow injection is really to take advantage of its controllability, is what I would say.
[Dr. Christopher Beck]
Let's say you're in a tiny branch, like it's a renal trauma, and you got a little pseudoaneurysm and you're deep into the vessel, maybe a two-miller vessel. How slow? How long would it take you after you have the solvent in to embolize that vessel?
[Dr. Gary Siskin]
If my catheter is well into that target vessel, then I would use a higher viscosity formulation. When it comes to LAVA, there's two viscosity formulations. One is called LAVA-18, one is called LAVA-34. Those are actually the units of viscosity, they're millipascal seconds. If I get my catheter right where I need it to be, I would slowly administer the LAVA-34. I would have it reflux back or pull my catheter back until the space was completely filled and I avoided refluxing into a vessel that I wanted to preserve.
If my catheter could not get that far, let's just say I couldn't make that last turn or that last division, and I'm in more of the parent vessel for that branch, then I would use the lower viscosity formulation because I would want that to do the work for me. That's a phrase I like to use. I don't have to go all the way out there because the lower viscosity formulation will do that work for me. It will travel, it will bridge that gap between the target and the tip of my catheter. That's how I make that decision, but that's how I would handle it.
One of the things that a new user would probably be surprised at, and I remember being surprised at this, is how little you actually need. When you're talking about a two-millimeter vessel, maybe it's a centimeter, maybe two long, that's it. You're going to use 0.1 ml, maybe 0.2 tops. It is a remarkably small amount of material that you need to treat a vessel like that.
[Dr. Christopher Beck]
Getting into that. What do you inject with? 1cc syringe, 3cc syringe? Because if you just need 0.1 ml, like that's pretty controlled.
[Dr. Gary Siskin]
Yes. Very. These materials come packaged with 1 ml syringes. Typically what we're talking about for dead space volume or the internal volume of the catheter is about 0.5 ml. You're first filling that microcatheter with 0.5 ml of DMSO, and now you're then going to be chasing it with the LAVA, let's say. What you're doing is initially first pushing out the DMSO, and we do that just as slowly as the embolic. That could be a source of error for some people because you might be thinking, "Well, it's not the embolic yet, it's just the solvent. Let me just push that in." That can because spasm and it can hurt patients. Patients feel that.
We want to go very slowly. We want to avoid that. Some people will use nitro first to help prevent that. Whatever you're doing, you're first pushing in the DMSO, and now you're going very slowly. You know that the first things coming out of the catheter are the DMSO. You don't have to fluoro right away until you get close to completely injecting that dead space volume. Then you start looking, and if you go slow enough, you'll see it begin to accumulate near the tip of the catheter, and you'll see it push forward as your injection force increases. It does not take a lot. You're right, it's a 1 ml syringe. It's very controlled. For most of the bleed cases that the average person is going to do, it's less than 2 ml of product that you need to administer. In fact, I would say less than 1 ml.
[Dr. Christopher Beck]
How far can you get the embolic to run, I guess, like if you're using the lower viscosity formulation?
[Dr. Gary Siskin]
I think the way to account for that is to get your catheter as close as you can to the target. That's going to take care of all of that, because now you'll be able to put the material in. As it starts to reflux, you can begin to pull your catheter back while you're injecting to fill all that space.
[Dr. Christopher Beck]
[W]hat's the end point? Sometimes when I'm doing coils, I'll squirt and I'll say, okay, well, it's not quite occluded. Maybe I'll launch a couple more coils and then I'll retake a picture. What's the endpoint with liquid?
[Dr. Gary Siskin]
I think what you need to do is to have a really good pre-embolization angiogram. You have a reference picture from a pre-embolization angiogram. You know exactly what the target vessel looks like, maybe the previous order branch that it came from. You know what that part of the tree looks like, essentially. When your cast looks exactly like the angiogram did, that's your endpoint. It would be extremely unusual for you to fill a vessel, have reflux come back, maybe into another vessel that you were totally willing to sacrifice and include in your field, and then do an angiogram and see that there's still flow through that target vessel. That is extremely unusual.
