Skip to content
Prolonged ECMO episode artwork

May 31, 2025 43 min

Prolonged ECMO

from Maryland CC Project

Episode details come from the podcast's RSS feed.

Embed this episode

Ready to play

Prolonged ECMO

0:00 43:31
of matches

Transcript Auto-generated, may contain errors

I'm pleased to welcome Dr. Adam Green, who is an associate professor of medicine at Cooper Medical School, Rowan University. Dr. Green is a critical care intensivist at Cooper University Healthcare, has held many roles in different medical director ships at multiple hospital sites. Dr. Green received his combined MD MBA at Drexel University, completed internal medicine residency at North Shore, Long Island Jewish, followed by primary critical care at Montifor Medical Center.

Dr. Green has been a part of Extracorporeal Membrane Oxygenation Team at Cooper since its inception. He's actively involved in medical education, serving as teaching faculty for a critical care fellowship at both Cooper University Healthcare and Inspira Health, and has been a core faculty on the fellowship program. He is involved in a lot of clinical research involving ECMO, and he's going to be talking about his research work today regarding survival of prolonged ECMO for ARDS. Please join me in welcoming Dr. Green. Thank you very much.

Thank you. I really appreciate that. It's always so embarrassing listening to those. And I guess I did those things. So I really appreciate this opportunity. I was kind of honored to receive the email. And I hope that you guys can find this work as interesting as I find it. So the title of the talk is Prolonged ECMO, specifically BV ECMO. And when should we stop? And so...

So I should say I have no relevant disclosures. I'm a paid consultant for Cerebell, which is a point-of-care EEG company. But I have, in terms of this talk, nothing that's pertinent. And there's three major objectives I'm hoping to achieve in this next hour. Number one, I hope that we can talk about mortality in association with prolonged ECMO. And as ECMO duration increases, what does that do to mortality? Number two, I want to talk about the use of prolonged ECMO and has that changed over time? Are we doing more of it, less of it? What is the trends there? And then third, I want to talk about the complications that occur in this cohort of patients and does that really matter?

Let's start with a story. I think it's always good to start with a story. So this is a patient I took care of. She was 41 years old. She'd recently been diagnosed with multiple sclerosis. It's on a variety of immunological agents. She was obese with a BMI of 39.

And she came to us with kind of a sub-acute presentation. She had developed respiratory symptoms maybe four weeks prior to her coming. She had gotten an oral course of outpatient steroids and antibiotics. Symptoms progressed. She was admitted to an outside hospital for a couple days, got IV antibiotics, continued to get discharged, continued to get worse, represented at a hospital 11 days ago, was found to be influenzae positive at that time, and had progressive hypoxemia, high-flow nasal cannula for several days, EPAP for several days, eventually intubation, at which point she was transferred to us for the reason of ECMO eval. I'm sure you guys get the same, but the number of patients that come for ECMO eval never surprises me. And so this is what she looked like when I saw her.

She was undergoing her first prone session. She was sedated, fully paralyzed. You can see just bilateral dense consolidations, probably worse in the left than the right. You see the endotracheal tube in the central line. And that's how she came to us. Her vent at that time, she was on lung protective ventilation, the total volume of 380, respiratory rate of 32, PEEP of 8, because she had very stiff, noncompliant lungs. So despite only a PEEP of 8, her plateau was 35. And you can see her PF was about 100, and she had clear inability to ventilate despite that high minute ventilation. I should say that she was in a touch of shock. Levo was less than 5. Her creatinine had been 1 the day before, and it is now 4 with about 400 cc's of urine output. So it was in renal failure. She had a white count. Her coags were okay. Her platelets were okay. The rest was relatively benign. Our infectious workup showed she was flu-way positive, had COVID, and had some rare MRSA in her sputum.

So what do we do? Right. And I hope this projects well. And this isn't necessarily our criteria. I think our criteria changes. And it's like it's always evolving based on on the literature. But this is from Aeolia, which I think is a good starting point just for conversation. So what do we do? Do we cannulate her? So I think you can make a strong argument. She has ARDS. I will admit the subacute nature of her disease scared us a little. I much prefer they get sick on Tuesday. And I'm looking at them Thursday. This was over some month of disease. So that's always.

scary like is she going to be is this a reversible process but she meets the criteria her pH is less than 7.25 despite a high minute ventilation and even here her plateau was was higher than what's there so she meets criteria most likely for inclusion to cannulate however there were some things that scared us i know i highlighted mechanical ventilation over seven days and she was really only ventilated this would be day two however specifically in covid the use of high flow and bpap has been shown to be analogous due to patient self-inflicted lung injury to mechanical ventilation so that was frightening and that does she have irreversible damage um also she's not a lung transplant candidate and i'm sure you guys talk about that what's the destination if you don't get better her bmi I would exclude her most likely from most transplant centers. She was immunocompromised, and she has this multiple sclerosis and chronic CNS disease, all of which scared us. So what do you think? Should we cannulate her? This is an ECMO talk. So, of course, we did. We cannulated her. So on hospital day two, she had a right IJ return 20 French cannula and a right femoral 25 French drainage cannula. She was started on CRT through the ECMO circuit. She was then excavated on hospital day three, got re-intubated on hospital day four. She then had a pretty uneventful couple days until on probably hospital day seven, she had a hemodynamic collapse, and POCA showed a severely dilated.

