So I was asked to talk a little bit about ECMO and just really some basics for the boards. ECMO, as we all know, through COVID and most of us have worked through, has really sort of exploded. So I'm just going to really sort of cover some very basics for people who are not working on ECMO for things to sort of know and also think about referrals and stuff like that. I'm going to talk a little bit about some of the data. With all due respect to one of the speakers yesterday, as somebody who spends a lot of time doing ECMO, I think you have ECMO doctors who believe very highly in ECMO and then maybe people who aren't doing ECMO as much who sort of may be having a slightly less than thrilled approach about ECMO. So I'm going to have a little bit of bias towards ECMO, and I'll explain to you maybe why I have that. I'm also going to show just a couple of cases to help illustrate some of the points that I want to make. I'm not going to ask you any questions per se, but this is a case that I actually was involved in just about 10 days ago. So a 42-year-old male with no known past medical history, sorry, with a known history of OSA, tonsil hypertrophy, tonsil megaly, hypertension, GERD, allergic rhinitis. And he came in for an elective bilateral tonsillectomy, a partial uvelectomy. In the night after the surgery, he started having extensive amounts of oropharyngeal bleeding and went back to the OR for re-exploration of all of this. And then they attempted to extubate him in the OR, and almost immediately, he developed really severe hypoxic respiratory failure, required re-intubation, went to the ICU, got rapid escalating vent requirements for severe hypoxemia. His ABG, when they called the ECMO team on 100% FIO2, PEEP of 20, inhaled lupoprosinol deep sedation and neuromuscular blockade was 7.314944. And his test X-ray pre-op now looks like this. OK? So this patient aspirated a huge amount of blood that was in the posterior pharynx when they attempted to fix all of this and extubate the patient in the OR. And consequently, his lungs were super thick. I mean, his lungs were super stiff, and he wasn't oxygenating. When I arrived at the bedside, so we decided we were going to cannulate this patient, and they were basically bagging this patient continuously to keep a SAT of even 70%. OK? So the first point in this is to say that this is a patient who would not have been randomized into any of the clinical trials that have been done about ECMO versus no ECMO. OK? So that's a very, very key point, because this patient is going to die without something being done. Does that make sense? OK. So we cannulated him. He went on ECMO. No problem whatsoever. About very steadily over the course of several days, his oxygenation improved. His pulmonary compliance improved. He broke down that blood. He was decannulated from ECMO. And in the days prior to me coming here, he was extubated, and he's doing fine and will survive all of this. OK? So extracorporeal life support, these are your sort of various flavors. You've got your VA-ECMO, your VV-ECMO hybrid, which is what we call veno-arterial venous, or some people call veno-veno-arterial ECMO, where we're doing both cardiac and pulmonary support in some hybrid fashion. May or may not also be using like an Impella or some other LV-assist device at the same time or for decompression or a balloon pump. eCPR, which is your not electronic CPR. Well, I guess it's electronic CPR. But it's a ECMO-assisted CPR. This is a growing thing. There's a lot of fantastic videos on YouTube of the French-Parisian EMS systems, like cannulating people on the streets, in the subway in Paris, and in the Louvre Museum, and stuff like that. eCORE, it's estrous corporeal CO2 removal. So you don't actually need a huge amount of blood flow just to get rid of CO2. So you can, throughout the rest of the world, and they are coming to the US very soon, there are devices that you can just add on to a CRRT circuit that can ventilate your patient on a standard CRRT circuit with standard CRRT blood flows. For those of you who use the Baxter brand of CRRT devices, the Prisma, Prismax, or the PrismaFlex, there is something called a Prisma Lung that is undergoing its final FDA authorizations and approvals to be used in the United States. And then there's DCD-ECMO. So this is growing of a way to potentially support organs after donor, after cardiac death, so that they're not rushing as much to harvest the organs, but they can perfuse the organs and stuff like that. I'm not going to talk about that really at all. All right, so let's talk a little bit about ECMO. So this is ELSO's data. So this is just as of a few weeks ago. So the number of centers is in red. You can see that there's been an explosion of centers that report to the ELSO, which is the Exocorporeal Life Support Organization. There's plenty of