Hello, and welcome to today's Journal Club Critical Care Medicine webcast. This webcast, hosted and supported by the Society of Critical Care Medicine, is part of the Journal Club Critical Care Medicine series. This webcast features two articles that appear in the May 2022 issue of Critical Care Medicine. This webcast is being recorded. The recording will be available to registrants on demand within five business days. Log into mysccm.org and navigate to the My Learning tab. My name is Jeremy Nielsen, the Associate Director of Journals at the Society of Critical Care Medicine. I'm speaking to you on behalf of Dr. Zakmani, who you see here, and is our session organizer. Thank you for joining us. We do have just a few housekeeping items before we get started. First, during the presentation, you will have the opportunity to participate in interactive polls. When you see a poll, simply click the bubble next to your choice. There will be a Q&A session at the conclusion of both presentations. To submit questions throughout the presentation, type into the question box located on your control panel. If you have a comment to share during the presentations, you may use the question box for that as well. And finally, everyone joining us for today's webcast will receive a follow-up email that will include an evaluation. Please take five minutes to complete the evaluation. Your feedback is greatly appreciated. Please note the disclaimer stating that the contents to follow is for educational purposes only. And now I would like to introduce today's presenters. Pontus Hedberg is a medical doctor from Karolinska Institute in Stockholm, Sweden. He is currently pursuing a PhD focusing on epidemiological research of various acute, post-acute, and long-term effects of SARS-CoV-2 infection. Recent scientific publications include work on acute effects of SARS-CoV-2 compared to other respiratory viruses, bacterial co-infections, perioperative infections, and ventilator-associated lower respiratory tract infections. The latter will be presented during the upcoming webcast. His ongoing research projects focus primarily on post-acute and long-term effects of COVID-19, including occurrence of post-COVID-19 condition and healthcare use and disease trajectories before and after acute infection. Dr. Peter Teinman is an EDIC-certified intensivist working at the University Hospital at Amsterdam University Medical Center. He has a master's degree in epidemiology, and his research focus is on point-of-care ultrasound, ARDS, and mechanical ventilation. He is the co-founder of the A-Life Study Group and free educational website alifeoffocus.com. He has co-authored over 120 peer-reviewed manuscripts. Dr. Teinman teaches point-of-care ultrasound both nationally and internationally. Furthermore, he is director of the intensive care training program in his hospital. Thank you both very much for joining us today. I will now turn the presentation over to Dr. Hedberg. Thank you so much, Jeremy, for the kind introduction. And first of all, I want to make sure, can everyone see me and hear me? I assume yes. And so my name is Pontus Hedberg, and I'm a medical doctor. I'm currently wrapping up a PhD at Karolinska Institute in Stockholm, Sweden. So today I have the honor and the privilege to present our work on the ventilator-associated low respiratory tract bacterial infections in patients with compared to without COVID-19 in an ICU setting. So it's a great privilege, and I'm presenting today on behalf of all the study authors for this study. So moving on to the next slide, some disclosures. This work was supported in part by the Swedish Innovation Agency, which in Swedish is called Vinnova, as well as the Region Stockholm and the Emil and Vera Cornell Foundation. My supervisor, who, interestingly enough, also called Pontus actually, received funding from the Swedish Innovation Agency, Region Stockholm and the foundation I previously mentioned. I was supported by the Karolinska Institute, which provided funding for a combined clinical studies and PhD training program. And Dr. Tarnov received funding from Vinnova as well. The other study authors did not have any disclosures or conflicts of interest. So first of all, connected to this article, as Jeremy mentioned, that was published in the May 2022 issue of Critical Care Medicine, I think an excellent and highly relevant editorial was written by Dr. Andrej Kalil and Dr. Jonathan Ryder about the current understanding of ventilator-associated pneumonia and COVID-19 and what we do and what we don't know. So I highly recommend you to read that in conjunction to the article I will present today. So just giving you a brief introduction to the ICU patient from a clinical VAP as well as an epidemiological perspective, it's a multi-state process where you have patients admitted to the ICU, which, of course, can be treated with or without mechanical ventilation during the course of their ICU stay. What you see here in this multi-state flow diagram, basically, is that the patient can transition from the state of being not treated with mechanical ventilation to receiving that treatment. And then you, on the bottom side, the three bottom boxes, you basically have absorbing states with regards to the specific study outcomes we investigated, which is that the patient can develop a ventilator-associated lower respiratory tract infection, the patient can, of course, die during the course of mechanical ventilation, as well as not receiving mechanical ventilation, and the patient can be discharged from the hospital. And this is important to keep in mind when studying these outcomes from an epidemiological