Good morning everyone. Thank you for being here and thank you to support our presenters. We'll get started on our early morning session and I hope everybody has had their coffee. The session is about advancements in ECMO and we have excellent presentations in store. We have five presentations and presentations will be seven to eight minutes and we'll take two to three minutes for questions after each presentation. We'll start with Dr. Wong who will be talking about predicting mortality in ECMO patients. All right. Good morning everyone. Let me get this loaded. So my name is Beverly Wong. I'm a pulmonary critical care fellow at Tufts Medical Center in Boston and I have no financial disclosures. So today I'm going to be talking about the results from our retrospective chart review on patients with severe ARDS requiring ECMO. This project started after we saw significant rise in the patients transferred in for ECMO evaluation but the decision to initiate ECMO is institution specific. Over the years there's been several mortality scores created to help in this decision and the studies found that actually survival was correlating mostly with extra pulmonary factors because deaths were usually related to complications from ECMO or multi-organ failure except for maybe mechanical ventilation duration. And the other respiratory variables were just not documented enough to study retrospectively. But recently there were two other studies that came out in the ECMO COVID population that did show that pulmonary variables do matter for ARDS and this included things like driving pressure. So for our project we wanted to use another marker, respiratory marker that's been shown to be associated with mortality and this is the venatory ratio. It's defined by the patient's minute ventilation multiplied by their PaCO2 divided by the constant values of their predicted minute ventilation multiplied by the ideal PaCO2. So it's a unitless value. The value of 1 is normal. Greater than 1 represents increased pulmonary dead space. And a value of greater than or equal to 2 was shown in a post hoc analysis of the ROSE trial to be associated with mortality. So VR hasn't been looked at in the ECMO population except for one that looked at it with relation to ECMO decannulation but not prior to its initiation. So for our study we started with 64 patients with respiratory failure on ECMO from January 2017 to September 2022. We excluded 9 people who didn't have ARDS as their primary ECMO indication and then another 3 as that was their second run but we included their first runs. So we did full chart reviews on 52 patients and then excluded another 8 for missing data and then we analyzed a total of 44 patients. We abstracted baseline demographics, comorbidities including the Alex Hauser Comorbidity Index, the severity scores using the SOFA and the REST score, and the values were gathered within 24 hours prior to ECMO initiation except for the GCS score for intubation. And then the respiratory data we started with the gas also within 24 hours of ECMO initiation with the highest PaCO2 and then we used the correlating vent setting and the time of that gas and then used the exhale tidal volume for the VR calculation. So here are our patient baseline characteristics. We had 31 survivors to discharge and 13 who died in the hospital. Our survivors were a little younger, 39 of mean age versus 47 in non-survivors. They had a similar comorbidities and the unequal Alex Hauser score but the incidences of these comorbidities listed were a little lower in the survivors. There is no difference in the SOFA score or the REST score and then we had about two-thirds of patients who were transferred in from a community center to test for ECMO evaluation and this was similar in both groups. For our respiratory results, we had similar duration of mechanical ventilation prior to ECMO at two to three days but we had a slightly higher proportion of survivors who were started on ECMO within 24 hours of intubation and this was 42 versus 23%. COVID was a predominant cause of ARDS in our group with 45% of survivors and 69% of non-survivors and they all had severe ARDS with P to Fs in the 60s, 70s and then there were similar rates of proning, paralysis, inhaled vasodilator use and then low tidal volume strategy at six to seven cc's per kilo in both groups. We had a slightly higher minute ventilation in the non-survivor group at 12 liters versus 10 but this evened out to a similar ventilatory ratio that was not significantly different between the two groups. So to recap again, our survivors are as expected, they're younger with fewer comorbidities and COVID was a predominant cause of ARDS in our cohort. There was no significant difference in ARDS management and the REST score was not significantly different but it was a little higher in the survivors which is on par with the original study where higher score is a higher survival. And our study has an overall mortality of 30% and our main outcome of interest which was the ventilatory ratio was not different between the survivors and non-survivors at 2.7 versus 3.0 but the difference came up in multivariate analysis and we found that every increase in the unit of VR was associated with increase in hospital mortality with an ARDS ratio of 2.65 and we adjusted for gender, initial admission to an academic versus community hospital, COVID, Alex Hauser, SOFA and the REST score. And then if we reran for mortality for a binary cutoff of VR of less than two