centers that are not part of ELSO as well. So this is just ELSO's reported data. And you can see the number of cases has totally exploded. Obviously, with COVID, it really went up. And it has thankfully backed down just a little bit, because we're not doing quite as much COVID ECMO. But nonetheless, the growth of ECMO is really, really brisk. Why is that? Well, because we've had significant improvements in technology in the last 40 years. So that's like in the 1980s. And that's now, where we have patients that are ambulating and up and about and have much better equipment. We have better pumps. So we now use centrifugal blood flow pumps that are similar to temporary LVADs, or even implantable LVADs use centrifugal magnetic levitating pumps. And we have integrated circuits, where the pump and the oxygenator are all in one. There's other devices on the market as well that are very similar. I could have showed you lots of different pictures of devices. Please don't interpret that I'm advocating for this brand versus another brand. The way these magnetic levitating pumps work is that they basically create a vortex. They suck blood in. And then due to the spinning of this thing, they create a lot of lateral centrifugal motion of the blood and it ejects the blood out. The advantage of these pumps, as opposed to other types of pumps, is that they don't touch the sidewall at all. They don't have any internal components. They are being driven by an external electromagnet, similar to like remembering your beakers in chemistry, where you would sometimes drop that thing in the beaker and you put it on your warming plate and you turn it on and the whole thing would be coolly spinning. It's very similar to that. And because they don't touch any sidewalls, they have much less hemolysis and damage and everything like that. They don't heat up as much. This is how we've been able to basically develop all of the HeartMate LVADs and everything else like that. So it's all basically the same techniques. The Centromag external LVADs that people see nowadays, instead of the big Pneuma Thoratex that we used to use for LVADs that in Grey's Anatomy she cut the lines and everything, we now have all of these type of pumps. So they're really great, okay? We also have improvement in oxygenators and gas exchange membranes. So I'm gonna show this to you in a little bit, but just so you know, the blood flows around the fibers and the air flows through the fibers. And I just wanna point out that that's different from dialysis, okay? So when you look at your dialyzers, whether it's CRT or intermittent hemo, and you see the big column of blood, the blood is actually traveling through the fibers and the dialysate or the effluent is bathing the outside of the fiber. So this is opposite. And the reason for that is that these tubes, whether it's a dialyzer or an ECMO machine, these tubes are really, really narrow, right? And so they are gonna have a lot of resistance to flow and blood is quite viscous. So that's why we can't flow more than, say, 300 to 350, maybe 500 mLs per minute of blood flow through a hemodialyzer is because the blood is flowing through these tiny fibers, the blood is very viscous and it has a hard time moving. So if you're gonna have to do, excuse me, if you're gonna have to flow four, four and a half liters for ECMO, you're not gonna be able to do that through those fibers. So we reversed the configuration because there's not a lot of resistance to blood flow outside the fibers and then the air is not very viscous and it can run through the fibers. Does that make sense? We'll talk a little bit more about this in a second. But the other advantages that we also have now with better membranes and better devices is that we also integrate, we also sandwich in part of the ECMO circuit some tubes that just carry water so that we can do heat exchange as well. So we can send water of different temperature through there and then as the blood is flowing past it, we can heat or cool the blood and the patients don't get as cold and as uncomfortable. We also have a huge improvement in vascular access catheters, which has helped a lot because now we have catheters that are very reliably and easy to place percutaneously at the bedside. These are drainage cannulas, they're called multistage cannulas because they have holes at the end and along about the first 25 or 30 centimeters of the cannula and we put these in the IVC. And then you've got our return cannulas up there and we also have dual lumen cannulas now that we can use for veno venous ECMO where we can drain from the SVC and the IVC and then return to the right atrium so that we only have one cannulation site. And it looks a little bit like this when you do that. So this is all of this equipment improvements is why there has been such a huge growth and explosion of