RV, and we thought this was probably RV shock. precipitated by the ECMO flow, hypoxia, et cetera. So we made a decision to try to transition her to VP ECMO. So the way we did that is we converted her first to FEMFEM to open up the IJ, and then we put a pulmonary artery return cannula in. During that process, she had a cardiac arrest that was brief. We got ROSC and seemed to be about the same neurological status. She then had an abrupt circuit thrombosis in hospital day 10, had another cardiac arrest, again, seemed to have recovered. So at this point, this is what she looks like. We're at hospital day 11, ECMO day 10 or 11. She have a dense bilateral consolidated process. Seems to have progressed. You have a VP. You can see that PA return cannula. She's requiring full sedation.

She's on APRV now and really non-compliant lungs and is fully ECMO-dependent. At this time, I think her sweep was probably about five and a half liters. She was 100% on the blender, 100% on the vent. In addition, she started showing kind of dusky distal extremities, her right hand, the worst, and just very poor perfusion throughout. So what do we do? I think you have a patient who's been supported for nearly two weeks and has failed everything despite that. And there's an argument to be made that we should talk to the family and discuss really what the next steps are. But what do we do?

We continued, of course. So here's hospital day 12 to 27. A couple events to do. Hospital day 19, she gets an open trach. In the OR, on the way to the OR, the battery on the ECMO dies. She requires hand crank. And, you know, the hypoxemia recovers from that. Hospital day 25, she has another oxygenator failure that requires an emergent change. She develops MRSA bacteremia, maybe from her lungs, maybe from the cannulas. On hospital day 27 that we treat, she has fluctuating thrombocytopenia, as all of these patients do. She develops the right thermal DDT. And then there's this, she develops this peripheral.

eosinophilia in this rash that we thought was maybe duress associated to due to the Cephapine she'd been receiving. But a statement can easily be made that she's clearly fully ECMO dependent. Her compliance is unchanged. It's dismal with tidal volumes in the 50s to 100. You can see the trend of her sweep gas from day 12 to 27. She's up to 12 liters of sweep and her FO2 in the vent is 100%. So what do we do? Again, a month of ECMO, support her the best we can and really no significant improvement whatsoever. And so we continued.

She has 10 days of quiet. And then we're working on sedation. We're trying to wake her up. We're trying to really mobilize her, do all of the things that we do kind of specifically for our BV patients. And then she has an abrupt worsening her hemodynamics. And I made it bold and red because it truly was dramatic when it happened. She had precipitous drop in her hemoglobin. And she had this x-ray, which shows an even worse consolidated process on that left hemithorax and CT, which shows an active hemothorax accumulating in the left side. What do we do? Well, we stopped her intercoagulation. We did not reverse her. She had already had several thrombotic events that led to catastrophic problems with the circuit. We resuscitated her and we waited. And miraculously, she got back.

better. She got better. And we had then kind of another period of quiet between hospital day 38 and 75. During that period, she developed multi-drug resistant Klebsie LA and her sputum. She developed a sacral wound that required a wound vac and frequent debridement by our surgical colleagues. Eventually we were able to get her off all of sedation. She was awake. She was participating with PT. She was, her compliance had improved, but she was still definitely ECMO dependent. So, and you can see her tidal volume is now 500 with a peak of 24. And I really attribute that to being awake and working with physical therapy, but still ECMO dependent. So what do we do, right? She's

75 days into her course, she's requires mechanical support and do we continue? And so I took this blurb because it was like a narrative about COVID and when you transplant, when you should refer to transplant. And I just liked the way it was worded. And it says patients were considered eligible for lung transplant in case they had shown no longitudinal evidence of lung recovery. or clinical improvement despite mechanical ventilation, ECMO support, and a combination of those, right? After four to six weeks, she clearly qualified and met those criteria. You can see the citations of that statement are highly regarded. And so the problem is that she's not a transplant candidate. We talked her into multiple centers and they had said no. And so I have a woman who's awake. She's doing those thousand piece Lego sets. She's watching movies. She's able to communicate. She's walking down the hall with PT, but she's stuck on ECMO. So we did like anyone would do is we just gave her more time until hospital day 76. Oh no, not again. Yes, she developed again, acute hypoxia, acute hypotension, and a cardiac arrest.

