So, good morning, mahalo, welcome to Hawaii, welcome to CHESS 2023. My name is Kai Hsu. Barbara Stewart. We're your moderators. We're going to, I'm going to give a few minutes before we get started, you know, mainly because this session is set for six and we have five presenters. So, we have a little leeway. So, I appreciate you all attending this morning. So, while we're getting to, I'll just provide you a little information about the session. So, for the presenters, yeah, we want to make sure you, first of all, thank you for coming because unless I'm mistaken, nobody is from Hawaii in this room here. Oh, sorry, excuse me. Sorry, no presenter? Are you? No, no presenter, but those, we have one, somebody who is local and everyone else, I guess, had to travel here, right? Right? Yes. Is this your first time, first timers? Oh, excellent. Number two times, three times, four times, five times, you know, ten times, I don't know, I've been here not that many times, but if you ever go to these, while we're waiting, these introductions for the meetings and they ask you the same question, I don't know if you've ever been, ask the same questions about how long you've been here and everyone, the ten times, the people who have been here ten times, they ask you to stand up, I'll tell you, they're rich. Well, with that in mind, I think we'll get started. So, I want to ask everybody, this is our CHESS COVID-19 session, it's on COVID-19, which is still here, it's maybe not as prominent as in the past but it's adding salt to the wound, we have other things that we've identified and want to promote for all the attendees. I want all the presenters to understand that please read out your disclosure slide or either state verbally what your conflicts are or not. Make sure that we will give you a one-minute warning if once we hit the seven-minute mark because we do want to stay at least with eight minutes and two minutes. Please stay for the entire session, presenters, because then we'll maybe able to have more questions, maybe questions between abstracts. And please, those who have questions, ask them in the microphone. And so, I think that without further ado, we'll get started and I will do introduce the first speaker. Oh, sorry. And the topic of the presentation is Prognostic Biomarkers of Disease Severity for Patients Affected with Coronavirus COVID-19. Irina Timokht, if I pronounced that correctly, thank you, will be presenting and welcome. Aloha. So, the title of our presentation is Prognostic Biomarker of Disease Severity for Patients Infected with COVID-19. I receive research grants from National Institutes of Health and National Academy of Medicine. The objective of the study are to identify biomarker of COVID-19 severity and develop microbiota-based target for diagnosis of treatment for patients with severe infection. As we all know, identifying patients at risk for severe COVID-19 complication following COVID infection continues to be a significant challenge. Emerging evidence is showing that microbiota in various body sites, oropharynx, upper airway and gut may play a critical role in development of progression of severe complication after COVID-19. Therefore, we hypothesize that understanding the shift in microbial population of patients with severe COVID infection can identify pathway involved in development of severe complication after COVID-19. And this is not new. Manipulating the gut-lung axis was previously proposed as treatment for different lung diseases and previous study have demonstrated changes in the composition of gut microbiota in COVID-19 patients. Therefore, our study is trying to define this relationship that could be responsible for varying the severity of patients with COVID-19. Stool, oral and nasal sample were collected from COVID-19 patients admitted at University of Maryland Medical Center between January 2021 and June 2021. The sample were analyzed using whole community shotgun metagenomic sequence on Illumina 6000 platform. Clinical factors including the ICU status, survival and the requirement of mechanical ventilation were correlated with microbiome characteristics. We profiled the composition and structure of the microbiome. We evaluated the community alpha diversity in order to determine biomarkers significantly different by clinical features. We found that stool samples of patients requiring ICU admission, the patient that had severe COVID infection had significantly more parasuterella, odorebacter, staphylococcus and selimonas. And this is similar to other GI dysbiotic condition like inflammatory bowel disease or itch trouble bowel syndrome. As mentioned before, analysis was conducted using the phylosec R package that demonstrates a significant difference in community alpha diversity for patients that require mechanical ventilation versus patients that did not require mechanical ventilation. This heat map is showing the microbial community composition and structure for oral, stool and nasal samples. What we found that both patients with severe infection admitted to the ICU as well as patients admitted with a floor, but the one that requires supplemental oxygen, demonstrated significantly lower community diversity in both GI and oral samples. And this is showing us the difference between patients that require mechanical ventilation versus patients that did not require mechanical ventilation. And patients that were on the ventilator had significantly more odorebacter, parasuterella, staphylococcus and selimonas, as is underlined in this slide. What we also find that oral microbiome of patients with severe disease had significantly increased abundance in staphylococcus and enterococcus. In conclusion, there is a significant difference in microbial composition, diversity and richness in patients who had severe COVID infection. There is a significant difference in the gut microbiome composition in patients with different levels of COVID-19 severity. That made us conclude that there is functional dysbiosis related to varying level of COVID-19 severity. Patients with severe COVID-19 infection had lower microbiome diversity and significantly altered microbial