versus greater than or equal to two, the ARDS ratio jumped up even further to 10.1 but this was not statistically significant at a P of .054. We hope that this suggests that it's just a little underpowered but we do have several limitations. One is that we did obtain VR a little differently than the original study and we allowed for some spontaneous tidal volumes to be obtained when the patient was clearly desynchronous from the event but this was not very common or frequent as the overall low tidal volume strategy was appropriate in most of our patients. The biggest limitation is of course it's a retrospective chart review. We had a lot of missing data and a lot of things that we wanted to look at as well including driving pressure and VR trajectory over time but we didn't have enough data points to analyze this. And so we hope to see that there's more coming out about VR because I think our study suggests that this could be a very useful clinical tool to help guide in this kind of decision for this patient population. It's very easy to get. All you need is a minute ventilation and then a gas. Thank you. Next we have Dr. Ho. He'll be talking about utilization of ECMO during the 2020 first year of COVID pandemic. Yep. All right. So good morning and welcome to the last day of CHESS. Thank you for coming to the talk. So today I'm going to talk about the national and regional trends in the ECMO use during the first year of pandemic. So interestingly a couple days ago a Nobel Prize in medicine was announced and it went to the two scientists who pretty much invented the COVID vaccine and I think that's important because it's really relevant to my study today. So my name is Cam. I'm one of the third year fellows at University of Maryland. I'm also an NHLBI track. I have nothing to disclose today. So these are learning objectives for this talk. So in 2020 there are about 1.7 million hospitalizations for COVID-19. Among the patients who were admitted to the hospital the mortality was about 13% and this is unadjusted. And among those patients who were requiring mechanical ventilation the mortality jumped to 55%. And as we know the pathophysiology of ARDS in COVID-19 leads to disruption of the lung epithelial membrane leading to increased permeability to fluids and proteins. So one of the end results is inability to have gas exchange and leading to refractory hypoxemia. So one way to overcome this is with ECMO where you have the exogenous oxygenator providing oxygen to the body thereby bypassing the lungs. The evidence for COVID-19 ECMO use comes from several clinical trials as well as society guidelines. So we have the CSER trial and the ELO trial which provides, although they are pre-COVID error, they do have some suggested effectiveness in ARDS. And then in terms of society guidelines, what that really boils down to is among patients with refractory disease or refractory hypoxemia due to failing conventional medical treatment, those are the people we should think about ECMO. So as highlighted by the ELSO guidelines, that includes a PF ratio less than 150. Those who fail proning, inhaled vasodilators, as well as neuromuscular blockade. And there's no contraindications to ECMO at that time. Those patients should be considered putting on ECMO. So based on the last talk, we already heard that this initiation is really based on different centers. So the challenges with ECMO is that it is a very intensive procedure and it requires a lot of resources. So including profusionists, all the staff, transportations, ICU care, as well as the ethic consults. So we know that in 2020, we were already in such a health-limited situation. So my study is actually to look at how we actually utilize ECMO uptake during that time. It is out of scope in my study to talk about effecting ECMO because we don't have the granular data. And we're not trying to talk about when to initiate ECMO for these patients. So the database I utilized was the National Inpatient Sample Database, which is the largest all-pair inpatient healthcare database. The inclusion criteria includes everybody who were admitted with COVID-19 as a primary diagnosis during that year. Unfortunately, the database does not link to a different year, so you can't really say Somebody who was admitted in December was also discharged in January of the next year. The limitations of this study, just to highlight some of the things, includes there's no vital signs data, no laboratory data, or medication data. So that's one of the drawbacks of using this database. In terms of our overall findings, these are the tables for the patient characteristics broken down by overall national and regional. So overall, to no surprise, most of the patients admitted who were initiated on ECMO were falling between 40 to 60, about 60% of them falling between that age category. Less than 50% of patients were over the age of 60. Predominantly, there were also 70% were male, and that could be due to the severe disease in male patients, as well as the higher amount of comorbid conditions, such as hypertension, hyperlipidemia, and diabetes. There's also a racial ethnicity discrepancy. Surprisingly, there's a lot of, you know, 35% were white and 34% were Hispanic, but in the West region, you know, there were around 60% who were categorized as Hispanic patients. Majority of them also had Medicaid, Medicare, and 10% were uninsured, and that can have some implications for healthcare resources. To no surprise, these are the breakdown for