ECMO because we can do it so much more safely and so much more easily over time. So now, what about ECMO? Why does it work and how does it work? So this is sort of our equation of life, right? We need oxygen delivery and we know from lots of things in critical care, like 20, 30 years ago when I was initially training there was all this talk about we need to maximize delivery of oxygen. And that's not really the case but we do need to deliver oxygen. We have our cardiac output and then we have our oxygen carrying capacity of our blood, right? We have your hemoglobin, your oxygen saturation, and then your amount of dissolved oxygen in your blood. So we need to deliver oxygen, we need cardiac output, and we need oxygen content within our blood. So if our lungs are the only thing that have failed, we can provide oxygen content into the blood with an extracorporeal device and then we can rely on the heart to pump the blood everywhere. If your heart has failed and or your lungs have failed, then we need to potentially do one or both of those and that's gonna determine sort of our configuration. I also just wanna point out that all of you generally have a good sense of what CVVHD or CRT is, CVHDF, so ECMO is basically exactly the same thing. It's two solutions, in this case air and blood as opposed to dialysate and blood separated by a semipermeable membrane and diffusion occurs down concentration gradients for substances that the membrane is permeable to. So in dialysis we use quite leaky membranes that are permeable to lots of different things and in ECMO we're gonna use membranes that are only basically permeable to oxygen and carbon dioxide. Does that make sense? So you have your ECMO membrane, we're gonna send air through there which we call sweep gas, have you heard that word? Sweep gas, essentially that's a dialysate, right? We know in CRT the faster we run the dialysate the more we're gonna clear, right? So the faster we run the gas, the sweep gas, the more we're gonna remove CO2. I'll explain why it's only important for CO2 in a minute. And then we bathe everything, we bathe these fibers in the blood and then oxygen goes down concentration gradients as does carbon dioxide. Generally make sense? Did I demystify it all for you? It's all basically exactly the same thing. All right, now the reason why sweep gas doesn't really make a huge difference on oxygen exchange is because depending on your circuits the oxygen content is so high compared to the blood flow content of oxygen like the PO2 is really, really low and the blood is very desaturated of oxygen and so there's always a really huge gradient from the oxygen side into the blood flow side even if the sweep gas is going really big or really slow. There's a huge concentration gradient from oxygen outside and oxygen into the blood whereas carbon dioxide, we know outside, right? That we have a partial pressure of carbon dioxide but even at its peak our partial pressure of carbon dioxide or dissolved carbon dioxide in our bloodstream is not super, super high. It might be 50, 70, 80. There's not a huge concentration gradient there and so when you run the sweep gas slower then you can basically equilibrate like we would in CRT and we lose our concentration gradient as this air is sort of slowly moving through here. It picks up CO2 and then you don't really have fresh concentration gradient. So the faster we move the air, the faster we're refreshing and we maintain that concentration gradient. Making some sense? Okay. All right, indications. So for VV first obviously we want something that's generally speaking, unless we're using it for a bridge to lung transplantation, we need some sort of reversible acute respiratory failure. Now I recognize sometimes we don't know if it's reversible but where the ECMO team is always gonna look for some potential harm marks of which it might not be reversible. You know, like is there any signs of interstitial lung disease or something else like that? Severe high-threatening life hypoxemia. Okay, inability to achieve safe ventilator settings. This was something that we did a lot in COVID. We were sort of stuck on very high vent settings and neuromuscular blockade and proning and maybe the patient wasn't failing all of that but they weren't getting better after say like 24 to 36 hours. And we sort of said, well, we don't really, you know, a peep of 20, you know, putting on my kidney hat for a little while, peep of 20 is gonna cause the kidneys to fail over the course of a couple of days. So keeping the patient as sort of a single organ failure with lower vent settings, lower thoracic pressure, peak thoracic pressures, all of that can be helpful. And then any time you have any airway issues, so if you've got a tracheal problem or an air leak syndromes of any sort, those are situations. And then oftentimes in