perspective that you do need to be aware of what are the competing risks or the competing events being present for your primary outcome. Moving on, mechanical ventilation and COVID-19, well, I would say it could cover an entire session, of course, obviously, so I will just highlight a few points I think is relevant given the presentation I will, and the results I will provide today. What we have seen, both from the clinical experience and from previous literature, previous scientific reports, is that higher ventilator-associated pneumonia rates have been observed in COVID-19 patients compared to patients without COVID-19. One problem, of course, with regards to VAP, which is the abbreviation for ventilator-associated pneumonia, is with regards from a COVID perspective, there is a great substantial overlap in the clinical presentation of VAP compared to the actual presentation of COVID-19. Those include, but are not restricted to fever, leukocytosis, bilateral infiltrates, and so on during the presentation. This, from an epidemiological point of view, when trying to understand the presence of ventilator-associated pneumonia and the occurrence and incidence of it, is a bit complicated. And due to that, we took somewhat a more cautious approach in this study and actually referred to it as ventilator-associated lower respiratory tract infections. I will describe why quite soon. One question, of course, is why do we see these higher rates in COVID patients compared to non-COVID patients? Well, there are, as has been outlined in previous reports and in the editorial I also recommended to you, there are various factors that could be involved here. One is, of course, viral immunomodulation. One is, of course, the very prolonged ventilation durations that we see in COVID-19 patients compared to other ICU patient populations. Another one is, of course, prolonged hospital stays before actually being admitted to the ICU. And then, of course, we have, which we will touch upon a bit, that I still think is a bit unknown, it's not fully elucidated, the role of pharmaceutical interventions and other non-pharmaceutical interventions in COVID-19 and their role in affecting the occurrence of that, basically. Prone positioning, of course, is also being used. And then, of course, we have more systemic healthcare-related factors, such as personal fatigue, personal experience, the use of personal protective equipment, strains on the healthcare systems, and so on. So, as you probably can appreciate, this is quite a complex topic, and to fully untangle what is what here, it's not very easy. So, let's proceed to the next slide. So, as I mentioned, basically, there is a challenge in defining ventilator-associated pneumonia in COVID-19 patients, as I previously mentioned, with regards to fever, leukocytosis, and infiltrates. Due to that fact, in this study, we used a strictly microbiological criteria to define ventilator-associated lower respiratory tract infections, where we used quantitative microbiological cultures and different CFU cut-offs, as you can see here, or that you can read in more detail in the manuscript. So, basically, all the microbiological cultures from the lower respiratory tract from 48 hours after intubation until extubation were considered for defining the outcome in the study. And we restricted for the definition of VAP or VALTRI, as I will call it in this case, we restricted the outcome to significant bacterial pathogens, which were classified by experienced clinical microbiologists and infectious diseases doctors involved in the study. And moving on, the aim of this study, hence, was to investigate the occurrence of microbiologically defined bacterial ventilator-associated lower respiratory tract infections among patients being mechanically ventilated with or without COVID-19. And this we did during the first 10 months of the pandemic, roughly. So, from March 2020 until December 2020. And we compared these to both the pool non-COVID-19 patient population and to specific subpopulations that I will mention more in a few slides. So, moving on, this is just a very schematic overview of the study. So, basically, as you can see for the non-COVID-19 patients, we stretched as far back as until 2011, up until the end of the study period, which was the end of 2020. And for the COVID-19 patients, those were included from the onset of the pandemic, which was roughly the 1st of March 2020. And dividing it up into two waves, basically, the first wave and the second wave. As you can see, August and September, in this case, it's not included in any of these waves. And the reason for that is that, basically, the amount or the number of COVID patients during these months were very low. And they were actually not attributed to any of these waves in Sweden. And we only, I should mention that we only looked at ventilator-associated events or infections here in adult patients, patients being 18 years or older. Just to give you a quick perspective of what the time period actually reflects with regards to variants, the left-hand side of this dashed line is the study period that was investigated. So, we did not look into ventilator-associated infections during, for example, the alpha, the delta, and now the unfolding Omicron wave, just for your reference. So, moving on, please, by any means, don't try to look at the entire table here. I will try to guide you through the important findings here. Basically, what we did was to, one could, from an epidemiological perspective, say that we restricted the study population to patients being ventilated for more than 48 hours, because those are the only patients being at risk for this