The way I phrase it is this, you want to inject through your base catheter or your parent catheter. We all like to get final pictures of what we did. It's good to show the kids and make a little photo album. You take out the microcatheter, you do that final angiogram, you're happy with how it looks, and you're done. Or you're not happy with how it looks, and now you have to go back in with another microcatheter and do a little more work there.
I will tell you, first of all, that's unusual. Second of all, it just makes sense to do it that way to avoid those potential risks of non-target embolization. If it was one of those cases where it took you forever to get the catheter where it needed to go, we've all been there. It took an hour and a half to get the catheter into that small branch. I think I would argue that's not a liquid case because you don't want to take that chance that you need to do more and have another hour and a half's worth of work. If it took forever, I think I would be comfortable just saying, let's just use some small coils and we'll see what it looks like. At that point, you can always fill in the gaps with some liquids if you want, but probably the coils will be enough, especially if it's a really small distal vessel.
Listen to the Full Podcast
Stay Up To Date
Follow:
Subscribe:
Sign Up:
Catheter Handling in Liquid Embolization
A common complication seen as a result of using adhesive agents like glue during liquid embolization is the catheter getting stuck. However, this challenge can be substantially mitigated with the utilization of cohesive, DMSO-based embolics such as LAVA. Unlike glue, LAVA does not bind to the catheter, reducing the risk of it becoming stuck in the patient. Mild resistance during withdrawal may also occur due to vessel spasm rather than product adherence; however, this can be resolved with forceful, controlled retraction.
While flushing with 0.2-0.3 mL DMSO may be used at the end of a procedure to finish delivery from the existing cast, full catheter clearing and reuse are discouraged, as residual volume in the dead space can result in unintentional non-target delivery. Consistent practice in injecting, embolizing, and removing the catheter promptly can help avoid procedural risks and preserve embolic cast integrity.
[Dr. Christopher Beck]
While I'm thinking about it, talking about reflux, I guess one of the dreaded complications that I have when it comes to reflux of a liquid embolic is the catheter getting stuck, like stuck inside of the patient. Is that a concern? If it's a concern, how do you avoid it?
[Dr. Gary Siskin]
Sure. That's a glue thing. That is one of the reasons why we converted from glue to these DMSO-based agents, because I just personally didn't want to have to think about a dilution factors that you have to do in order to tailor how far the glue is going to go in a particular vascular bed, but also having to navigate that risk of the catheter getting stuck, the adhesive quality of glue. That is not an issue when it comes to LAVA. LAVA is cohesive. It's not adhesive.
In the recently published trial, the multicenter trial that led to the approval of LAVA, there were no cases at all of a catheter getting stuck in the target vascular bed. We have not had a catheter getting stuck either. Sometimes what you might feel is some of the vessels spasm. You might feel that it can be a little difficult to initially pull. The best analogy I heard was from Dr. Fishman from Mount Sinai, who described it as quicksand sometimes, that you have to just pull back a little tighter, but you can get it out. It just doesn't get stuck. I want to make sure he gets credit for that because I like it.
Ultimately, we have not had any catheter get stuck to the point where we are unable to pull it out. I guess what I would say is, what I tell people as I'm training, let's say, or just talking about it is just do these procedures like a normal person. Just in other words, getting where you're going to go, give the embolic that you need to use, embolize the vessel, and then get out. If you're going to leave the catheter in and have lunch for a half hour, then you might run into problems. Just do what you need to do and get the catheter out, and you're going to be fine.
[Dr. Christopher Beck]
Okay, fair statement. After you're embolized with LAVA, is that catheter still-- say you have a pelvic or injury to the liver, and you've got two spots that you'd like to get super selective on and embolize. Can you embolize one location with liquid embolic and then redirect the microcatheter? Is the microcatheter still in play for a second embolization?