community composition and structure. COVID-19 infection appears to pose a great influence not only on the lungs, but also systemically and on the GI tract. We can conclude that changes in gut microbiome may present an indicator of disease progression. Our study has the potential to identify biomarker related to disease progression and severity. These biomarker are involved in local immune activation and contribute to high rate of death in COVID-19 patients. Our study has the potential to develop microbiota-based targets for diagnosis and treatment in patients with severe infections. Thank you. Thank you very much. Not only are you on time, we have a few extra minutes. So I'd like those who have questions to step up to the microphone and maybe I could start and ask, you know, your timing of your collection, your SU data and all these, were you sort of use these to predict severity or were they obtained in a sequential fashion during the course of the patient hospitalization? We collected longitudinal samples. First sample was collected within 24 to 48 hours from admission. Second sample, and thereafter, there are longitudinal samples collected at one week, two weeks, three weeks, depending on how long the patient was in the hospital. Fortunately, many patients were discharged within one to two weeks, but patients that required admission in the ICU had a longer length of stay and for some we collected more samples. Please come to the microphone. And then, so you suggested there's a change in the microbiome. So I'm just curious what changed because you identified certain bacteria that were increased in frequency isolation, but, you know, what was, you know, what sort of change did you see over the time since you had more than, you know, initial, you had some follow-up samples? So there was a change in the diversity of gut population. So there was a different type of microbiome that was associated with COVID-19 severity. And what we found that there were more prevalent were the species that were significantly different was odoribacter, parasutella, semolina, and staphylococcus. And this was for the GI tract. We also collected oral sample and nasal sample. Unfortunately, we didn't have enough funding to analyze the nasal sample. We just analyzed the oral and the gut samples. And there also, there was also a difference in oral microbiota. As I mentioned, there was significantly more staphylococcus and enterococcus in the oral microbiome samples from patients that had severe infection and required mechanical ventilation. Okay. Sir. Hi. My name is Dhaval Rawal. I'm from University of Alabama at Birmingham. So similar question on the same line. The data that you presented which shows there is a difference in the microbiome, was it the same patient on the sample one over one week later and two week later? Or this was like different set of population? Meaning when the patient came in, they had a rich microbiome. And then one week into the ICU stay, they got less. And that suggests that it is changing. Or we are thinking that they already had a poor diversity to begin with and that make them at risk of developing COVID infection? That's a very good question. And this would be something that we have to evaluate in the future because it's hard when you have samples at the time of admission, it's hard to say if the change in microbiome was related to underlying conditions that predispose the patient to have a more severe COVID infection versus COVID infection creating this biosis. Right. And I guess if we have enough sample size, can we look at the same patient over the course of their ICU stay, whether it's a one week or two week, and see how it changes? Like for example, now we know that if you do MRSA swab and you say, okay, your MRSA swab is positive, that can correlate with your high likelihood of having an MRSA pneumonia while you are in the ICU. COVID is a new in that term. So we don't know what that microbiome mean and the gut microbiome is not usually tested. But if you have a, you know, some control cohort, we can kind of compare day one within 24 hours. Hopefully it hasn't changed much unless patients spend a lot of time before coming to hospital with COVID infection. So that might be useful data to look into. Definitely, definitely. We collected sample from 130 patients, which we consider that is a significant number of patients. Almost all patients had samples of admissions and a follow up sample. There are only a few patients that end up having four longitudinal samples. Because as I mentioned, fortunately some of them were discharged. During this period of time for the same patients, we also collected samples to analyze the metabolome, cytokine and the T cell profile. So our next step is to correlate the cytokine profile with the microbiome, with the T cell profile. And that will give you, give us more insight in answering your first question. What was first, the dysbiosis or the COVID-19? Thank you. I appreciate it. Along that same, I have one more question. Definitely. Was it treated? Were the staphylococcus and the enterococcus treated? The patients were started on broad spectrum antibiotics. This was at the beginning of COVID epidemics. We didn't know much about it. Patients were not vaccinated. And at that point in time, we were just trying different type of treatments. But were all patients at that point in time, just because we were not sure what to do or started on broad spectrum antibiotics. Okay. One last question. Yes. On the same line of thinking, thank you for doing this talk. So the patients who went to this ICU with COVID, you had mentioned everyone's on empiric antibiotics. Do we kind of feel like the dysbiosis is actually us and just critical care? Like if you did, someone who wasn't COVID-19 was going to ICU with flu, do they have the same sort of profile? Because we, iatrogenically, are messing them up, basically. And I guess the second question I would have is like, okay, this is great. You're almost saying I can profile someone who's going to go down, which would have been helpful to us three years ago when it was like a disaster to pick out. Who's going to