Penn Medicine Anesthesiologist, Dr. Asad Usman, reviews how management strategies were both developed and refined during COVID-19. He will also speak about how these strategies have the ability to be applied to non-COVID populations and where he sees areas for future progress among various patient populations.
@PennMDForum I'd like to thank everyone for having me here. Um My talk is gonna be about Pen lung rescue experience and translating the lessons from COVID. I wanted to uh review the management strategies developed and refined during the COVID pandemic and how they can be applied to non COVID populations and really talk about how we uh managed and navigated some of the data that was just recently presented about COVID ECMO and about E C O in general. And then also take into consideration what Jack had spoken about with regards to resources and how we made decisions uh by the seat of our pants. Um And then I wanted to describe areas of future progress uh in certain patient populations. Uh Those are my uh I don't have any disclosures and my funding. So as Jack had put up, um I actually came to Penn specifically to work with Jack. Um I was at the University of Michigan and I worked in Doctor Robert Bartlett's lab. As Doctor Frasier had mentioned, he's like the founding father of E O and um and I, I still keep in contact with him every once in a while and he sort of mentioned that I should go and take a look at what they're doing at Penn. Um When I came here as a fellow, I would go out with these two guys, Jack one time, pick me up in his, in his car and we were driving and I was very nervous. Um uh it was one of our first ations and then when I was a cardiac anesthesia fellow, um towards the end of my fellowship, you know, the pandemic had started and I saw what was going on in, in terms of how these two guys were managing all these ECMO patients. And I knew that, um you know, they, that I was thinking in my mind that the uh CAL cal uh cavalry support is coming. So, um when I graduated, uh I started working um with these guys and then Audrey soon uh after joined and I still remember the things that I'm gonna talk about here. Uh The four of us would get on the phone pretty much 5 to 10 times a day. Um uh And, and Emily as well and we would um decide in very, very important decisions as a group um that, you know, I had never made in my life before, like such as one of the first patients Jack had asked me to take care of, we had to take her off of ECMO. It was one of the patients I had cannulated and um Doctor Bermudez was uh evaluating her as well for lung transplantation and her lungs had sensitized so we could not transplant her. And she had been on ECMO for more than eight weeks. But this is before we had figured out some of the, the strategies that can help these patients. And I had to talk to that family. I think about that patient. Um you know, all the time about having to take her off of um and what that did for her family, she had two kids and it is a very, very challenging time. So the time I spent with these guys, Jack and Bill and Audrey and Jack. Uh and uh Emily is very special to me. So I want to talk about mechanical circulatory support and sort of put ECMO uh in the frame and context of where it lies. We have a number of different tools that can support the lungs that can support the heart that can do subs uh segmental aspects of uh lung support such as uh CO2 removal as well as pumping function of the heart. So the key aspects that I want to talk about are looking at what happened during COVID. And some of the key decisions that we made regarding tracheostomy, chest tube placements, ventilator settings, lung mechanics, pron patients using a separate device called the hemo lung. And what we did with our anti coagulation strategies. So, Tracheostomy, so we know that patients who have tracheostomy have a tendency to bleed. Patients who are in ECMO have a very high likelihood of bleeding when they uh require tracheostomy. This is typically due to venous congestion. It's very unlikely that they have arterial bleeding that's going on. We have large cannula in the head, uh vessels that obstruct the flow and they uh they tend to have swelling in their head and that can result in bleeding. So these patients then subsequently, if they bleed, get increased transfusions, we typically will try to hold their anti coagulation, which is challenging because you're on an extracorporeal circuit. And then if they do bleed, they'll form clot in their airway. Um And the, but there are benefits of performing a trache while on ECMO because it allows for early mobilization and rehabilitation. It allows for a faster uh weaning of acma getting into physical therapy um and allowing us to communicate with patients assessing their neurologic function to make sure they're not having a stroke. Um And then there is some question if they have improved survival. Um One of the things we call it, uh some of us call it the Gucci trache. Some of us call it the Verni two step trike. I it's, it's, it's a debate amongst the group as who actually came up with this idea. But basically, what we started doing is doing a two stage trike. And so what we would first do is we would have our surgeons come and they would dissect the skin all the way down to the trachea. Ok. That uh way they had