outcome. However, we decided to actually include all ventilated patients. And the reason for that was kind of to highlight the long ventilation durations in COVID-19 patients. So, what you can see here is that 426 out of 479 COVID patients were treated in the ventilator for more than two days, two days or more, whereas the corresponding proportion for the non-COVID-19 patient populations were much lower. I will not go through the exact details here. I'm currently working on trying to understand the long-term effects of ICU treatment in COVID patients compared to non-COVID patients during the year following the critical care episode. And what we do see is that COVID-19, the COVID-19 ICU population is a quite distinct patient population. And that you need to keep in mind when trying to compare with other non-COVID-19 groups, that this is actually, in many ways, these are a case mix of patients that normally is not seen in the ICU setting so much. And that you need to be aware of. That includes both with regards to age and with underlying pre-existing health status. And that's exactly what we saw here as well, that the patients were younger, the COVID patients were younger and had fewer comorbidities. So, moving on, the left-hand graph here that you see is basically on the x-axis, you have the number of days treated in mechanical ventilation. And on the y-axis, you have a subset of the most common ICU diagnosis provided by the ICU physician here, ICU physicians in the ICUs. And what you probably already know and what is fully reinforced here from this graph is that COVID-19 patients are treated exceptionally long in mechanical ventilation. And that, of course, have a substantial impact both on the acute care and the following care after the critical illness. So, what did we find? So, what you see here on the right-hand side is basically the cumulative incidence with days at risk. So, basically, the time the clock starts ticking at day two after the start of mechanical ventilation and onwards. And what you see is that COVID-19 patients had a higher COVID-19 ventilator-associated respiratory tract infection cumulative incidence compared to the non-COVID-19 patient population. What you're not taking into account here is, of course, the risk time. How long is the patient being mechanically ventilated? As you see here from the black lines, basically, the COVID patients had substantially lower chance or risk depending on what the outcome of the extubation was, but they had a substantially lower probability of being extubated at any given time point, basically. Moving on to the next slide. So, here, basically, what we have done is that we have classified among patients being mechanically ventilated for two days or more, what was the occurrence, what was the incidence rate of ventilator-associated infections? So, in the top row here, you have the COVID-19 patients. So, you see that you had an incidence rate of 31 valvotries per 1,000 days at risk compared to the pooled non-COVID-19 patient population where it was slightly higher, 34 events per 1,000 days at risk. However, what you probably can understand and appreciate is that the pooled non-COVID-19 patient population is a very diverse patient population. And accordingly, we try to understand that by looking into specific diagnosis being either a diagnosis that we found to be of clinical relevance and of a comparative relevance or conditions which are normally seen or commonly seen in the ICU setting. So, what we did find was that compared to all other infectious diseases that we investigated, bacterial pneumonia, influenza, and severe sepsis, the incidence rate was substantially higher in COVID-19 compared to these diseases. However, if you look at non-infectious diseases such as acute renal failure, patients with heart failure, patients with injuries, non-cranial or cranial injuries, for many of them, the incidence rate was higher. And then, interestingly, what we found was that when we looked at patients during the first and the second wave with COVID-19, we found a much higher incidence rate during the second wave compared to the first wave. We saw tendencies of that in the non-COVID-19 patient population as well, and I will try to briefly go through potential reasons for that. I will just highlight these. I will go through quite briefly, but I will just highlight that basically what we found, which is in line with previous Swedish research on ventilator-associated pneumonias or infections, is that the most common agent detected was Staph aureus in this case. We have here stratified the etiological agents by time of mechanical ventilation. So, the upper graphs here represent the microbiological findings during the first five days of mechanical ventilation, and the lower represents more than five days. You can have a look in the manuscript if you want to have a more detailed description of the etiological agents here. I will more try to spend the last few minutes on trying to explain potential reasons for why we saw an increased incidence rate during the second wave compared to the first wave. One thing that you can appreciate from this table, which is a supplementary table from the manuscript, is that the case mix with regards to age and also sex was different during the second wave compared to the first wave. So, we had lots more elderly patients during the second wave, more often female sex, less comorbidities. Also, if you look into the description of the ICU state, this is highly interesting, and this is still, I would say, an open question of the importance of this, is that steroid treatment before ICU admission was provided to almost 90 