asthma, it's the hypercapnia that ends up getting us to cannulate patients. And ECMO for severe, very refractory asthma exacerbations is fantastic. You can get your patients extubated very quickly and sort of move forward. Okay, cardiogenic shock for ECMO. We had a great talk by my predecessor on all of this. The main thing when you're thinking about VECMO is you need to have a bit of an end game. Much more so than in VVECMO, you really do need to think ahead of time. What is the opportunities for VADs or transplants? Or is this a really a bridge to recovery? We do that sometimes. We sometimes put patients on that we know they have no option for a VAD or a transplant. We have to tell the family ahead of time, look, this is not a permanent thing. This is a bridge to see if your loved one can recover from their massive MI. They have other contraindications to why we can't do this. They may not work and they may not get better, but it's too early for us to tell. And then massive pulmonary embolism and an absence of significant respiratory failure. And then we'll talk about when we might do VAV. So I want to make a point, because we make a point of this to say in our program that we have very few at this point absolute contraindications, okay? We've put in people who are older than 75, but we do think about it. Irreversible issues is definitely one that is probably a very strong contraindication. Unwilling to accept blood products. Most of the time these patients do need transfusions, so we have to think long and hard about someone who's unwilling to accept it. If they're metastatic malignancy that's failing treatment, that's definitely a very strong contraindication. But metastatic malignancy that's never been treated before is not necessarily a contraindication. So we put somebody on several years ago, was 18-year-old who had widely metastatic, newly diagnosed testicular cancer with mets to the lungs, couldn't breathe anywhere. We know that's a very chemosensitive tumor, right? So we put that patient on, initiated chemotherapy. The patient recovered very nicely and went home. And then obviously things like advanced liver disease. Now, relative contraindications, if they're on the vent for a very long time, like two weeks or something, we don't like that. There's oftentimes a lot of irreversible inflammation there in the lungs. Severe multi-organ failure, a very profound distributive shock would be probably not a good thing for VV ECMO. We would have to sort of think about VA ECMO or some sort of thing. We personally have had a lot of trouble with patients that are severely immunocompromised, especially PJP pneumonia. I cannot recall the last time we've had a PJP that has successfully liberated from ECMO. Even patients that had not been treated or failing management, like there's a new diagnosis, they just showed up, they had never been on ARVs. You think you do all of that, they should get better. They have not survived. So we have a lot of concern about PJP. And then somebody who may not be a transplant candidate we have to think about. Okay, so in the interest of time, these are just some areas where it might not be successful. But again, we do individualize a little bit on this. All right, so let's talk about the evidence. So we have very few randomized control trials. Okay, let me take a step back. How many of you use CRRT in your ICU? Do you believe very strongly in CRRT? Yes or no? Okay, would any of you be willing to get rid of your CRRT and go back to intermittent HEMO for all of your patients in the ICU? So we have no studies that have demonstrated that there's a clear survival advantage of CRRT either. Okay? It's very hard to do studies that look at whether or not a device clearly is the rate-limiting step to survival or not survival, right? We know that CRRT is better than intermittent HEMO for many things that we care about, right? Especially fluid management and potentially kidney recovery. So even if it doesn't necessarily drastically change survival, if it helps the patients get out of the ICU faster because they get volume off, they maybe not get a trach as well, those are all things that are positive potential things for it, okay? I'm just gonna touch base about this. So the CSER trial was a large trial that looked at transferring patients to ECMO centers that were advanced ARDS centers versus treating them in their local community. Not everyone who went to the ECMO center got ECMO. Only about 50% of the patients got ECMO, but a lot of the other patients got managed at advanced ARDS center. This trial showed pretty definitively in the UK that being cared for at a high volume ARDS center that has ECMO as a possibility was better than staying in a less advanced ARDS hospital or not as experienced center, okay? In H1N1, we had a ton of propensity case match trials that showed the