Of course, this happened during our critical care fellowship dinner when it was a trauma surgeon watching her, but we all rushed there. We resuscitated her, stabilized her, and we kind of entered the same period where she was back to the same neurological status miraculously. And she was able to wean off sedation, start working with PT. The culprit ended up being that same Klebsiella in her lungs was now growing in her blood. So she developed a bacteremia and acute, acute RD failure, probably from the hypoxia associated with that. And so here we are, we now have another 20 days where we slowly wean her sweet gas. And by slow, I'm talking about a quarter liter per minute, like every three or four days or every week, very slow period until we could get her to recirculate. And we were able to then decannulate her in hospital day 97.

She was transitioned intermittent hemodialysis on hospital day 103. She went to tricolor on day 120. She had renal recovery. Her permacath was removed, and she was discharged to an LTAC on hospital day 131. She was there for about two and a half weeks where her drink was decannulated and ultimately she was discharged home where she currently is with her family. And so truly a remarkable case. So did we just get lucky or do these patients that have bad ARDS and are able to avoid catastrophic events or do they recover? So what I did then is I decided, let's look at our data. What have we done? Now, this is very rudimentary approach, but I just looked at lung, the run length and the percent alive and percent deceased. And you could see this over the last several years for VP for ARDS. And you can see that still the most patients were, of course, shorter runs, but maybe what?

28 of those patients, was that, I don't know, do the math, 30, 40% of them had runs over 20 days. And quite a few of those, maybe 30 or 40% lived. So that made me decide, I should really, we should really look at the literature. What does it say about this population and what I'm going to call long run BP ECMO? So this was published in 2011. It was a single center retrospective study. They divided it into group A, which is less than 10 days, group B, which is 10 to 20 days, and group C is over 20 days. And they looked at survivorship and what they found.

was that group A had about 59% survived and group C about 52% survived. Now, group B, for some reason, only 31% survived, but I'm going to attribute that to low sample size. But for the most part, their conclusion was that length of BV ECMO, the length of the run didn't correspond to the outcomes and that that should not be a clinical factor. They found age did, and I think they found the use of inhalometric oxide was associated with outcome in their multivariable logistic regression. But for the most part, the length of the run did not.

But this was old in a small sample size. So what else did we find? And this was pre-COVID. So there's a handful of case reports during the COVID. You guys would probably find even more. But I have three that I'm going to put out. This was a gentleman who was on for 111 days on VVACMO for COVID ARDS who had full recovery. This was published to the gentleman who, actually I think it was a woman, who was on for 136 days. I don't believe this was for COVID. But still, survived a very long...

run and have a full recovery. I think this, oh no, this was a gentleman and he had refused lung transplant. It had been offered and he had declined and he waited it out and he was correct. Had a couple of complications as you could see, but, but made it through. This one is a little bit different. They were on for 265 days and this patient was ultimately had a recirculation trial and then for several days and then was decannulated, but ultimately died three or four days later due to, um,

due to hypercarbia and just increased dead space fraction. And so this was not necessarily a successful case in terms of mortality, but what it showed is that 265 days of ECMO occurred without a single major catastrophic ECMO event. And I remember during my training, the thought was always, you couldn't support these patients super long just because something was going to happen. And I think this kind of debunks that. And I think we've started seeing that. We are really good at putting patients on DVECMO and supporting them for very long periods of time. So is there anything more that case reports in more recent time? And this is kind of a choppy slide. And this is looking at...

patients, single center retrospective. You can see there's 140 patients, 12 of them were on for over 50 days. The mortality of that group was 42%, so lower than the 118 group of 53%. And the reason for that is of those, I think eight that survived, six of them had lung transplant, something like that. So that kind of artificially skews the mortality. And several of them had a recovery, 11 of these 12 were COVID, but still it's kind of showing the same picture is support these patients, either get them to lung transplant or hope for native lung recovery, all the same message. The two tables that are two figures there, A and B, we don't need to get too much into the details. I just found it a little bit interesting. A showing the number of patients experiencing

at least one event. Spoiler alert is that all of them have at least one event. And the time-to-first event too is shown where different events kind of have different implications, but they all end up having multiple events in multiple time periods. So this study here, and I promise this is the last study that I will actually get to my talk, is looking at the ELSO database. And we will get into the ELSO database and what that means because I think that's, and I'm sure many of you are familiar with it, but I think it's an important part of understanding this talk. And what they did was they looked between 1989 and 2013.

and they defined prolonged ECMO as over 14 days. I think the overall mortality was roughly 45%. And when we look at figure one, you can see on the y-axis, the total number of patients, and then on the x-axis, the number of weeks supported. So as you go right and have longer support, fewer patients are on for that long. That makes sense. And the mortality, when you look, there seems to be a downtrend in mortality, or survivorship, I apologize. And as you're on longer, you're less likely to survive. However, that was not statistically significant. So overall, about 45.4%. Those over six weeks, 41%. And really clearly, there's a trend there. They also showed you...