comorbid conditions. You know, the most common ones were also hypertension, hyperlipidemia, and diabetes, and we know as now that these are the risk factors for severity in COVID-19. In terms of the hospital characteristics, majority of them identify as a large hospital, and that means they have more than 500 beds, medium falling between 200 and to 500. So those are the majority of the, you know, ECMO cases where they were admitted. You know, majority of them were also in the city area, and they identified as teaching hospitals. The average length of stay for someone who were admitted for COVID-19 on ECMO use is around 35 days. That includes the patients who survived and didn't survive, too. Surprisingly to us, you know, standard deviation is quite low, which is one to two days. The cost, on the other hand, was quite high. It was about, you know, a quarter of a million for this hospital admission, and there are a lot of complications, but based on ICD-10 coding, most common ones were sepsis, renal replacement therapy, and VTE. Overall, our findings, we adjusted for age in terms of mortality, and, you know, comparing the difference between COVID-19 for those who were using mechanical ventilation and compared those with ECMO. So, for those who require mechanical ventilation, the adjusted mortality is around 37%, compared to the 55%, which is cited from other paper. And in terms of COVID with ECMO use, that was about 45%. So essentially, half the people who were placed on ECMO eventually pass away, unfortunately. This is a graph showing the cases of COVID admissions, as well as cases of COVID ECMO initiation over time from January to December. So on the left side of the graph, you have the absolute numbers for COVID admissions. On the right side, you have the absolute numbers for ECMO use. So just focusing on the COVID cases, we kind of break it down to urban and suburban hospitals. What we see over time is there are a couple of peaks during that year, specifically around April, July, and then a resurgence around December. And at that time, this was the era before vaccinations, effective therapy, and the only thing we had was the dexamethasone trial that came out around June. Surprisingly to us, we see the ECMO cases were highest initially around April, and then it gradually decreased over time. And we speculated that because this is probably due to the different variants in terms of the strains that came out. I mean, the strains came out from like alpha, beta, delta, omicron, and there are a lot of strains that are still coming out today. The second thing is, as I have highlighted, the dexamethasone trial came out in June, and which showed a decreased mortality from 25% to 20% in the UK among patients who are using dexamethasone. And then over time, we also speculated that systems kind of adapted to the utilization of like their health resources for COVID, and some were preparing better over time and managing patients better over time, possibly leading to the decreased of ECMO use. So we were also able to kind of divide up the cases by regional, and according to the database, they're divided into four regions. So what's interesting, I know there's a lot of graphs here. For me, it's the incidence kind of like circling around the US and rotating clockwise. So if we start with the northeast on the top right, we see that the highest incidence then in like around the New York states at that time was around April. And then the COVID kind of cases kind of spread around the country from there, going to the southern states, and then to the west coast after that. Surprisingly, the Midwest state never really had, you know, much of a peak in the early of the year until November or December. As we see here, there are some similarities between these graphs and the national ones too, where the overall trend kind of surges, you know, over time. And the cases for ECMO actually decreased over time. So these are just similar things that I wanted to say. So I think overall from this study, we can't say, you know, whether ECMO is effective or not, but we do see a very high incidence. Around 45% of patients who were initiated in ECMO eventually passed away. The total cost for ECMO during that year was around 900 million, and we see about 10% of patients were uninsured. And we're not totally sure how, you know, hospitals, health system can recover that. You know, the hospitals were all under a lot of stress. At the same time, we have a shortage of, you know, nursing care, ICU docs. You know, it's really, it'll be interesting to do a follow-up in terms of my end. I would like to do an economic health analysis on like, you know, whether it's, you know, health effectiveness on like patient, placing patients on ECMO, and what the downstream effects are. But I think there's more to come on this study. So I think that's pretty much it, and these are my references. Thank you. Thank you, Dr. Ho for the excellent review. Actually, I'm happy you had much better survival than we did. But yeah, we're happy to take questions. Any questions? Yeah. Thank you so much, once again, yeah. Next we have Dr. Altaie. He is here. Yes. He'll be talking about outcomes in ECMO during the COVID pandemic. Good morning, everyone. Sorry. Good morning, everyone. Thank you so much for joining our talk here today. So today I'm going to be talking about the outcomes of the ECMO in patients with COVID-19 versus patients who were admitted with ARDS from other causes. We utilized the national