use of ECMO to be beneficial in H1N1 in 2009, 2010, and we have a ton of data in COVID patients, okay? The EOLIA trial, which was talked about yesterday by one of my predecessors that was talking about lung failure, this was a study that did randomize patients to ECMO versus no ECMO in ARDS. Now, let me just talk about why. So this trial showed no difference as the top headline of the trial. Now, let's talk about this. So in this trial, they enrolled about, I think it was about ultimately about 300 people, 250 people that were randomized into the two different arms, okay? They screened about 1,000 patients, and so they basically excluded 700 patients, 750 patients. Of those 750 patients, 250 of the 750 were excluded because they were already on ECMO, okay? So essentially, they randomized a bunch of patients that the ICU team was like, eh, I could go either way on this one, right? Is it any surprise that it was hard to find a benefit if you're looking at the patient and saying, eh, I could go either way on this one, right? Most of the patients that we get called on, we're not like, eh, I could go either way on this one. Most of them we're being called on are people who are not gonna survive the night if they don't get placed on ECMO, okay? So we have to be very careful not to over-interpret the fact that we have negative randomized control trials, okay? I think what you can say is if you're standing in front of the patient and you're sort of like, eh, they're sort of okay, but they're on a lot of vent settings, you can maybe say, let's see how this goes for the night or the next 24 hours. If we're making progress, we're getting better, okay, good, the proning's working, I'm able to come down on my vent settings, you know, they're still urinating and stuff like that, great, but if you come back in the next day or the end of the next day, you've tried to unprone them, they've catastrophically failed the second you've unprone them again, you're putting them back, their PEEP is 20, they're no longer urinating, their creatinine's getting worse. These are patients that you probably need to say, they're failing and we need something else, okay? Make sense? Okay, all right, so let's just talk about a couple of things that you may run into. So remember with VVECMO, we've got deoxygenated blood coming back to the heart, which we are draining out to the ECMO circuit, okay? And we are taking that deoxygenated blood and we are making it oxygenated blood. But we're not draining all of the blood that's returning, we're only draining some fraction of the blood, right? Their cardiac output might be six liters a minute, we're draining four liters a minute. So when we put it all back in, we sort of get purple blood. Does that make sense? It's oxygenated well, but it's not like it has a PO2 of 300, it has like a PO2 of like 100 once it mixes together, okay? All right, and so then it goes to the lungs, which aren't really doing anything, right? Comes back to the heart, the heart pumps out this sort of purplish blood out to the patient. Now, so that means that the determinants of oxygenation, this may come up on your boards, okay? So patients who are on VVECMO can become still refractory hypoxic, all right? And if they are, if you're having trouble oxygenating on VVECMO, it's almost always a problem where your extracorporeal flow fraction is not keeping up with your cardiac output. So these hyperdynamic septic pneumonia, SIRS-E COVID patients have a cardiac output of maybe say like 10 liters a minute, and we're only diverting say four liters a minute. So they're only getting, you know, 40% of their cardiac output is getting oxygenated. Does that make sense? Okay, so the way that we have to deal with this almost always on your boards is going to be to increase the blood flow rate. So we need to increase the flow rate through the ECMO circuit so that we can be diverting 60% or 70% of the flow. Make sense? Okay, so the first thing is always gonna be increase the blood flow rate on VVECMO. That's different than VECMO, okay? On VVECMO, it's always gonna be increase the blood flow rate. If you're still having issues, then you may have to use the lungs a little bit more. We usually try to rest the lungs a lot, but you may have to use the lungs a little bit more. And then you may think about things like, is my, am I a little anemic and other stuff like that. And if all else fails, you will sometimes see us introduce beta blockade to these patients, okay? To artificially lower cardiac output a little bit. But we need to be mindful and careful when we do that. All right, so these are our single lumen cannula strategies where we will sort of have a femoral drainage cannula and an IJ return cannula. And then these are our dual lumens where we'll either do a dual lumen in the neck or more commonly, oftentimes under the subclavian