They showed on the right, if we look at figure two, I think this is really neat, is it looks at the total number of cases in red based on year and the blue, which is prolonged cases. And you can see through time, more ECMO is being done and more prolonged cases are being done. And when they broke these up into two time periods, I forget exactly what it was. I think maybe 1989, 2006, something like that. They found that the cases in that first time period, the first half of that chart did worse than the second half. And that mortality was actually worse in the earlier group. I think that's a good thing, right? That tells us that we were getting better at doing ECMO during that time period.

Good. So what is the ELSO database? And so I want to spend just a little bit of time going through this because it's kind of the crux to how we got our data and about our study. There's 580 centers, maybe even more by now, in the database. They collect a variety of data. They collect demographics. They collect the physiological data pre-ECMO, the type of ECMO, the type of equipment, the complications, and the overall kind of outcomes of these patients. And it does, I will say, it does a very nice job of being granular enough.

that they can convince people to put data in, but not so granular that it is too cumbersome to collect the data. Because obviously, this is all sweat equity, people putting it in. I've put a handful in, and it does take time. And in that database, there's almost a quarter million patients. And you can see broken up into age and type. And really, there's a wealth of knowledge in there. When you want to do a study using the database, there is a form you fill out. You must be an ELSO center, which you guys are. And you must be an ELSO center. You must then submit it. Exactly. There's much more of the form than just this about what the specifics of your study are. It goes through an initial committee. And if it's considered to be a large database, which all of our studies usually are, COVID will be large.

large data, BB areas, they're mostly all large database. There's a second committee that looks at it. You must have at least three centers on it, and you must have someone who's at least a master's level statistician to make sure that it's feasible. If you get approved, the data is yours for that topic for a year. So they are very particular about that. I've submitted proposals and had to wait several months to resubmit as that that was already being investigated. And the number of publications is really impressive. Many of these are in high-impact journals, and you can see that recent times, 2022-49 publications, 2023-34. So it's really a great source of information.

So that gets, that brings us to my talk. So this, this is where we publish in critical care of medicine. You can see the coauthors there. Just want to highlight a couple of Dr. Barry and Dr. Damon, two of my colleagues, Dr. Racklees from Duke. Dr. Conrad is a legend in the field and has a master's in stats and is an intensivist down at LSU Shreveport. Dr. King, I've worked with a ton at Innova and is really just an amazing position. And then the lead author is a intensivist. I went to Cooper Medical School and did training at Walter Reed and is also has a stats, a master's in stats. So this is the paper we published recently in critical care of medicine. And so what do we do? We did an also registry analysis between January 12th, 2012 and December of 2022. We looked at all patients with the ECMO where their initial support was.

BVD ECMO for ARDS. And we really want to look at the impact of ECMO duration on subsequent BVD ECMO decannulation survival. There's a whole bunch of data I'll show you in terms of secondary analysis, but we wanted to know how the rate changed over decades or time. We wanted to look at the outcomes in those in the extreme use of ECMO, which we called over 50 days. We wanted to look to see if case volume mattered. It didn't matter if you do a lot of ECMO in terms of the probability you're going to do a prolonged case, so you're more patient, and the probability that those prolonged cases would survive. And then we did sensitivity analysis where we removed all of the COVID ARDS just to make sure that wasn't influencing our results.

So I don't want to get too into the stats, but we divided the groups and we talked about timeframes into three different groups. The latter group is the 2020 to 2022, which is really the COVID period. And then the other two are kind of divided into half. So there are three timeframes, which you'll see on subsequent slides. We did death and discharge between the groups using a multivariable just regression.

We didn't do the total REST score, but rather the components for a variety of reasons. And then to determine case volume, we used 2012 to 2017 as our benchmark for the number of cases those centers did and compared it to their outcomes in 2018 to 2022. So that's how we did that. And we looked at, as I said, both the probability prolonged use and the outcomes of those patients. And we looked to see if the reason they discontinued changed over time, but I don't think I present the results. So here's the first table. And so this is kind of our all overall patients.

analysis, there were 13,861 total patients. Of those, about 30% of them were supported for over 21 days. That's 4,000 patients. And I think that that alone to me is a remarkable statement to make, is that almost a third of the patients that get put on ECMO for ARDS are on for over three weeks. And that probably mirrors my experience, is that the idea of putting them on and having them off in a week, it seems much less frequently. I frequently have it. I feel like every time we cannulate, I'm in for the long run. And then about 1,000 of those patients, or 7.1%, were on for 50 days or more. And I think that these are common realities of ECMO for ARDS.