inpatient sample from the year 2020. So I'm Mustafa. I'm a third-year resident from Montefiore Medical Center, and I have nothing to disclose. So the objectives of our talk today is to investigate the utilization of the ECMO in patients with COVID-19 versus patients with ARDS from other causes and to assess whether the ECMO is associated with lower mortality in these patients. So the ECMO has been used to treat patients with ARDS and it has been used during the year 2020 and 2024 for the patients with COVID. But the effectiveness or the efficacy of the ECMO in this population has not been well established. So we utilize the National Inpatient Sample Database from the year 2020, which represents all the hospitalizations in the United States that year. And we looked at the patients who were admitted with ARDS and we stratified these patients into those who had diagnosis of COVID versus other causes. And then we compared their baseline characteristics and we looked at the patients who had ECMO placement and then we performed multivariate regression for these patients. So the total patients that were admitted with ARDS were about 130,000 patients in the year 2020. And the majority of them were COVID patients, almost 75%, almost three quarters of the patients. And then 26% were admitted with other causes. Among the patients with COVID-19, about 2.4%, almost 2,500 patients had ECMO place as part of their management for COVID versus the other causes had 4% higher percentage and then there were almost 1,500 patients. So we looked at the baseline characteristics for the patients who had the ECMO versus the patients who were treated without the ECMO. And it looks like the ECMO patient had a median age of about 65. So they were a little bit older than the patients without ECMO. And then the majority of the patients were males, almost 70% of them. But looking at the other comorbidities, looks like the patients who got the ECMO were less sick from the patients who did not get the ECMO. Except for obesity, which was higher on the ECMO group. So by analyzing the patients using the unadjusted mortality, it looks like the patients who were treated with ECMO has a lower mortality. The odds ratio were lower in that group. So the mortality rate was 47% in the ECMO group who had COVID versus 51% who did not get the ECMO. And the non-COVID patients, same thing, also had a lower mortality on that group. And mortality rate was 33.8% versus 42. The p-value was less than .001 for both. And then after performing the multivariate analysis and looking at the comorbidities of both groups, ECMO is actually associated with higher odds of mortality in the COVID patients versus the other patients in the non-COVID group, which had a lower mortality with an odds ratio for the COVID patients of 1.4. And the confidence interval was 1.3 to 1.5 with a p-value of less than .001 versus the non-COVID patients had the odds ratio of .83, a confidence interval of .73, .95, and a p-value of .007. So in conclusion, there's a lot of limitations like Dr. Ho mentioned earlier for using the national inpatient sample because a lot of the patient characteristics are lacking using these data. But we think or we concluded that ECMO effectiveness and the COVID patients might be questionable and further research needed to look at this population and see how effective that is and if that is associated with lower or higher mortality. And I think that's all for me. Thank you. Next, we have Dr. Han to talk about proning during the VV ECMO. Good morning, everyone. My name is Susan Han. I'm assistant professor in pulmonary and critical care at Tufts Medical Center. Thank you for the opportunity to discuss our work today on the impact of proning on respiratory physiology in patients on VV ECMO with ARDS due to COVID-19 pneumonia. I have no disclosures to report. So I think we're all pretty familiar with the benefits of proning in patients with ARDS from mechanically ventilated. However, less is known about the utility of proning on patients who are also requiring VV ECMO support. Some studies have shown that proning is safe and feasible and that it is associated with the reduction in hospital mortality. However, this has not been a consistent finding across different studies and sites. Less is known about the effects of proning on lung mechanics and oxygenation, particularly in a setting of some of these ultra lung protective ventilation strategies that the ECMO patients frequently receive. And then lastly, data is even more limited for patients with COVID-19 under VV ECMO and proning. So to further investigate this, our study aimed to determine the effects of initiation of proning on lung mechanics, gas exchange, and degree of ECMO support in patients on VV ECMO for the management of ARDS secondary to COVID-19 pneumonia. This was a retrospective single center study and an academic tertiary care center of all adult patients with COVID-19 pneumonia who developed ARDS or cannulated for VV ECMO and then subsequently prone while on VV ECMO. The study period was March 2020 to February 2022. So we collected data at two time points. For the pre-proning variables, we chose the data that was closest to the initiation of proning with a limit of 24 hours prior to proning. For our post-proning variables, we looked at data that was collected approximately one hour after initiation of the first cycle of proning on VV ECMO. So this data included variables of respiratory mechanics. So as you can see here, tidal