ICU? Who's, you know, figuring out who is the person you're going to accept for transfer, you know? And so do you think that these things could have been added to the biomarkers we were already doing, the CRPs and ILD, you know, IL-6, to profile who's really at risk of going down? Yeah. And I think that's a very good question. Dr. Kim and I, we are transplant doctors, so we are doing lung transplantation. And I think for us, that would be extremely helpful in the future because that can tell us that might be a biomarker to tell us which patients are going to have more severe COVID infection and maybe help us prioritize which patient needs to be admitted to the hospital. Because as transplant doctors, we always get those calls. I got COVID infection, and I'm having minimal symptoms. What should I do? And you're very worried about your transplant patients because you know that they can have severe complications. And actually, recently, we had patients with minimal symptoms that the next day end up having significant infiltrates and oxygen requirements. So having a biomarker of the disease severity would definitely be helpful because we will try to keep the patients that don't need to be admitted, we can try to keep them isolated at home. Okay. Thank you very much. Thank you. Our next speaker is Dr. Lee, who is talking to us on concomitant COVID-19 infection and pulmonary embolism incidents and in-hospital outcomes in a nationwide cohort. Thank you for coming today. It's not Dr. Lee. It's, sorry, it's on a, we have a different list, but it's a. Please tell us your last name. Last name is Padavel, but I go by Rana. Rana. Okay. Good morning, everyone. I'm Rana. I'm one of the first-year fellows at SUNY Upstate Medical University. And today I'll be talking about a little work that we did on patients with concomitant COVID-19 infection and pulmonary embolism. And we looked at the incidents and in-hospital outcomes in a nationwide cohort. And I have nothing to disclose. So after coronary artery disease and stroke, acute PE is one of the largest causes of cardiovascular death and in-hospital mortality. And some studies have shown rates up to 30%. So existing studies have shown that COVID-19 infection is an independent risk factor for the development of pulmonary embolism. So our study was designed to see if patients hospitalized with pulmonary embolism and coexisting COVID-19 infection, whether they had a higher mortality as compared to patients who are admitted with just pulmonary embolism without a coexisting COVID-19 infection. The study design was a retrospective cohort study, and the sampling was done on the National Inpatient Sample Database. I would like to take a minute to talk about the National Inpatient Sample Database. It's run by the Agency for Healthcare Research and Quality. It's a stratified sampling of all of the inpatient hospitalizations in the United States. This data is stratified sampling from all of the CMS and reporting to the Center for Medicaid and Medicare. And from there, they have sampled 5 million samples of inpatient hospitalizations. Those are the number of samples available in the 2020 version. The variables that are available in this database includes some of the basic patient demographic characteristics, include age, race, sex, and median household income for that particular zip code. But they also stratified it according to the hospital status, whether it is a teaching hospital, non-teaching hospital, and also whether that was an urban or rural hospital. We also have some of the other variables available, including the outcome of that hospitalization, whether it was discharge or death, whether the length of stay, and some of the severity and comorbidity measures, which are basically derived from the ICD-10 codes. So we have up to 30 ICD-10 codes available, and based on that, we can make some severity measures of their comorbidity burden. So for this study, we kept an inclusion criteria of pulmonary embolism as the primary admitting diagnosis. We excluded patients with age less than 18 years of age, and all of the non-emergent admissions, that are the elective admission for procedures and such. So when we did the stratified sampling, there is a variable called discharge weight, which we can use to expand the cohort to a nationwide cohort. When we did that, the 2020 version could represent about 32 million hospitalizations, out of which adult non-elective admissions were 26 million, and acute pulmonary embolism as the primary admitting diagnosis, we got 425,000, which we separated into patients with COVID-19 infection and patients without COVID-19 infection, based on their ICD-10 coding, if it had COVID-19 in it or not. So we found that approximately 11% of the patients had a simultaneous COVID-19 infection, and the rest, 89% did not. In terms of the outcomes we looked at, the primary outcome we were interested in was in hospital mortality, but we also looked at some of the other secondary outcomes, such as length of stay, total hospitalization charges, requirements of vasopressors, which we thought would give us an indicator of the severity, whether or not ECMO was utilized, and interestingly, we also looked at whether or not they had underwent a thrombolysis or an embolectomy during that particular hospitalization. Statistical analysis was done with Chi-square test and Mann-Whitney test, but the multivariate analysis was also performed for the outcomes, and we adjusted for age, sex, race, some of the comorbidities, such as hypertension, diabetes, smoking status, et cetera, and also the hospital characteristics, which has proven to be affecting the outcomes in patients with COVID-19. So I'm not talking about all of the comorbidities, but some of the variables in the baseline characteristics we showed that age was similar across both groups. We had a female predominance in patients without COVID-19 infection. In terms of race, we have a Caucasian predominance in patients without COVID-19 infection, but African-American and Hispanic races were higher in the PE with COVID-19 group. The comorbidities we looked at, hypertension was similar across both