And so a couple things from these tables. This table is going to be showing you all patients less than 21 over 21 and the peak value for a handful of characteristics. So I want to highlight. Just a couple of them. We don't need to go through all of them. The first is mechanical ventilation. So this number of hours pre-cannulation. And you can see that in the shorter run group, they had a P-value of less than 0.001 in terms of they had less mechanical ventilation pre-ECMO cannulation than the longer group. So the question is, why is that? Could it be because there's more ventilator-induced lung injury in the longer group? I think it's fair to say that if you're going to go down ECMO, you should probably do it sooner and convert to lung rest than to weight. Or is it that in the longer group are less desirable candidates and we're trying to optimize medical management and those comorbidities or reasons that make them less desirable is what's influencing the need for long runs. Who knows? Either one or maybe a combination. The second, I think,

think is really interesting is this acute associated non-pulmonary infection and that they're much more likely to be in the long run. And I think maybe this is analogous for just complexity of disease. I think about that patient with influenza ARDS and what it's like when they go on ECMO versus the patient with pancreatitis induced ARDS. Clearly that non-pulmonary pancreatitis, that is a much more complex patient. So maybe that's why.

The next thing I want to highlight is this, which is looking at different characteristics between the two groups. And so the prolonged group had more paralysis and inhaled nitroxide. I think those are equivalent to degree of hypoxemia, more likely to need paralysis and inhaled pulmonary vasodilators. So that makes sense to me. You're more hypoxic. You have more severe disease. You need ECMO longer. Number two, they're less likely to have cardiac arrest and bicarb. That also makes sense. Those tend to be fatal events. If you have a cardiac arrest or pushing bicarb because of really severe acidosis, you're more likely to die. So it's hard to need ECMO for 22 days if you have a cardiac arrest that you don't survive at day three.

They had a higher PaCO2 in the prolonged group. Again, indicative of just bad ventilation. That makes sense to me. And then the peak airway pressure are the same. And so, you know, if I was presenting to my fellows, I'd say, why is that? And I think unless you kind of know what the recommendations and the approach to ventilating patients on the neck mode, you wouldn't be able to come up with the answer. And the answer to me is relatively simple, is that we tend to ventilate our patients on pressure control. We tend to do 15 over 10. That tends to be our approach. And we just want to eliminate barotrauma. And if that results in hypoventilation, that's okay. We'll just use sweet gas to account for it. So that's probably the explanation there.

So this is looking at a couple things. So on the left, this is figure one. So on the left is figure A. And what this is looking at is the average run length based on year. And you can see that as you go through from 2012 all the way to 2019, there really doesn't seem to be a difference in the average run length. However, 2020 and 2021 happened, and there seems to be an increase. I think you can attribute that solely to COVID. We all know that it was a different beast. It took longer to recover. And then you can see in 2022, as COVID goes away, we start to trend back down towards the baseline. And that is statistically significant. I'll say several times, I think, throughout this is that those patients were on for longer than when compared to the other group.

we look at table or figure B, there's a couple of pieces of information here. So the x-axis is the year of support. There's two y-axis. So on the left is a proportion of cases that are supported for over 21 days. And that refers to those blue boxes along the bottom of the histogram there. And so what it's showing you is that it's a proportion of patients that are supported for over 21 days based on time. And once again, you can see that there are more percent on longer duration during those COVID years than other years. Now on the right-hand side, y-axis is mortality. So the orange dots is the actual value with the green kind of fitted line.

And I think I could make the statement that it's relatively unchanged throughout the years. Maybe there's a downtrend in the beginning, but for the most part, it's unchanged. And more importantly, it's unchanged in that 2020 and 2021 year where there was an increase in duration, that that increase in duration doesn't, you would think that there would be a dip in mortality because they're sicker, but there in fact wasn't, that they stayed the same. So that brings us to this table, and you'll notice these tables, and you'll notice that I'm not going to have any writing on there, and there's no boxes, because I think I felt obligated to put this on, but I don't want to spend a ton of time looking into this and really trying to dive into it, because I think it detracts from really what the point of the study is. But I think they're necessary to go over. So E-table 4, the one on the left, is from our supplemental data, and you can see it's broken down into three time frames that I mentioned. COVID on the right, 2020 to 2022, and the other two divided, 2017 to 2019, and 2012 to 2016. And these are all patients supported with ECMO.

in our data set. Table two is from the actual manuscript, same time periods, but just the prolonged periods. And the statement I'd like to make is that the age, the gender, the race, the BMI, a lot of the respiratory components were the same between the groups. If I were to highlight things that are different and we're looking at the one at the left, you can see that again, the percent of prolonged cases is higher in the more recent. group, the 2020 to 2022. And also in the total cohort, not in the long cohort, so I don't want you to get confused with that, there was a high mortality in the more recent group and a less likely to be discharged alive when compared to the other more historic time periods. Interesting, and I think this is shown in multiple of our tables, is that the second to the bottom, I don't know if you can see my thing, the second to the bottom row, which is ECLS complication, is incredibly low. So these really catastrophic ECMO-related complications, we found to be like 1% of the patients. It almost never happened. Those accidental decannulations or the abrupt circuit change that leads to mortality, really, really unlikely to happen.