volume, tidal volume for predicted body weight, plateau pressure, PEEP, driving pressure, and FiO2. We looked at gas exchange as reflected by pH, PCO2, and PaO2. And finally, we looked at the degree of ECMO support required with the flow rates and the sweep settings. These variables defining lung mechanics, gas exchange, and ECMO support were our primary outcomes of interest. We had a fairly straightforward analysis in this study with just parity tests, and Wilcoxon banked some tests as appropriate to compare the pre- and post-proning observations for all of our patients. Now onto our results. So our final study population included 12 patients during this time period. The patients were predominantly male. As you can see here, 75% male with a mean age of about 51 years, mean BMI of 30. The mean number of days from intubation to ECMO initiation was about five, and then there was another subsequent seven days between ECMO initiation to the very first proning session. Our patients underwent a median of about 3.5 proning cycles in total during their courses of ECMO. In terms of patient outcomes at 60 days, 58% of our patients were alive at this time point with 33% still hospitalized and 25% having been discharged either home or to a rehab facility, and 42% had died in the hospital during this time period. Now onto our respiratory mechanics. So as you can see here, there was no statistically significant difference in any of these variables pre- and post-proning. You can see that these patients really were, again, on their really true lung rests, very long protective ventilation strategies with quite low tidal volumes of about two, both pre- and post. Driving pressure and plateau pressure were both appropriate. Plateau pressure was 23, 25. Driving pressure was around 12, and these patients tended to be on higher PEEP settings of 11 to 12. With regards to gas exchange, we also saw that gas exchange was fairly similar between the two groups pre- and post with relatively normal and appropriate-looking PAHs, PCO2, and PAO2 for both time points. And then lastly, looking at ECMO support, we saw that there was no significant difference in the level of support required by these patients in the pre- and post-proning time points. We saw that the flow rates were about four liters pre- and post, and that the sweep was 5.5 pre- and 5.1 post-proning. So in conclusion, based on the data that we had available to us, it seemed that the initiation of proning did not lead to any early changes, at the very least, in lung mechanics, gas exchange, or degree of ECMO support required. Our study did have multiple limitations. This was a single-center study with a very small group of mostly male patients, so generalizability is quite limited. In addition, we studied the very first proning cycle. There certainly are changes that could have manifested with multiple cycles of proning that we did not capture. Additionally, the ideal time point for reassessment of these sort of physiologic and support variables after initiation of proning is unclear. We chose one hour because this was a retrospective study, and that was really the clinical practice pattern our institution was to do a comprehensive reassessment one hour after proning. However, there may have been changes that would take several hours to develop, and that may have manifested later on after that first cycle. So in conclusion, the initiation of proning did not significantly lead to immediate impacts on respiratory mechanics, gas exchange, or the degree of ECMO support required. However, we did find that proning was safe and feasible in our study population. There were no adverse events that were related to the proning sessions in our patients. Clearly, further studies in this area are very necessary. I would say that prospective studies and randomized controlled trials in particular, one of the challenges of these retrospective studies is that it's extremely difficult to get consistent, comprehensive, granular, and longitudinal physiologic data or data about support or respiratory mechanics in these patients. So I think having predetermined variables and time points for assessment will be really critical to truly understand the benefits of these interventions. I think it would be fascinating to look at the impact of multiple proning cycles, and also to look at the ideal timing of proning initiation after maybe ECMO cannulation. As you can see, our patients underwent proning a median of about 12 or so days after they were first intubated, so relatively late in their course. And I do wonder if initiating proning earlier would lead to more physiologic response and potentially more benefit. So I want to acknowledge, particularly our fellow Katie Waybill, who really spearheaded this work and sadly couldn't be here at CHESS this year to present this, I'm presenting on her behalf. I'd also like to thank our MICI director, Anthony Fogno, and the entire large multidisciplinary ECMO team at Tufts Medical Center. So here are our references. Thank you guys for your time and your attention, and happy to take any questions. Next we have Dr. Rema talking about Impella versus ECMO in cardiogenic shock due to STEMI. Hi everyone. My name's Victoria Reimer. I'm pulmonary critical care attending at Jefferson, New Jersey. I don't have any disclosures. And I'm gonna talk about in-hospital outcomes of Impella versus ECMO in patients with STEMI complicated by cardiogenic shock, looking at a population of patients over six years. So