groups, diabetes was more in the PE with COVID-19 group, but interestingly, all of the other comorbidities, such as CH of smoking, malignancy, COPD, and a prior history of VTE, they were all higher in the PE without COVID-19 group. So the primary outcome we looked at was in-hospital mortality, and we found that mortality was higher in the PE with COVID-19 group, approaching up to 20%, whereas the PE without COVID-19, the mortality was roughly around 7%, which is similar to some of the national estimates. In terms of secondary outcomes, we can see that requirements of vasopressors need requirements of ECMO, average median length of stay, as well as total hospitalization charges. They were all high for the patients with simultaneous COVID-19 infection, as opposed to those without. This should be interpreted in light of the fact that this group had, in general, lesser comorbidity burden than the other group. The other interesting outcome was that the procedure utilization. So we looked at systemic thrombolysis, catheter-directed thrombolysis, and thrombectomy. We found that if there was a coexisting COVID-19 infection, they were less likely to undergo any of these procedures. So the thrombolysis, embolactomy, and all were higher in the PE without COVID-19 group. So just to summarize the results, patients with PE and COVID-19 infection had, in general, lesser comorbidity burden than those with PE without COVID-19 infection. And in PE with COVID-19 infection, they had a higher mortality, higher total hospitalization charges, and a longer length of stay. And they were less likely to undergo systemic thrombolysis, catheter-directed thrombolysis, and thrombectomy. So we are not sure why the outcomes were skewed towards this way, but our guess or anticipation is that the last personal protective equipment potentially opposed patient burden. As you can see, this was only in the 2020 sample. That was the beginning of the pandemic where we had a lot of patients coming together at once. And concerns regarding evidence potentially resulted in the PE and COVID-19 group receiving lesser interventions as compared to the PE without COVID-19 group. It's unsure if the reductions in intervention resulted in a higher mortality in the group, though the mortality could simply be attributable to COVID-19 itself. But what we would like to do further is the 2021 NIS should be coming out soon, and 2022, the year after. It will be interesting to see once we have a better understanding of the disease process itself, where we're more willing to do thrombolysis or embolactomy in these patient groups as opposed to how it happened in 2020. So that would be something interesting to look at. Also, this should be interpreted within the limitations of NIS itself. So as you can see, NIS, it is data collected from the Center for Medicare and Medicaid. And the data is only as good as the coding that was used. So it is always limited by the ICD-10 codes that were used for billing for patients who are hospitalized. It also doesn't give us any severity information, how severe the disease process was. So it doesn't give us any laboratory data, doesn't give us any imaging, so we couldn't adjust it for the size and extent of the PE itself. And I think the most important one is we couldn't have any information regarding to anticoagulation use. Of course, that's another important factor that if I had the choice, we would have loved to look at. And the analysis is always also limited to the in-hospital only. So this is an in-hospital database, so we don't know what happened to those patients after they were discharged home. So I think a longer follow-up and data of out-of-hospital, out-of-discharge outcomes, that would be interesting as well. Thank you. So I know that you weren't able to look at that and I wasn't quite clear, but so in addition, were there other thrombotic complications? Because it's very hard to tell, determine whether the pulmonary embolism patients passed away or had excess morbidity and mortality from PE or they just had the PE as part of their whole constellation of disease. And I'm thinking of myocardial infarctions and strokes, are there thrombotic complications that we've seen with COVID? Right. So we only selected patients who had the primary admitting diagnosis of pulmonary embolism. So we assume that that was the reason why they were brought into the hospital and the PE, the COVID-19 was just existing with them. Unfortunately, with the, for us, NIS doesn't give us the reason for the mortality or the disease process that actually caused the death. So that level of granularity, we could not do with the database that we had. Sir. So I guess I was going to go on the same line, but I guess you mentioned that there was no cause of death data available and there was also no data available on the treatment of those patients. Right. So, I mean, within the limitation of the design, you guys did a really good job doing that. But I guess going forward, if you're trying to look at it in the next year, maybe try to use, I mean, it's okay to have a smaller sample size and look at a patient on whom you can get a more detailed data. This is useful. I mean, we know that COVID increases the risk of PE. We know that COVID itself has increased mortality. The question we couldn't answer here is that when both of them got combined, how bad really things got and how much of that because of the PE rather than just having a bad COVID because we don't know the COVID severity in this, right? So maybe in the next session, when you try to look at the data, instead of having the same data source, go to a smaller source where you can get more granular information and that will be more helpful to know. I agree. Thank you. Thank you. I think that we should move along. Let's go on to our next presenter, which I have here as Dr. Young-Seok Lee speaking on respiratory complications after Omicron variant infection in healthy population under 60 years of age. Thank you. I am Dr. Young-Seok Lee from Korean University Grove Hospital in South Korea. The title of our study is respiratory complication after Omicron variant infection in health