Thank you so much for having me. Obviously, an important topic, complications and bailout strategies. I'll start by saying that you may be the most at risk in the hospital to hurt the patient for pulmonary embolism for a couple of reasons. One, unlike a lot of the patients we treat, these people were oftentimes healthy yesterday, right? So this was the 17-year-old that was playing soccer yesterday, not the 90-year-old getting valve and valve tabber. Second of all, there's no class 1 indication for any of the stuff that we're talking about doing over the last couple of hours. This is a space where 20% of clinicians in the country think you're doing this for RVUs, and if you kill someone when you're doing it for RVUs, that's pretty darn bad. So you want to make sure you know what you're dealing with as you get into it. We've seen this slide a couple of times, what happens with RV physiology and pulmonary embolism. It's important to remember that before we even start these interventional procedures, we could hurt people with the procedure. If you're sedating someone, if you're intubating someone for these procedures, you'll drop their blood pressure, you'll drop their cardiac output, you'll increase their intrathoracic pressure, and they'll die. And they'll die while you're sitting there with a catheter doing a procedure for a non-class 1 indicated procedure. Bleeding is obviously one of the most important things that we talk about. So you've got to start with that full-dose systemic TPA. People bleed. If you give them 100 mg of TPA over 30 minutes, we've known that forever, they bleed a lot. So if you're pushing systemic TPA, you're giving 100 mg, number needed to harm for a major bleed, about 18. Number needed to harm for intracranial hemorrhage, about 78. But we know that. And we've tried for a long time to get away from full-dose TPA beyond the highest risk of patients. We've talked a little bit about catheter-directed thrombolysis. And as the world has changed over the last 5 to 10 years, we've used lower and lower and lower doses of TPA. The reason why that's so important is because the bleeding risk has dropped precipitously. So if you look at knockout PE, which is supposed to be 500 patients representing current practice, we see that 70% of folks are using less than 20, less than 15 mg of TPA. And as a result, the bleed risk has dropped precipitously. In 489 patients with that series, we saw no intracranial hemorrhage. And we saw 1.8% major bleeding within 72 hours by a very conservative strategy. So ISTH major, if your hemoglobin drops from 13 to 11, you were called to bleed. And that only happened 1.8% of the time with modern doses of TPA using the ECOS catheter. Flash, as we look at the flash registry, looking at the ANARI device, large-bore thrombectomy catheter, we see a low 1.4% risk of bleeding. A little bit less conservative categorization of bleed, so BARK3B, you've got a hemoglobin drop of greater than 5, but we're seeing bleeding that happens there. That's true when we're doing these procedures in these patients. We talked a little bit about Optalyse. You know, as we talk about how we're trying to improve and how we're trying to get better, like discussed earlier, Optalyse helped us get to lower and lower doses of TPA. So those small numbers, it showed in these series of patients with as low as 4 mg over 4 hours, improvement in right ventricular dysfunction. That's what drove us to knockout PE. That's how we started getting to using these really, really low doses. On the mechanical thrombectomy side, trying to reduce bleeding, we started to do some things procedurally. You know, we'll use upstream pre-closed and per-closed devices before venous thrombectomy cases. It's always a little interesting, you know, to put a suture in a vein in someone who has a DVT, not certainly indicated for that, but it's been safe. Some combination of venous pre-closed plus or minus figure eights at sheath removal work to help decrease that bleeding and reduce that bleeding in mechanical thrombectomy. The companies have also come out with some exciting tech. So the FlowSaver device, less about reducing bleeding than reducing blood loss, the FlowSaver device for Menari is meant to be able to give blood back into the system, as opposed to early on where we're really removing hundreds and hundreds of cc's of blood, you now have the ability to give it back to the sheath. On the Penumbra side, they developed the capability to limit the actual blood extracted from the catheter using the lightening device, which helps the tubing actually recognize if there's clot or free flow of blood in the device, so it can try to tamp down the amount of free flow blood being extracted. All steps in the right direction to minimize bleeding. The stuff that we worry about the most is even beyond bleeding, though, right? It's what we're doing when we're actually in the pulmonary artery. An ounce of prevention is worth a pound of cure, right? And so we want to be making sure we're accessing the PA safely. We've seen strategies to do that, whether it's using pigtails or balloon-tipped right heart catheters. Personally, I use pigtail catheters when I get up and in. You know, the reason why is, one, because it's pretty easy and it's pretty quick, two, because that pigtail is not going to sub-select some of those small branches the same way a pulmonary artery