the objectives, I'm gonna take a retrospective analysis of patients with STEMI and cardiogenic shock and try to compare the two of Impella versus ECMO, looking at things like mortality, length of stay, hospital charges in patients who received either mechanical support. So a little background. Cardiogenic shock happens in about 10% of patients after acute MI, defined by SPP less than 90 for more than 30 minutes, or requiring inotropes to maintain adequate MAP of over 65. Even with early revascularization with PCI, the mortality remains high in these patients and can approach 40% at one month and 50% at one year. So the IABP Shock 2 study led to a decline in the use of intraaortic balloon pumps, which increased the use of ECMO and Impella. ECMO provides its support by bypassing the LA and LV, resulting in non-physiologic flow, and then gas exchanges provided by the oxygenator put into circulation. ECMO can contribute to worsening LV failure due to the increases in afterload and increased diastolic pressure, leading to increased ventricular expansion. So it's kind of said that ECMO shouldn't be used alone in severe LV failure, whereas the Impella preserves physiologic flow and pumps blood directly from the LV into the aorta, resulting in a physiologic flow compared to ECMO. So methods, again, retrospective analysis using the Nationwide Inpatient Sample from 2015 to 2020. We used ICD-10 codes to identify patients with the principal diagnosis of STEMI and cardiogenic shock who received mechanical support with ECMO or Impella while hospitalized. Parameters were analyzed using propensity matching, utilizing the kernel method, and then multivariate logistic regression was used to adjust for patient demographics, hospital demographics, and then other comorbidities. The primary outcomes were inpatient mortality, length of stay, and total hospital charges. Secondary outcomes were respiratory failure, GI bleed, stroke or TIA, ICH, acute kidney injury, VTAC, and atrial fibrillation. Out of the 235,110 patients with STEMI and cardiogenic shock, 5,244 or 2.2% received Impella support, and 1,260 or 0.5% received ECMO support. So the primary outcomes, ECMO use was associated with more significant in-hospital stay mortality, increased hospital length of stay, and total inpatient hospital charges. As far as the secondary outcomes, the ECMO patients also had increased frequency of respiratory failure, respiratory failure requiring intubation, and then VTAC and AFib. Patients with Impella were found to have higher acute kidney injury as well as cardiac arrest. As far as the CVA, TIA, or ICH, no statistical significance between the two. Females had higher mortality compared to males regardless of device used. So there have been previous trials that kind of revealed similar outcomes in discussing ECMO or Impella for patients with STEMI complicated by cardiogenic shock, and our study suggests that inpatient STEMIs with cardiogenic shock, patients on ECMO had higher mortality, higher length of stay, higher hospital charges. And then again, further analysis showed that patients on Impella had higher cardiac arrest and acute kidney injury. But further randomized trials and studies are needed to compare further these different circulatory devices. So clinical implications. Our study would show that Impella would remain the gold standard for patients with STEMI complicated by cardiogenic shock as compared to ECMO. These patients are critically ill, they're gonna need early mechanical support, Impella can provide that quickly. And that's important because obviously the longer that you're delaying the cardiopulmonary support can lead to worsening infarction, necrosis, and more tissue death. So future aims. More randomized trials looking specifically at females I think would be beneficial considering that both had higher mortality regardless of the devices. And trials evaluating short-term ECMO followed by Impella, just combination may have greater benefit than either individually given that, like I said earlier, ECMO has the risk of worsening LV failure due to the increased afterload. And a combination of ECMO and Impella could potentially kind of work to lessen that. Limitations of the study. The nationwide inpatient sample, it doesn't identify individual patients and recurrent hospitalizations. So it can look like different distinct observations. And one patient can contribute to multiple entries if they're hospitalized more than once in the same study period, which over six years is likely. One patient can also have a different outcome if they were transferred at one facility and die at the next. It doesn't capture any of the outpatient visits and procedures occurring at different hospitalizations in different hospitals can be underrepresented. It also only collects 20% of the actual data. Weights are used to generate national estimates. Race category, very high missing entry rate as far as in the national inpatient sample. Retrospective study, it comes with its own limitations and it's totally dependent on the data that was entered into the clinical database. Some data can just be missing if it wasn't pre-designed format or specifically entered from the beginning, knowing that this was gonna be looked at for a study. Lack of follow-up, lack of homogeneity. And that is all I have. So if there's any questions, thank you for listening.

ECMO

mortality prediction

COVID-19 pandemic

venatory ratio

proning on respiratory physiology

Impella versus ECMO

ARDS