population under 60 years of age. I intend to explain KT's respiratory complication following Omicron variant infection within a healthy individual under the age of 60 in South Korea and to discern predictor associated with a medium to high health impact resulting from respiratory complication. And I would like to engage in discussion regarding the management measure for addressing respiratory complication associated with COVID-19. The mortality rate associated with COVID-19 has rapidly declined owing to advancement in vaccine development and diversification of treatment strategy. However, the post-COVID-19 condition remains a Northwest public health concern. I think that the explicit mechanism underlying post-COVID-19 condition encompasses organ dysfunction. Elderly patients have experienced severe respiratory complication attributed to the advanced disease state during COVID-19 infection. Therefore, most of elderly patients experienced COVID-19 sequelae, subsequent to the initial infection. However, the extent of respiratory complication in health population under the age of 60 remains unclear. The Omicron variant manifests a less severe clinical profile marked by a heightened prevalence of respiratory tract infection compared to the preceding variant. So patients with Omicron variant infection exhibit a low instance of pneumonia compared to preceding variants, particularly among the young and middle-aged adult patient. The purpose of study was to examine respiratory complication following infection with Omicron variant with health population under the age of 60 and to identify predictor associated with medium to high health impact resulting from respiratory complication. This study was a prospective observation study and was conducted at 12 university-appealed hospitals from January and December 2022. The Omicron variant has been a prevalent species in South Korea since the first case was reported on November 2021. Inter-participants took part in telephone interview about their respiratory complication using well-valid CAT score by a research nurse at three and six months after COVID-19 diagnosis. The CAT score was developed and validated by GSK to assess the impact of COPD on individual's life. The CAT has a scoring range of zero to 40. The total score of then 10 points indicates clinical significance necessary for COPD. During the study period, 696 patients were included. The medium age of the participant was 32 years and 21% were male. The medium duration of isolation at home during COVID-19 infection was six days and 33 participants were re-infected with COVID-19. None of the patients were diagnosed with any disease prior to enrollment. Results. Three months after COVID-19 infection, 64% of patients reported dyspnea and 48% reported impregnancy with cough and 42% exhibit sputum production. The prevalence of respiratory symptoms did not improve at the six months following COVID-19 infection. Three months following COVID-19 infection, the proportion of patients with a total CAT score exceeding 10 points was approximately 24% and this ratio did not improve statistically significant. Three months following COVID-19 infection, approximately 13% of patients initially classified as low risk experienced on escalation to medium to high impact at six months, whereas 44% of patients initially categorized as medium to high risk demonstration on improvement transitioning to low impact at the six-month interval. Attention should be directed toward these patients to mitigate respiratory complication following COVID-19 infection. Multivariate logistic regression showed that older age female and longer isolation period were significant predicts of medium to high impact due to respiratory complication. In conclusion, this study revealed that individuals within the health population under the age of six years exhibit various respiratory symptoms such as dyspnea, cough and sputum. Approximately 23% of participants reported a medium to high health impact attributed to respiratory complication. And practices such as advanced age female gender and prolonged isolation were associated with diminished quality of life due to respiratory complication. I want to discuss with you for this topic. How can managed strategy be optimized to mitigate respiratory complication associated with COVID-19 infection? What is your response to this question? I see a correct response to this inquiry. I ask this question to CHAT-GPT. The response for CHAT-GPT is as follow. Enhanced vaccination, early detection, integrated case system, capacity building, research and development. I think that CHAT-GPT is amazing tools. Thank you. Thank you very much. If you have questions, please step up to the microphone. I wanted to ask you, I was initially curious as to why you used the COVT, the CAT test as an evaluation for respiratory symptoms since, you know, this is designed for patients with chronic obstructive pulmonary disease. And I don't think that was probably, you know, part of the underlying diagnosis in many of your patients. Although it's good that you have identified significant parameters. Yeah. And previous study showed that simplification for respiratory complications. And I think that well-variated questionnaire is accurate than previous studies. And so CAT score consists of eight items. And each question is focused to respiratory symptoms. So I used to CAT score. Please. Hi, Doc. I'm Korean as well. Welcome. I was taken about by two things actually on your slides. Number one, that the average Korean's BMI, that huge group was 20s, which I wonder to myself like, oh, you know, we talk about weight as having an influence in COVID-19 respiratory symptoms. But from what you describe, I guess not so much. I mean, even in a different country where everyone's BMI is slimmer, you have a lot of respiratory complications from COVID-19. So I took away that point. And the second point, I wonder if some of this reporting is cultural. Like, why would women complain more than men? You know, I don't know if that's a, you know, if there's a correlation to how they did. Like, do the women do, you know, like clinically much worse, not