So this table is an important table to look at. And this is looking at those on shorter run ECMO, less than 21 days, comparing mortality to those on longer run ECMO. And I'm going to spend this whole time talking about how long run ECMO is safe and we need to do it and don't give up on your patients. And that's definitely the message I want you to take away from this. Short run ECMO, they do better than long run ECMO. So the less than 21 days de-ceased or 33%, a third of them in the registry died versus almost 50% in the over 21 group with a P-value of less than 0.001. So mortality is certainly lower in the shorter run group. And this is seen in discharge alive. And once again, we look at the complication rate incredibly low. This is probably my favorite figure or maybe my second favorite figure in the paper. And I think it's just has some really neat information in it. So once again, just because I can, I know there's a lot and it's hard to look at. I'm just going to go through what the figure is showing you. So on the X-axis is duration of ECLS from zero to 150 days. There are two Y-axis. The one on the left is a probability.

of discharge to live. That's reflected again in those orange dots in the blue line. And then the one on the right is the total number of patients, which corresponds to that dotted green line. And so there's two ways to look at this and kind of divide it into two ways. First, the patients that were less than, that were supported for less than 21 days. And we found that the duration of ECMO was associated with mortality. So there was, they adjusted for respite score, odds ratio was 0.97 with a P-value of less than 0.001. And I think it came out to something like every additional day you're on ECMO, in that less than 21 day group, you had a 3% chance increase of dying. So in that shorter run, as you're recovering from your acute illness,

Patients that are on for less time do better, and that the longer run implies degree of sickness and less likely to survive. However, when you look at the greater than 21 days, the duration of ECMA was not associated with mortality. Here, the adjusted odds ratio for respite score was 1.0 with a p-value of 0.221. You can see it crosses one. So it was not. So once again, that same message, which is...

When you're on for a long period, once you break that 21 days, the mortality really doesn't change with time. And I think this is really highlighted when you look in that blue box on the bottom right. You can see these are the percent of patients discharged. So if you're on ECMO for over 21 days, 46.5% of patients were discharged alive. I think that's amazing. If you go to greater than 50 days, 47.5%, even higher percent. And over 120 days, it's 46%, really unchanged. And I find that phenomenal, that you're on ECMO for, what is that, like four months? Yeah, four months, and you have almost a 50% chance of being discharged alive. Really important message, I think.

But where do these patients go? What happens to them? So you have a supplemental table. They're eating table three, which is where they discharged. And about 27.5% of them were discharged straight home from the hospital. Another 22.8% to an LTAC. And that usually means to me that they're weaning from the vent, that they're awake, that they're participating with care, that they have a favorable prognosis. A handful went to rehab. Some probably went to another hospital looking for a transplant evaluation. And then there's a very small missing data that we don't know where they went. When we talk about lung transplant, unfortunately, it didn't really enter as a variable, a mandated variable in the ELSA registry until 2016. So that limits a little bit of the analysis. But all cases that we found were about 2.3%.

3% of them had lung transplant. Those that were on for over 21 days had 4.4% had transplant. And those over 50 days, 10.9% went on to have lung transplant. Now, not included in that is what happened after lung transplant. And how did those patients fare 30 days after transplant? It's not necessarily synonymous with survival, right?

So now this is looking at the outcome of the patient's mortality, the patient's based on year. So really unfortunate, not like that also registry. When we look at the year periods, 2012, 2016 and onwards, there's no change in mortality that the PBI was 0.95. So maybe that means we're not getting better. I have another hypothesis, which we'll talk about later, but there was no change in discharge alive or deceased. So it did not matter based on time period. And that was confirmed. If we remove the COVID group, still there was no change in mortality. And if we just looked at the greater than 50 day, also no change in mortality based on time period. I really liked this graph and I'm going to try to kind of explain it to you. So this is looking at stack probability plots based on time period. So each of the time periods is on the top. You can see what it's showing you is the predicted.

the probability of each of these outcomes, whether to stay on ECLS, whether to be recovered, or whether you are expected to die. And what you see is a couple points. Number one, if we look all the way to the right, kind of the amount that's blue is more than the other two. So time discontinuation increased in the most recent group supported by COVID. We've talked about that. The survival of the prolonged group has not changed between the groups. If I kind of look at the trajectory of that between the orange and the red line, it seems to stabilize about the same point. And mortality really seems to stabilize after the initial 21 days, like we've been saying. So that all makes sense. I think that's a really nice way to kind of depict and look at that.