on the left side. All right, what about VA? So in VA configurations, we are draining deoxygenated blood, we are oxygenating it, and then we're returning it into the aorta somewhere, okay? Now, the challenge here is that in the most common configurations, we are doing peripheral VA cannulations. If someone has had CPR or they're in the CCU with an acute MI, we haven't cut open their chest and put the cannula in the proximal aorta. So the blood is returning into the distal aorta and you're gonna have retrograde flow up the aorta of blood. And that's going to meet at some point blood coming anti-grade out of the heart. And you get sort of like a mixing cloud of blood coming up from below versus blood coming out from in the anti-grade fashion. Making sense? That mixing cloud that happens can impact how we're getting, I'm just looking at time, are we good? Am I a little over? Okay, sorry. Okay, sorry, I lost track of time there. I'll speak a little quicker. Okay, so if that mixing cloud is where we have to do potentially VA, VECMO, if the lungs aren't working and we're ejecting from the heart deoxygenated blood, then sometimes we need to oxygenate that blood better and do that. Okay. We always need anticoagulation. We tend to turn the vent settings way, way down to try to rest the lungs, but we don't have any good studies on what is the perfect setting for this. So we try to usually keep the lungs open a little bit with a little bit of peep, but then we turn the driving pressures way down. And then you need to think about tracheostomy and early rehabilitation of your patients. The complications are primarily bleeding-related complications, sometimes complications from the cannulas, and then hemolysis is something we watch very closely because it's a big risk factor for AKI. Okay, so in summary, I just wanna show you one other case real quick. So in summary, newer technology is better. ECMO is showing that the lungs have an amazing capacity to recover, and I just wanna put this out there. There's not a year goes by that we don't get reports in Atlanta of somebody dying from flu, like that was 20 years old. And it's like all over the news, how tragic this 20-year-old died from pneumonia or from flu. And when we talked to our other ECMO centers, no one heard about this patient. That's just unacceptable in the modern era in the United States, okay? We have plenty of remote cannulation options. You should not be working at a hospital where you are not willing to call and refer these patients for ECMO. You should not have patients dying from reversible problems without an ECMO center at least hearing about it. In my personal opinion. All right, so let me just show you the last case, okay? This is a 28-year-old pediatric ICU nurse who fell and broke her collarbone. She developed progressive multifocal bilateral necrotizing MRSA pneumonia. She basically was colonized with pneumonia, probably because she, MRSA, because she worked in a PICU. She broke her collarbone. She wasn't breathing very well to clear her secretions and everything because of a lot of pain. She got a terrible pneumonia, okay? Let me see if I can get this to play. This was her CT scan. Okay. That looks terrible, right? Okay, she developed persistent hypoxic respiratory failure. We couldn't clear this infection whatsoever. And I just wanna again show you this case, and I'm sorry that we're going over. Okay, so we consulted. We consulted. We placed this patient on ECMO. We adjusted the antibiotics. And quite honestly, we were viewing this as a bridge to lung transplantation, okay? So we got her sedation off. We began doing PT, OT over time. But while she was going through this whole process of getting stronger, looking like a good candidate for transplant rehabilitating, her lungs were magically starting to get better. And she ultimately did not, over many weeks, she was on ECMO for about 12 weeks. Over the course of that time, she continued to recover and recover and recover. She never transplanted, and she walked out of the hospital. And six months later, this is what her CT scan of her chest looks like. Okay? So the point of this is to say that the lungs have a huge capacity to recover that I think we didn't give it credit for. And I think we've learned during COVID that even the worst lungs with enough time, many of them can get better, okay? We put a ton of patients on ECMO for COVID. Most of them recovered, and the percentage that we had to transplant was really quite few and far between. So don't sleep on ECMO. If they're gonna die without it, then it's better to at least think about it. And I don't think you're ever gonna get any clinical trials that are gonna randomize patients who are about to die to ECMO or death, okay? All right. Thank you.

ECMO

COVID-19

respiratory failure

VA-ECMO

VV-ECMO

CESAR trial

EOLIA trial