the airway itself protected. And then two days later, they would come back and complete the trache. So if they bled, they would not bleed into the airway. And this tremendously reduced the amount of airway bleeding and compromise that we had. Um, there's one of the fellows that is actually looking at this. Currently, in addition, uh what did we do with chest tubes? So for patients who get chest tubes again, it's the same thing. Instead of putting the uh you know, the source of bleeding in, in the trache in the airway, now you're causing you bleeding into the in the lung. So those patients, if they bleed, increase risk of sensitization, you have to stop the the anti coagulation, which results in the amo circuit going down. Um There's an increased risk of kidney injury and then they would require multiple trips in the operating room. And then it would make it extremely challenging. I mean, we have the world's best lung transplanting surgeon. I watch him every day and I have hundreds of videos of him doing lung transplantation, but you know him going in and trying to clean these lungs out and then do a transplant on top of that is extremely challenging and very risky. So chest tubes became a very, very important decision in how we would manage these patients. So this is work done by one of our fellows that was presented at the SI O meeting last year. Um And this uh demonstrated that chest tubes and COVID-19. This was a title um called Chest Tubes and COVID-19 patients on Extra corporal memory and oxygenation Time to revisit our approach. And as you can see here, um we had three different approaches. One was the surgical chest tube that we had our C T surgery colleagues placed at the bedside for, you know, pneumothoraces and plural of fusions. We had a very high risk of bleeding, 55%. Then we looked at percutaneous placements. We did 20 of those seven of those blood, um that was 35%. And then we decided, you know, why don't we wait? Some of these pneumo authorities were trapped air and we realized patients were not having hemodynamic compromise or oxygenation compromising. And so we then even sometimes waited through the weekend and got I R to, to manage these and we really drastically reduced the amount of bleeding and um surgery, uh you know, explorations that required um in the chest which then subsequently resulted in improved survival. And, and, and to mention, we actually published this work in J T CBS. This is one of the works that um we used the, the database of Doctor Frazier. Um Jack and I published in J T CV S um looking at the patients who did get chest tubes actually had increased risk of mortality and it was a surrogate marker of increased risk of uh um death while on ECMO ventilator settings. So this is a patient that we were consulted for, for ECMO evaluation. And we've seen some of these x-rays before they all have new authorities. There's air consolidation, you can see um air in and around the, the apex and and around the heart and then after 36 hours of lung rest and low tidal volume ventilation, those new authorities improve and and we can um then start recovering the lungs. So what was our preferred strategy? Um We went to ultra low stretch strategies using a inspiratory plateau pressure less than 25. Whereas the acceptable range is 30 C M H2O, we had a peep around 10 to 10 to 15 that we typically used, we really uh dropped the respiratory rate and the reason we did this is there's a new concept called mechanical power, which is a composite score of, of taking a look at the respiratory ray and the peep and the plateau pressures and then trying to, to manage these patients. So our preference was to use pressure control which allowed for us to avoid um you know, injury to the lungs. In addition, um we titrated our sedation because we found that some of these patients were very compliant uh in their lungs and they would generate title volumes of 809 100 CCS um which would then subsequently result in injury to the lungs which goes into lung mechanics. So, um like I said, there was a, a gain who was uh that picture of him was up there earlier, uh wrote a paper describing it. It was a theoretical paper describing two different types of, of A R DS and COVID A R DS. There was a very high compliant, low recruit, low recruit ability, lung or the H type, which was low compliance and high recruit ability, lung. And so we demonstrated early on that these patients if you tried to quickly wean them off of ECMO. Um, as Corey was saying, um, the originally with A R DS, we typically would have a seven day run and they would come off ECMO. So in the first wave, we really were like, ok, we put them on ACL for seven days, we start waking them up and they'll get better, but these lungs were highly compliant and then at day seven, they would completely bloom their A R DS again. So we had to come up with a sort of uh unique sedation strategies. Um Audrey Jack and I uh wrote a paper in an anesthesiology journal J CBS regarding using alternative agents such as methadone, gabapentin, lidocaine, inhaled lidocaine, um, you know, uh using, um, Precedex and we, we have over 30 medications that we have used to try to get, um, acma patients awake and uh recovered. Uh, so using those sedation mechanisms, we can try to avoid the P silly and then their, their uh their lungs would