will. You'll see by deformation of the pigtail occur in a small branch, and that's where a lot of your problems come with large board mechanical devices. Pulmonary angiograms, that was one of the best talks I saw in the last session ever about pulmonary angiography. The ability to really know where you are becomes extremely, extremely important. But bad things happen. So anomalous anatomy exists. One in four people sitting in this room have a PFO. The first time I saw this picture, I had no idea what was going on. This was someone who had kind of done their first large board thrombectomy catheter. They managed to stick it across a PFO. They took a picture. They thought it looked good. They didn't see any clot. They didn't see any massive seizure on the table. And it turned out they put a 24-frame sheath across a PFO into the left atrium, and they blew a bunch of air into the brain. And so if you don't know what you're looking at, really, really bad things can happen. And one in four of your patients have a PFO. Will every wire across a PFO? No. But when it happens, it can be absolutely terrible. So you want to make sure you're comfortable with the tech. This wasn't even recognized on the table. They stopped the case because of the seizure, and then went on with their day. Not all clot can be withdrawn. We love thrombectomy. We love how it looks on the back table. Sometimes it's not going to come out. There's fancy names for that. You can call it lollipop clot, clot sitting on the back of the catheter. This was thrombus in transit in the right ventricle. It's a great picture of a huge clot bigger than my thumb, right? But you're not supposed to see this picture. This was clot that didn't come out through the cannula. This is clot that got stuck. And thankfully, it all stayed organized as we dragged it out. It made for a nice picture, but that clot's not supposed to be in the cannula. This was actually a bear to deal with. We'd aspirate onto it multiple times, lost it multiple times, and then managed to drag it out. Lollipop clot is also sometimes even worse. This is a slightly different case. This was an angiovac case that was done by our cardiac surgery colleagues for someone who was an active drug user with massive tricuspid valve endocarditis. The angiovac, they thought they got it out. It was TE-guided. You know, high-fiving in the cath lab. You know, 24 hours later, they tried to die. They were newly hypoxic. We did a CAT scan, and all that stuff that was sitting on the tricuspid valve is now sitting distal. Same thing can happen with tronclansit. In this case, we went and got it with an NRE catheter. We took it out. It certainly didn't look like clot, but we found the tricuspid valve and a bunch of bacteria. This one's even worse. So, you know, we talked about utilization of thrombectomy in the setting of the highest risk of cases. This was a patient who actually was on ECMO, and so we tapped into the circuit. We did multiple passes of an NRE. We kept feeling like we were latching onto something. It wasn't pulling back. We would drag it back. If you've ever done an NRE on ECMO, you know that you have to turn off your venous outflow cannula at the time you're dragging something back. Otherwise, something bad can happen. The clot gets sucked into your venous cannula. We did this three times. On the third time, you know, we still couldn't get it out. Immediately, our ECMO flows dropped. We brought our catheter back up. We took a picture, and the clot was gone. And so that's pretty good, except now the patient's ECMO circuit is collapsing. So where's the clot? There's the clot. So the clot got sucked over onto the venous outflow cannula of our ECMO, and we still couldn't get it, right? We weren't able to drag this out. We tried three or four times. We upsized the sheath. This was before the 26-wrench sheath existed for an NRE, so we put a big 26-wrench dry seal. Still couldn't get the thing out. We finally got this thing out by actually latching the NRE catheter onto the clot, leaving the wire up, taking the clot and the sheath out, and dragging the whole thing out of the leg. The next step was doing a surgical cut down in order to get it. It came out. I thought we'd ripped out the IVC, and we got a big chunk of clot. But while we were doing this, the patient was dying because we broke the ECMO circuit, and we had no way of fixing it until we got that off there. That was pretty rough. So not all lollipopped clot is easy or, you know, something inconsequential to deal with. Best case scenario, it doesn't come out. Worst case scenario, it starts to come out, and it goes somewhere bad. Here's another case. Expert operator went up, tried to get clot out, grabbed onto it, lollipopped a couple times, got some of it out. So there's some improvement in distal perfusion. But then at the end of the case, she couldn't move her arm, activated a stroke code. The only nice part is we were in the cath lab. Our neurointerventionalists came in. They were like, you're doing what? Why are you doing this to this young woman with a P? And it's like, well, you know, we're trying to prove that we can treat patients with pulmonary embolism in this fashion. She had a big stroke. So she had a big M1 MCA, total occlusion. This patient died of this, actually. Terrible. So, you know, they managed to do a thrombectomy. They managed to get a large