[Dr. Gary Siskin]
In my opinion, it's not in play. I think the question that gets asked commonly is, can you clear the LAVA from the microcatheter using DMSO? In other words, can you essentially flush the catheter? That's a problem, especially if you think about the small volumes that I said you typically use. If you're in a pelvic branch, let's say, and you want to embolize, you keep in mind that you can always have 0.5 ml of embolic in that microcatheter. Flushing out 0.5 ml is actually a lot of liquid to introduce into the system, and there's almost no doubt that you're going to get non-target embolization. Then the question comes up, can you just take the catheter out, flush it on the table, and then reuse the microcatheter? I will tell you, of course, it's possible to do. What worries me about that is the assumption that you've gotten out all of the liquid. I think that may in fact be an assumption that's not always correct. What I worry about is that you go back in with that microcatheter, maybe you can use it, maybe you can't. Now you try and put a coil through it, maybe you can get it through, maybe you can't. I just think it's not worth it. I would not recommend doing that.
The only thing I do during procedures is at times you come up to the end of a vial of product, and there's always a decision, do I need to open up another vial because I just need to use a little bit more? Then you remember that you have that 0.5 ml still in the microcatheter. What I would do is take some DMSO and clear some of it, but not all of it. I might take, let's say 0.2 or 0.3 ml, get that little bit out just to top off what I've done, and then get out. I just don't like the idea of having DMSO sitting behind the cast I've created. I just worry about the stability of the cast in that environment. I don't even let it happen. I want the proximal end of the cast to always be the embolic and not DMSO.
Liquid Embolic Use in Large Vessels & High-Flow States
In addition to catheter handling, compatibility with vessel size and flow dynamics are other essential factors that must be carefully considered when using liquid embolics. Liquid embolics should not be injected through large-bore (5-6 Fr) parent catheters, as this approach requires more embolic volume, jeopardizing catheter positioning, and reducing control. Additionally, vessels larger than 5-6 mm in diameter are not ideal candidates for liquid embolics only, since it carries the risk of the flow carrying the agent further away. In these cases, using coils or a plug as backstop can help to create a containment zone.
In high-flow conditions like AV fistulas or pulmonary AVMs, liquid embolics should be avoided unless distal catheter placement and backstopping can ensure complete control over flow direction and embolic deposition.
[Dr. Christopher Beck]
Going back to the setting of trauma, ballpark, what's the biggest vessel you can use it in? Hold on, here's another question, because I just want to throw 100 questions at you and see how good you can do. Can you inject it through your parent catheter, like a 5 or 6 syringe catheter?
[Dr. Gary Siskin]
I would not. I think you're dealing with a lot of volume, and you're also jeopardizing all your work that you did to get that parent catheter where it needed to go. Even if it's pretty straightforward, I still prefer working through a microcatheter just in case something happens with the catheter. I want to just be able to pull it out and replace it. I would not recommend putting it through an L3-5ID catheter. I think it's going to take a lot more embolic, and you run into a greater potential for problems with that.
[Dr. Christopher Beck]
Biggest vessel potentially you can use it in?
[Dr. Gary Siskin]
We've done very large pseudoaneurysms that could be upwards of two-plus centimeters. It can be used in a lot. In those cases, what I'll usually do is put a couple of coils in there to frame out the aneurysm so I have some sense of where it's at, and then just use the liquid to fill the spaces there. That's obviously a very rare case. That's not the norm. I would personally say that I try not to use it as a primary agent in vessels that are larger than five or six millimeters in diameter. What I worry about is the flow carrying it further than I want it to go. If I'm in a large vessel and, for whatever reason I want to use a liquid, I'll usually use a backstop of coils or a plug and then use it to fill space behind that backstop. I think as a primary agent, it's better in some of the smaller vessels.
[Dr. Christopher Beck]
No problem as far as like compatibility where if you want to do a backstop of coils or a plug, filling the void with glue. I guess there's no need to bookend it.