typically improve uh pron which has been demonstrated as we know, um prone poi. This is a paper in the New England Journal of Medicine is a classical paper by the pro study group that demonstrated that Pronin is beneficial in typical A R DS lung pathology. But um prior to the COVID pandemic, we once a patient got on ECMO at pen, our culture and our V IC U was not to prone patients. So um we had resource limitations. But however, when we looked at COVID, we had no other options. Even we went on ECMO and they were still hypoxic. So we, we were throwing in uh everything to try to help these patients out. And we found out that two things, one that Pronin really helped these patients. Second, we also found out that if you prone patients, let's say 30 days into their acma run 40 days into their ACMA run, they still improve. That is something new um that we have not really experienced prior to COVID. Um And here's a paper that demonstrated that there was an improved survival uh with lower mortality rates uh for patients who are prone to ACMA. We also use the hemo lung, which is a Co2 dialysis machine. Um We found that patients after their ECMO run would have CO2 clearance problems. And so once their oxygenation improved and we were ready to Decca, we would bridge to a hemo lung device to help CO2 removal. Because typically their Perma was so damaged, they were unable to remove uh the CO2 which, you know, uh is more difficult for the lung. Whereas oxygenation was uh preserved um in the, the 1st 10 patients uh that we had with COVID ECMO. Um we actually had 34 patients that had devastating. The first four out of the 1st 10 had devastating in intercranial hemorrhage. They, they ended up dying because of bleeding. One of them had a cran toy from this and we wrote this up, this experience and we know that COVID has a high risk of coagulopathy and has uh perturbations and alterations in their uh coagulation cascade. Um And so basically, one of the strategies we did was to switch to Velar Rodin. And we showed that Velar Rodin for ECMO, decreased um mortality, uh decreased, the major risk of bleeding events, decreased, ECMO, thrombotic complications and increased the time and therapeutic range. Um This has been demonstrated in publications outside of the COVID ECMO world. But in COVID, ECMO, we also started using bivalve and it is our current strategy for V V ECMO and we're considering it for the rest of our V V Acma population. So, using these six or seven strategies, what were our outcomes at pen? Um As you can see the, on the left, there is the S O database. Um looking at their mortality, we were able to beat the odds or the expected odds. Um According to the, uh if you use the Els O uh registry as a, as a standard, and we got our uh survival or our mortality down to 30% which is overall so high, but much better than when we started, which was 48%. So we had an 18% decrease in, in mortality just in the ECMO population. The one of the most pivotal things that we introduced is right ventricular cyst devices. Um Typically, we uh put patients on V V ECMO. Um But we had, we know that there is increased risk of R V dysfunction in A R DS. These patients have hypoxia, hypercapnia and increased ventilation, uh ventilatory settings which can be fixed by ACMA. But things that are not fixed by acma that impact the R V are vascular Alternations, edema and Micromuse. And so, up to 10-40 of patients who are uh on acma with A R DS manifest some degree of R V failure. Um And so the question then becomes, what is the ideal cannulation strategy for these patients. So um halfway through the pandemic, we uh saw reports from Chicago and I, I trained with Antoine Tito, who wrote the paper that looked at using right ventricular assist devices in patients with A R DS, uh in particular in COVID A R DS. And so they put this paper out um uh with Scott Silveri and uh Antoine from Chicago and from Florida that uh looked at a strategy comparing V V ECMO in COVID-19 versus uh V P, ECMO, Vino Pulmonary ECMO. And they showed about a 19% improval in survival um between the two groups. So we also used that strategy. We had a number of different um cannulas that we used um in in these patients. And uh these, we use dual cannulas, two stage Cannulas um and two separate Cannulas to help these patients. And we noticed that these patients had much better outcomes than our previous cohort of V V E patients as well. And that also contributed to our improved survival. So today, we have done 52 R ads and that has expanded the, the role of the right ventricular cyst support device, not only in COVID A R DS, but also in non COVID COVID A R DS. And we have used these strategies in um in ad surgeries for some of Doctor Bermudez cases as well as post heart transplantation and postcardiotomy uh shock. Um And so we have uh sort of a a diagram and a uh decision tree as to how we decide which Cand strategy we use. And basically the takeaways are that while challenging COVID provided an extraordinary learning environment, it was sort of a live in in life scenario and um experience for us to use these strategies and improve the way we can care for patients. Um And there's a lot of momentum that came out of that and I'd like to thank, uh, Jack and Bill and Audrey and Emily for, uh, being some of the best partners that I have.