in addition to reporting more? Do you know what I mean? Like, I'm just curious to know if there's like a gender bias on how they view their COVID-19, you know. I'm surrounded by Korean women in my family, and they're very dramatic. So I'm wondering if that's part of it. That's very key, because I think that, you know, men tend to minimize their symptoms. And so that may explain in part some of your findings, but it's not clear. But point well taken. Sir, please. Last year at these meetings, there was a lot of data presented regarding the pathophylogy of the disorder. And most of the arrows seemed to point to inflammation. Now, I've not really seen any studies that looked at treatment of these patients. I've seen a couple hundred of these patients, and none of them are being treated. So I'm wondering about why that is the case, and why we don't see anything about it in the literature. I don't know if you want to answer. I don't have a good answer for you either, but I do know that many of these patients that we see with what is now considered long COVID, because this is post-COVID, the symptoms that persist and have dyspnea, a lot of it is related to hyperactive airways. And many of these patients are treated with the standard inhaled corticosteroid lob accommodations. And they seem to do fine. Yes, please. Yes, please step up to the microphone. And one last comment. Microphone, please. We are at UT Southwestern, and we started a COVID clinic, and we are seeing patients. Oh, please sit a little closer. So I was saying that we started a COVID clinic, and we are seeing patients with interstitial lung disease, post-COVID ILD. And we developed immunosuppressive treatment for them, combining high-dose steroids with anti-metabolites with CellCept. We have very good results, and we are going to present our results during the poster session on Wednesday. Okay, thank you very much. Thank you. I don't have, I think our next presenter is, let me make sure it's Max Yang. I don't want to. Yes. Our next presenter is Max Yang on Spontaneous and Orthorax Immunocyte Incidents and Outcomes in COVID-19. Do I get a pointer on this? Yeah, just double-click on that. And the last one is, that's the last one. Okay. Cool. All right. Good morning, everyone. Thanks for being here today. Our presentation will be discussing our work on characterizing the incidents and outcomes of spontaneous pneumothorax and pneumomyasdinum in adults with the COVID-19 pneumonia. My name is Max Yang. I'm a third-year internal medicine resident at Los Angeles General Medical Center. I have no financial disclosures. Okay. So spontaneous pneumothorax and pneumomyasdinum, which I'll hereby henceforth refer to as PTX and PM from here on, has been observed to have increased incidence in patients with COVID-19 pneumonia. However, today only a small number of case series and cohort studies exist to describe the phenomenon, as well as its risk factors and outcomes. They're not particularly well characterized at this point. So in order to address some of these knowledge gaps, we designed a study to examine, first off, the prevalence of PTX and PM in the adult COVID-19 patient population. We also wanted to examine independent risk factors for developing this complication. And finally, how this complication affects patient outcomes. So our study was a single center retrospective cohort study at the Los Angeles General Medical Center. We looked at all adults who were admitted to our hospital who tested COVID-19 PCR positive during or upon their admission between the months of January 2020 to March 2021. So that mostly encompassed that first and second surge. We then looked at coding information in these patients to examine who in our COVID-19 population had a coded instance of PTX or PM. And we manually reviewed their charts to look at, you know, time of onset, in-hospital mortality, what kind of procedures they got within 24 hours of onset, et cetera, et cetera. We adopted a two-tiered statistical analysis approach to answer our second and third study questions. To look at risk factors, we first set our primary outcome to an in-hospital PTX PM event. And then we used a multivariable logistic regression to see if there was any independent correlation between these co-variables here. Then to look at patient outcomes, we actually set our primary outcome to in-hospital mortality. And then we used a similar multivariable logistic regression in addition to PTX PM to try to see if there were any independent correlates there. We also looked at secondary outcomes of homebound discharge as well as 90-day hospital free days. So what did we find? So this is our table one, kind of breaking down our patient cohort. So it's kind of a busy slide, so I'll just point out some salient details here. Our cohort identification strategy identified 2,812 cases of patients who tested positive on COVID-19 PCR. And out of that cohort, we actually managed to identify 89 patients who developed either a PTX and or a pneumo PM, which represented a 3.2% event rate. Kind of going down to our cohort characteristics, I wanted to point out that our patient population was predominantly Hispanic and that we actually captured a wide range of COVID severities with patients on room-air-upon-admission all the way up to and including patients who are on invasive mechanical ventilation. We also identified who among these patients were on, were initiated on steroids within 48 hours. And then looking at our mortality data here, we saw that our overall mortality was 12.6%, which is roughly comparable to the National Healthcare Survey data from that time. I will also foreshadow here that our mortality in our PTX PM population was a staggering 67.4%. Looking a little bit closer at our case population, there were a few things that I want to note here. So 67 of our 89 cases had onset while on invasive ventilation, which left another 22 who developed a pneumothorax and rheumatoid myelostatin, not even on any sort of positive pressure ventilation. Among those 67 cases, our median time on ventilation until onset was just short of like a week. In addition, we also saw, interestingly