Now, how about case volume? So case volume between that earlier group, 2012 to 2017, was inversely related to the prolonged support. So what does that mean? That per 10 cases, the rate of prolonged support decreased by 4%. So the more ECMO you do, the less likely you're going to do prolonged ECMO.

stay true when we adjust it for the rest of the score. And once again, I'll kind of get into why maybe that is later. Now, how about mortality? So this is looking at if you do more ECMO in that reference group to 2012 to 2017, what is the mortality look like in the prolonged group? And what we found was that with the unadjusted analysis, that if you did more ECMO in the reference group, your patients in the prolonged group did better. A non-ratio 0.96 did not cross the confidence interval, doesn't cross one. P-value is less than 0.05. When you adjust for rest, that's not true. But still, I like to believe that message. I think if we do more ECMO, we're likely to be able to handle prolonged ECMO better. So I like that message, and I think that makes sense to me clinically.

So that's kind of all of the results from the paper. Now, this is a little bit something different that I'm going to add, that this was an analysis we did. So we did this in response. And when we published the paper, we had several people email us saying, how about the complication rate? I get that you're not having fatal complication, but how many go acute kidney injury? How many have hemorrhage? Whatever. And so we went back and looked at the data scene and tried to analyze what's the percent chance of complication based on ectomotoration. So what these categories are is based on the ELSO registry. And I really recommend when you're interpreting this to go to the registry and see what it means. Gastrointestinal hemorrhage has a very specific definition in terms of interventions, in terms of transfuse, amount of time period, et cetera. So that's an important.

to know when you're analyzing and looking at what this means. And what we found is that every single complication increases with time on ECMO. So all of these things happen more frequently when you're on ECMO for a longer period of time, except for neurological and limb. And that makes sense to me. First off, I believe both limb complications and neurological complications are likely to happen earlier on the ECMO course. And when they happen, are usually going to be catastrophic. You don't really have a major CNS hemorrhage and then go on to stay on ECMO for 30 or 40 or 50 more days. And so that is why they did not increase with time.

The limb you'll see is the lowest complication. I think that would be very different if we're talking about VA population or those with arterial access. The most common complication was mechanical, and you'll note that circuit change is included in that. And I can argue that maybe circuit change isn't really a complication, but just a necessary aspect of caring for these patients on ECMO. It is a step to maintain the circuit, not necessarily a complication of it. And when we look at the percent of patients with at least one complication, in less than 21 days, 54.6% had one complication. In the 21 to 50 days, three quarters had a complication, almost 90% in over 50 days. I removed.

The mechanical, as I said, because I don't know if that's necessarily complication, and 71.7% of patients in the over 50-day group had at least one complication. So what does this mean? This to me means if you're on ECMO for a long period, expect complications. And I think it's super important to keep in context that the mortality is still relatively low, in my opinion, for this group. So even though complications are happening, it's not necessarily affecting the overall outcomes of these patients. We also divide complications per 100 ECMO days to look at the rate, because the statement of saying if you're on longer, you're more likely to have a complication is kind of analogous to saying if you shoot more free throws, you're more likely to hit one. It's obvious. And so we looked at complications per 100 ECMO days.

In the less than 21-day group, it was 9.9 complications per 100 ECMO. That decreased to 5, and decreased to 2.6 in over 50 days. And that declining complication rate, to me, is truly remarkable. And I should admit, like I said before, I understand the concept of time-related incidence or mortality bias, and that if you were to have a catastrophic CNS hemorrhage, intraventricular hemorrhage on day 4, you're unlikely to need a circuit change on day 50. So there's certainly a component of that here. But to me, what this says is if a patient is tolerating ECMO, then they're increasingly likely to continue to do so. So give them time.

So what do we take away from all this? Number one, the use of prolonged ECMO is common. 30% of patients for ARDS require prolonged ECMO, and that stays true even with the extra longer, greater than 50 days. And the duration of support of this group is not associated with mortality, and so duration of support or inability to recover should not be the sole reason that you take someone off of DV. Less important, but still worth mentioning, is that mortality rate has not changed significantly in the prolonged group over time. And, you know, I would like to think that it did. And I would like to say, oh, we're getting better at supporting these patients. I think the reality is we're probably putting sicker patients on. And I think that there is a component of as the number of ECMO centers increases, there's some inexperience that contributes here as well. But I do think we're probably getting better at what we do. And finally, ECMO discontinuation due to ECMO-related complications is really quite rare.