thrombus. This was another patient with a large PFO. I know in my hospital, like in many of our hospitals, sometimes you just don't have the echo before you start doing these therapies, especially in the modern era where it's hard to get echo techs around for everything you need. You just don't always know if someone has a large ASD or a large PFO before you get started. And this is rare, but it happens, and it's something for us to think about as we go. These are the pictures of aspiration of that MCA thrombus. Perforation. Man, perforation is terrible in this bed. So what are the parts you can definitely do if you have a perforation? So if it's bad, you have to do things quicker than you can do anything interventional. So you're putting your bad lung down, you're selectively intubating that patient. As you're intubating them, you're thinking about my first slide, which is you can kill someone with intubation and positive pressure. You can decide, is it a wire perforation, can you wait and see? Are you holding anticoagulation? Are you pushing protamine? And this person that came in with an intermediate to high-risk P is like the worst patient possible to give protamine to. If it's proximal, you may be calling cardiac surgery. We can think about covered stents. I don't know if anyone's ever managed to put a covered stent in. Jay Geary, I'm looking for you. I've never seen anyone do that in the midst of an acute bleed. Small vessel perforation, you know, plug occlusion or coils. I've never had to, thank God, have a real proximal perforation that I had to covered stent. This is the closest I can get that tells me I might be able to do it in the future. This was someone who had a massive lymphoma that they were kind of drowning in back in 2018, and so it was just occluding that left main pulmonary artery. We managed to get up, wire blindly, sub-select a large branch, and then deploy a covered VBX stent. When we did, we actually got some flow back into the lung. Maybe if you had a proximal perforation, you could do something like that, maybe. I don't know. Never seen it reported. The nice thing is I can tell you that after this, a very different disease, five years later that covered stent is still open. This guy got enrolled into an immunologic trial for lymphoma and he survived. So can you put covered stents in the lung? The answer to that question is yes. Can you do that in the midst of someone who's hemorrhaging blood into their lung? The answer to that is I have no idea, and if anyone else in the room knows, please let me know. Small perforations that you want to coil or plug. You know, I've coiled and plugged a lot of things in my life, soft, non-dense. Pulmonary tissue would be a really hard one to do it. You know, we talk about low-resisted tissue. The lung is about as low-resisted as possible, and you're just going to ooze and bleed around those coils. You know, the reason why I mentioned the good Dr. Geary, who was our last moderator, is because he actually has a case of managing to get a plug into someone. Those BPA talks on pulmonary angiography is the most important part of this. Those heavy lateral or those ipsilateral oblique projections to let you really see what you're doing. In the midst of an acute bleed, that's really, really, really, really difficult. So you know, these are bronchiolar arteries that we managed to plug in the midst of bleeding malignancies, kind of a similar concept. You can get a catheter and you can introduce—remember your narrow catheters are 90 centimeters. You can introduce coronary guides like we do to subselect, and then you can deploy plugs and coils if you need to. But man, in the midst of an acute bleed, that's really, really, really, really hard. And outside of that one case, I haven't seen anyone else kind of do that successfully. Now, we're only treating 4% of patients, roughly, right, with interventional therapies. In the years that come, you guys and other leaders throughout the country are hopefully going to be expanding that rapidly, especially as we start to get to class 1 indications. And as we've seen in every procedural field, as it expands, more and more terrible things happen. And so we've just got to be real thoughtful about what we're doing in these patient populations. So in conclusion, this is the worst group of people in your hospital to hurt, because a lot of folks don't believe you should be doing this in the first place, and because of the fact that they were healthy before, and because of the fact that there's no class 1 indication. Two, bleeding is a risk for all PE therapies, though we're getting better. You know, catheter-directed thrombolysis has gotten down with ECOS and ultrasound assistance to really low doses with effective outcomes. Corporombectomies had improvements in terms of blood loss and trying to mitigate bleeding as well. Ounce of prevention greater than a pound of cure, by a lot, because we're not really good at treating these things. I could perforate a lot of stuff and fix it, this I can't really do. Bad lung down, selective intubation of the good lung, really important. Lollipop clot can be annoying or it can be a nightmare. And then last but not least, plugs and covered stents, maybe, I think, maybe, probably, I don't know. If anyone knows, please tell me. So thank you, guys.

complications

bailout strategies

pulmonary embolism

risk

bleeding