enough, that 25 cases developed in the PTX or PM within 24 hours of an insertion of an ETT. We captured 463 ETT insertion events within, or 463 patients who ended up getting intubated in our patient cohort. So this actually represents like a 5% event rate associated with intubation. Looking at this chart on the right, we broke down our patient cohort by the amount of respiratory support they needed on admission. And we also saw kind of like a dose-dependent behavior of the rate of PTX PM in each subpopulation of respiratory support. Moving on to the results of our multivariable logistic regression, we saw two main independent predictors of developing a PTX PM. The first one was actually admission respiratory support, which we kind of simplified to whether they were on room air, noninvasive ventilation, or invasive ventilation. And then we also saw that steroid usage early on in their hospital course also seemed to be an independent predictor of developing a PTX or PM. Okay, going back to our table one, now broken down by alive at discharge versus deceased in hospital, we have our data here. Looking at the results of our multivariable regression, we actually saw, we were able to actually reproduce some of the existing literature on mortality risk factors in COVID-19, for example, we saw that older patients tended to have a higher propensity of death with COVID-19. But the new thing here that I wanted to point out was that PTX and PM also have a very strong independent predictor, independent correlation with in-hospital mortality, independent of the rest of these co-variables. Looking at our secondary outcomes, since, you know, we saw so much mortality in the case population, you know, it wasn't really a surprise that we also saw independent correlation with lower rates of homebound discharge, as well as a lower number of hospital free days. We also did a sensitivity analysis to take a closer look at our data, first what we did was we excluded all of our patients who did not need any supplemental oxygen on admission, and this did not appreciably change the outcome of our analysis. We also did a subgroup analysis of patients who were, of only looking at patients who were on mechanical ventilation on admission, and that also demonstrated similar results as well. So again, going down to our discussion here, to summarize our findings, we found a high rate of PTX PM in our patients with COVID-19. We saw maybe one case out of every 30 patients who tested PCR positive on admission. Our risk factors that we identified was admission oxygen requirement, which seemed to follow, which seemed to show that the more oxygen the patient was on admission, the more, the higher propensity they had to develop a PTX PM. And we also saw early steroid initiation as another risk factor as well. We also saw that PTX PM incidentally noted that it was, it seemed to have an increased complication rate with intubation in COVID-19. We, again, to remind you all that we saw one PTX PM with, for every 20 intubation events. And then lastly, we saw that PTX PM was independently associated with worse outcomes across the board, higher mortality, lower rates of homebound discharge, and lower number of hospital free days. Going down to our strengths and limitations of the study, we managed, I think we managed to capture the largest cohort of PTX PM associated with COVID-19 in a single study in the existing literature. We also covered a broad range of COVID-19 severities, which allowed us to capture a significant amount of PTX PM cases in patients who are not on any sort of mechanical ventilation. Some of the limitations here, I have to give lip service to some of the inherent limitations of having a retrospective single center study, which was relying on coding data. Our single center setting limits the generalizability of our data and the coding data may underestimate incidents of PTX PM depending on, you know, depending on gaps in reporting. And I think our biggest limitation here was that we unfortunately did not include a validated disease severity marker such as a SOFA score in our analysis. So unfortunately, while we were kind of able to scratch the surface of that question, we were unable to see if COVID-19 or PTX PM was simply just another marker of disease severity. I want to take a moment to acknowledge everybody who were able to, was able to make this study happen, including my faculty mentor, Dr. Huerta, who's sitting in the back left over there. And I'd like to thank you all for being here and for your attention. Thank you. I want to ask, because I think this is, you've hit on it, because I think we've all seen the marrow trauma, pneumonia, myasthenia, in these patients. It's often a marker of the severity of the disease. And I noticed that the high flow patients also had a higher, and since I wonder if there's, you know, if you, in your analysis, were able to link anything with, you know, the pressures that you would get post-inhibition. I don't know if you looked into that detail, you know, ventilator pressures, peak airway pressures, mean airway pressures, and the patients who were, yeah. Yeah, so we did look at this. We didn't have time to get into it in too much detail here, but I did include this slide here because I was going to, I figured that somebody was going to ask. Unfortunately, because it was a retrospective study, we did have a lot of non-uniformity in recording a lot of this ventilator data. So as a result, you know, when we attempted to do, you know, our non-parametric paired testing, we actually had, we were unable to see anything, first off, but I think the reason why that was was because we were only able to identify, like, maybe like 50, 50 patients who had paired ventilation data before and after pneumothorax and pneumometastinum. And so I don't know if it's, like, maybe just like a type 2 error. Maybe we didn't have, like, enough paired data to try to really capture any signals there. So unfortunately, that study was not conclusive in that regard. Hi. How does your rate of ventilator-induced lung injury in PTX and PD compare to, say, pre-COVID-19? So have you looked back at the data to