A couple of other points that came up that's worth just stating one more time is that there's an inverse relationship between case volume and prolonged ECMO use and that the mortality is decreased in more experienced centers. Case volume, that's what that says. And then there's a couple of kind of, I think, interesting COVID pandemic lessons. And that not in this study, but we published a different registry study looking at the outcomes for BV ECMO in patients with COVID ARDS versus those in other viral ARDS. COVID ARDS is much worse than in influenza or other viral ARDS, induced ARDS. And that a higher percent of patients received prolonged ECMO during the COVID pandemic. And despite this, the rate of discontinuation, survival, and BV ECMO related complications were really unchanged in the prolonged group. So what that tells me,

is that even though COVID has worse outcome, once you hit the prolonged group, it probably no longer matters. And then finally, anticipate complications because they will come, and it's how you respond to them that's important. In summary, I would say that prognosis is probably less related to pre-ECMO prognostic factors. It's not necessarily related to the severity of the initial areas in sold. It's not necessarily related to the complications related to ECMO support, but really in this prolonged group, it's related to our ability to just allow for gradual recovery and respond to the clinical changes that occur. So patients that survive the initial inflammatory cascade of ARDS without clear contraindications to continue support, they probably will benefit from prolonged support, and we need to allow them to have respiratory recovery.

So where do we go from here? I think we need to understand the factors associated with survival in this group. There's probably aspects that occur or clinical situations that occur that infer bad outcomes versus those. We need to dive deeper into that. And I really would love to know longer term quality of life and functional status. Are we creating chronically critically ill patients that are debilitated? Or what does their life look like post long run ECMO? Third, I would say that we need to focus on the type of complications and how to prevent them. And finally, when do we transplant? I think this muddies the water a little bit. Who should go on to transplant and who should we continue to support?

So I started with a story. I'm going to end with a story. So this is a 59-year-old gentleman. He was a former ICU nurse at Cooper with us. He then left nursing and became a master woodworker making furniture and had a little shop and then went back to nursing during the COVID pandemic because of the need and was a nursing home manager during that nighttime nursing home manager. He got COVID complicated by MRSA pneumonia. He came in with ARDS. He was cannulated for ECMO. He had, of course, complicated by pulmonary hemorrhage. You can see the...

blood casts we pulled from his lung using cryotherapy. He had a hemothorax. He had a bronchopleural fistula. He had necrotizing pneumonia that needed lung resection. He got trachs. He got pigs. He got...

discharged to an LTAC. And since then, he has fully recovered. This is him on the left with his daughter cycling through Amsterdam and enjoying all that that has to offer. And then he and I do an annual triathlon on the shore to kind of celebrate his recovery and just how wonderful he's doing despite everything he's been through. So with that, I want to say thank you. Here's a couple of our team members, my boss on the right, me there in the middle. And that gentleman, the police officer, is our first VV ECMO ARDS patient that's fully recovered, gone back to work, and has received promotions for everything he's doing. So with that, I'll take any questions.

Beyond Hope Project with Jason Tharp Beyond Hope Project Welcome to 'Beyond Hope Project,' a beacon of inspiration in the podcast world. Join Jason Tharp as we embark on a transformative journey, exploring stories that awaken the will to start anew and find the strength to rise from setbacks. 'Beyond Hope Project' is about lighting the way to your starting line, not dictating your path. We delve into narratives that foster optimism and belief in a future where dreams become achievements. Each episode is a tapestry of positivity, enthusiasm, and comforting support, resonating with the essence of hope. This podcast goes beyond storytelling; it's a movement of positive change rooted in authenticity and transparency. We provide real tools for real journeys, infusing more hope and joy into the world. 'Beyond Hope Project' stands as a testament to the transformative power of hope and the limitless potential of the human spirit. Join us in transcending boundaries and breaking barriers. Here, we prove dai Veteran Chat Project Kyle Turner Our goal is to reach as many Veterans as we can. We noticed a lot of the podcast personalities tend to be unrelatable to the every day joe. Not taking anything away from anyone just hoping to give more people a place to feel connected to something. KZ and Kyle are Army vets who deployed together for OIF 07-09 and OND. Now close to 15 years later we are finding a way to open up about our issues, and find a way to build a community. Paramedic Philanthropy Project Melissa Ade Podcast by Melissa Ade Bigfoot Information Project Podcast Bigfoot Information Project Occasional conversations with leading researchers and investigators regarding the animal commonly referred to as bigfoot.

Frequently asked questions

How long is this episode of Maryland CC Project?

This episode is 43 minutes long.

When was this Maryland CC Project episode published?

This episode was published on May 31, 2025.

Is there a transcript available for this episode?

Yes, a full transcript is available for this episode. You can read the complete transcript on the episode page.

Can I download this Maryland CC Project episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.