[Dr. Gary Siskin]
No, I wouldn't. First of all, bookending is just going to complicate it. You're not going to want to put it through the delivery catheter. I would say not only is there no problem, I would actually recommend it, especially for a new user of a liquid like LAVA. In fact, it will make the average person feel a bit more confident that they're going to avoid things like non-target embolization, that it's not going to run away from them distally.
If you put, let's say two coils, they can even be pushable coils, just something that will create that backstop. For example, let's imagine the very bottom of the GDA in an upper GI bleeder, put a couple of coils right at that bifurcation at the terminal portion of the GDA. Now, if you separate your catheter, maybe by just a few millimeters, you'll fill that space. The coils will catch the liquid. It's very rare for it to go through, and you'll be able to watch it come back along the course of your catheter. Then all you really have to do is figure out when to stop. All right, that's the biggest task you'll have.
[Dr. Christopher Beck]
Thinking about like some high flow states that maybe you want to avoid, I was thinking also in setting a peripheral hemorrhage, like you get into, I don't know, a bleed, and it turns out to be an AV fistula. Maybe not a good case for-
[Dr. Gary Siskin]
I don't think I would recommend that unless it was very distal and you felt like you would have enough space before the fistula to fill that feeding artery. Let's just take a step back. The one thing you want to avoid is the liquid going somewhere you're not planning for it to go. Obviously, if you're in a high-flow AV fistula, the big worry is that there's going to be a lot of flow directed towards the fistula that's just going to carry the liquid right through the venous side and then God knows where it's going to go.
To me, that would be something I would avoid unless, for example, I was able to get my catheter all the way out there into the fistula itself or just before, fill that space with coils as a backstop, and then very slowly administer a more viscous liquid that the coils will catch. I don't think I would use it as a primary embolic product. Not only is it a traumatic AV fistula, but don't use it for pulmonary AVMs, kind of a similar idea. You just don't want it to go someplace that you can't control. That has significant implications.
Common Errors to Avoid in Liquid Embolization
Dr. Siskin provides a few examples of common clinical errors that occur during a liquid embolization procedure, which often stem from assumptions about distribution, volume, and flow behavior rather than technical errors alone. One case involves mistaking a false endpoint by assuming the liquid embolic has filled the entire space when it layers, especially in larger cavities such as pseudoaneurysms or endoleaks. Estimating volume before injection, even with rough geometry-based calculations, helps verify that the expected volume of embolic is delivered. For example, a 12 mm pseudoaneurysm requires approximately 1 mL of liquid embolic.
Another central concern is uncontrolled reflux, which may be mitigated by advancing the catheter as far distal as safely possible and avoiding initiating embolization in central trunks if reflux could affect critical branches. Most importantly, aiming for the slowest possible injection is the most effective way of maintaining control.
[Dr. Christopher Beck]
I want to talk about best uses and best practices, but the way I want to ask it is, how have you guys messed it up, if you don't mind talking about that? When have you seen it go wrong, and then we can take away, that way everyone can learn from, maybe it wasn't Dr. Siskin, maybe it was the fellow, maybe it was a partner?
[Dr. Gary Siskin]
All right, so let's imagine you're in a pseudoaneurysm, has a little bit of size to it, or better yet, let's say an off-label use in an endoleak, for example. You picture yourself in a large space, and you're trying to fill that space. We got fooled, and it's almost embarrassing to talk about, because we got fooled in terms of reaching an endpoint that was a false endpoint. If you just take probably five seconds to think about it, you'll realize what happened. The patient was laying down, it is a liquid. All the liquid was layering into the dependent portion of the space. As it layered, it looked like we were filling the entire space when in fact we were only filling the dependent half of the space.