see what the rate were in general on Philly? Yeah. So I did find, just to kind of give you an idea, we had 67 patients who developed this complication out of the 463 patients who needed intubation. So that's, you know, if you do the math, that's, like, roughly like a 15% incidence rate amongst our intubated patients. And I think in non-COVID arts, I read in a study that that rate was actually about 6.5%. So it was, we did see that it was higher. Yeah. I've seen numbers as high as 25%. Oh, wow. Okay. Yeah. Thank you. Got it. Thank you. I think we'll move on. Our last presentation. So here we have prevalence of pneumothorax in COVID-19 patients. A comparative analysis with other viral infections using the National Inpatient Sample Database. Dr. Sood, take it away. All right. Hi, everyone. Good morning. So my name is Gadiel Al-Sawidi, one of the formerly attending Jersey Shore, Hackensack Meridian, Jersey Shore Hospital. So we did encounter a lot of pneumothorax during COVID patients, especially during the first wave of COVID. So we looked at the national database for patient to see what's the incidence and what's the prevalence, sorry, of the COVID patient having pneumothorax and some etiologies of pneumothorax in COVID patient. So I'm going to skip the background as it gets because I just want to, pneumothorax, as you guys know, is a collection of air outside the lung and within the coronal cavity. It is a primary or secondary etiology and is traumatic and atherogenic. But pneumothorax in the setting of coronavirus has been described in case reports on small observational studies. However, the associated between the two conditions have not been established yet. So we aim to investigate the prevalence of pneumothorax among COVID. We compared it with the prevalence among other viral respiratory infection. So it's a retrospective on cohort study. So we're using the National Inpatient Sample Database 2020, which represent 20% of all hospitals, all pair hospitalization. I think my colleague talked about that a little bit. So I'm going to skip that for a second. So, so patient diagnosed with COVID were compared to those with other respiratory viral infection, including flu, RSV, part of flu on human metanemovirus. The primary outcome was the prevalence of pneumothorax. So multivariable logistic regression analysis were performed to assess the association between viral infection on the development of pneumothorax, adjusting for potential co-founders. So during the data management measurement was the age was reported in years. The race also was categorized in white, black, Hispanic, Asian, and others. Primary insurance also was categorized, Medicare, Medicaid. The only thing I want to highlight is the medical condition were identified through the international classification of the disease. So using the ICD-10. So a total of over a million, over almost 2 million encounters were included. So almost like a million and 600, over a million and 600,000 patient with COVID on 253,000 with other viral infections were analyzed. So I think pretty much the other things as we divide out the same, the age of race, primary insurance, on the medical condition. So this is a total number of patients were been seen. We encounter like almost 2 million. So unadjusted rate for pneumothorax. So I'm saying that unadjusted because we look at the database. Well, the database is just saying, telling us the patients who had pneumothorax with COVID. But unfortunately we cannot go in deep. So a lot of co-founders like including like how long they've been intubated, how high the people were encountered, that any procedures has been done during the hospitalization, including thoracentesis or TLC placement, as well as steroids, all these things. So it's unadjusted rate of pneumothorax is almost 1.5% we found. With other viral infections, 0.7%. With a P-value less than 0.01. Mechanical ventilated patients, almost 9.1 patient, while other viral infection patient, 4.6%. Non-mechanical ventilated patients, we found like 0.5%. While with other infections, 0.3%. Also statistically significant. So after adjusting for potential co-founders, that's what we found. As I said, it's not everything, including the database, just the age, sex, race, insurance, hypertension, diabetes, CKD, obesity, COPD also is not classified how bad the COPD, how severe the COPD, but just COPD. Hyperlipidemia, alcohol, smoking, malignancy. COVID was independently associated with increased risk of pneumothorax. As you can see, it's like with 1.69, 95 confidence interval is about 1.6 to 1.7. That's what's compared with other viral infection. So potential theology of pneumothorax in patient with COVID is a formation of alliances. This is, I'm going to a couple of studies has been done by a son and his colleague who are demonstrating some radiological evidence for the progression of lung involvement in patient with COVID. We see ground glass capacities, then consolidation, then giant bullion, then pneumothorax. Other case reports have been shown the formation of cysts suggesting that the occurrence of pneumothorax is not solely attributed to barotrauma or mechanical ventilation. Similar observation of cyst formation has as a complication of acute respiratory syndrome caused by SARS have also been reported in some previous cases. So we concluded that the rate of pneumothorax is significantly higher in a patient with COVID compared with those with other viral infection. Clinicians should remain vigilant for the potential complication of pneumothorax with any COVID patient. Further research, as I told you, should further investigate the underlying mechanism on potential risk factor associated with pneumothorax in the context of COVID patient, especially the length of stay, how long they've been intubated, what's the PEEP been used, pruning, steroids, any procedures has been done. That's but not measured during the whole database. Thank you.

retrospective cohort study

National Inpatient Sample database

prevalence

pneumothorax

COVID-19 patients

viral respiratory infections

mechanically ventilated

non-mechanically ventilated

risk factors

complications