We thought that we had filled the space completely. It looked great. We were confident we were done. When we scanned the patient a few days later, we literally saw contrast just flowing right over the liquid. We were, needless to say, not happy with ourselves. We went back in and completed the job. What that taught me is simply to have a clue about the volume of space that you're trying to embolize. It doesn't have to be perfect. You just need to have some idea. As you do more, you'll start to know that. You'll start to know that this vessel is typically 0.8 ml or this size pseudoaneurysm. For example, one of the benchmarks I use is that a 12 millimeter pseudoaneurysm is 1 ml of liquid. It's just a benchmark that I know.
If I'm treating a 12 millimeter pseudoaneurysm and I only use half an ml, I know that something is wrong, that it's a false endpoint. That's why I tend to at least give a little bit of thought to how much embolic I think I'm going to be using. Sometimes it requires me to break out my middle school geometry textbooks and look at a formula for a cylinder for a vessel or a sphere for an aneurysm. Once you do that, it's not as horrific as it sounds to do that. It's just nice to have an idea before you start.
[Dr. Christopher Beck]
Any other scenarios that you can think of? That's a fantastic one, and it's a fantastic rule of thumb, the 12 mls per,--hold on, I've already-- [crosstalk]
[Dr. Gary Siskin]
1 ml for a 12 millimeter pseudoaneurysm.
[Dr. Gary Siskin]
Yes, just be careful about the reflux. Just be very aware of it. I think there are times when it might reflux faster than you think. A lot of that comes down to how quickly we're injecting it. It really comes down to the basics of your technique. If you're injecting it slow, and you're looking for reflux, you'll see it before it's a problem. I tend to like a scenario where I do these embolizations with the idea of controlled reflux in mind. In other words, I'll go further into a vessel. Now, when I see reflux into a more proximal vessel, that becomes my endpoint. I was willing to take that vessel out from the very beginning.
Whereas if I started in the main trunk of that Y, the reflux I saw would have gone somewhere I didn't want it to go. I think that's a very good technique, and it allowed me to really learn how quickly to expect reflux and how to pay attention to that. I would say that's a major step in learning how to use this product appropriately is understanding how quickly it might reflux and how to recognize that.
[Dr. Christopher Beck]
What, if any, problems have you seen with injecting too quickly? I can imagine that is something that you get in a hurry, maybe the adrenaline's flowing. What does that look like? How does the LAVA behave when you're injecting too quickly?
[Dr. Gary Siskin]
I think if you're injecting too quickly, it depends on the vessel that you're in. If you're in a very small vessel, let's say a small two millimeter vessel like you laid out before, and you inject it very quickly, it's going to fill that space quickly, reflux quickly, and go down to many more proximal vessels. That's what I would worry about. If you're in a larger high-flow vessel, you will be able to push this further than you may have intended for it to go. The slow injection will allow it to stay a little bit closer to your catheter and create a bit more of a controlled plug. I hate to give you a very weak, it depends answer, but it does depend on where you are, the size of the vessel, and the amount of flow in that vessel.
[Dr. Christopher Beck]
It probably also depends on the solution, whether you have the 18 to the 34 viscosity.
[Dr. Gary Siskin]
Yes, that's exactly right. Again, I want to reiterate this point because I think it's important that if I'm using the higher viscosity, it means that I'm very close to my target, and there's really no reason to inject it quickly. I think where you might want to even think about that sometimes is when you're further away. There's a sense that if I inject it just a little bit harder, will it bridge that gap a little bit more reliably? I think in many cases, it does bridge that gap more reliably, but you just have to be aware of the potential for reflux.
Podcast Contributors
Cite This Podcast
BackTable, LLC (Producer). (2025, June 3). Ep. 549 – Liquid Embolics: Practical Applications & Techniques [Audio podcast]. Retrieved from https://www.backtable.com
Disclaimer: The Materials available on BackTable.com are for informational and educational purposes only and are not a substitute for the professional judgment of a healthcare professional in diagnosing and treating patients. The opinions expressed by participants of the BackTable Podcast belong solely to the participants, and do not necessarily reflect the views of BackTable.
Podcasts