percent of patients during the second wave, compared to roughly one-fourth during the first wave. What you also see from this is that the mortality rates, both looking at the ICU mortality rates and the 30-day all-cause mortality rates, was substantially higher during the second wave compared to the first wave. We couldn't find any substantial difference or any difference at all, actually, in how much lower respiratory tract cultures were performed, which indicates, to some extent, that the results we found do not seem to be due to differential testing strategies during the different waves. And when then moving on and looking, using regression modelling strategies and CSHR, in this case, is an abbreviation for cause-specific hazards regression, and SHR is, in this case, sub-distribution hazards regression, what we could find that when we adjusted for age, sex, comorbidities, obesity, the use of prone positioning, and the use of steroid treatment, you still had a higher incidence or a higher risk of ventilator-associated infections during the second wave compared to the first wave in the COVID population. That could not be observed. I would say it couldn't be excluded as well in the non-COVID population. Just for your reference, I have updated this analysis now with new population-based data because this study was looking into the Kaliinsky University Hospital, which is a tertiary hospital here in Stockholm, and now I'm looking at the longer time period in more ICU settings in Stockholm, and the results are basically exactly the same with these differences observed between the waves, and I would say that we fully do not understand exactly why this is the case, how much of this is attributed to differences in case mixes and how much is attributed to the interventions provided for COVID-19, and that we need to understand much better, I would say. So, in conclusion, mechanical ventilation durations, they are exceptionally long in COVID-19 patients, and that has implications, of course, for the patient both in the acute and the post-acute and long-term period, and the incidence proportions was significantly higher. However, if you took time into account, risk time into account there was not an increased incidence of these infections. We observed differences between the first and the second wave of the COVID-19 pandemic in COVID-19 patients. And I think it's important to try to understand that both with regards to the specific interventions provided and with regards to structural aspects of the provided healthcare. And with that, I would like to thank you so much for giving me the opportunity to present this. And please feel free to ask any questions or contact me if any ideas or potential collaborations. Thank you. And actually, now I will hand over to the next speaker, which is Dr. Teunman. So, thank you so much. Thank you, Pontus. And thanks for the invitation. My name is Pieter Teunman. And I will present a study we did with a live study group. And I would like to mention that we started this group because we would like by free education and research, improve patient care using point-of-care ultrasound, but also increase healthcare workers' job satisfaction. I think everybody who started using point-of-care ultrasound is very enthusiastic about it. So, we also teach our nurses. And I think there's also an important aspect of point-of-care ultrasound. But let's talk about one of our projects, which was published last year in Critical Care Medicine. And it's called Extended Lung Ultrasound to Differentiate Between Pneumonia and Atelectasis in Critically Ill Patients. So, the outline of my talk is a bit like the outline of a manuscript. I will give you some background about the manuscript. And in the background, I will focus on the lung ultrasound and the techniques we used. So, about the color Doppler we used and the dynamic air bronchogram. And I don't want to go too much into detail into the clinical scores we compared ultrasound with. Like, for example, the clinical pulmonary infection score or the simplified version of it. Because I think most of you know that those scores had quite a low accuracy with a sensitivity of around 60 percent and a specificity a bit higher, around 65 percent. So, I consider that a bit of knowledge already present in the audience. Then we go to the methods of the study. We will discuss, of course, the results, but also later on go into how you can use the results of the study in your clinical practice. We will discuss some pearls and pitfalls. And at the end, hopefully, you have a lot of questions for us, because it's always nice to discuss the research and the topic. But let's start with a poll. And the question is, do you use lung ultrasound to differentiate between pneumonia and atelectasis? So, please click on yes or no. And let's see how it's used currently. So, already one-third of you is using lung ultrasound to differentiate between pneumonia and atelectasis. So, I think that's really a great amount. And maybe after this presentation, maybe after this presentation, we can repeat the poll and see if these numbers will change. So, let's move on to the next slide. So, as very nicely discussed by Dr. Hedberg in the previous presentation, there is in patients on the ICU and high prevalence of consolidation. So, both the high prevalence of pneumonia, so ventilator-associated pneumonia, hospital-acquired pneumonia, and even patients coming into the ICU with a community-acquired pneumonia. But also, a lot of patients have atelectasis because of the treatment in the ICU. So, because of the medical ventilation, because of sedation, paralysis, but also abdominal problems, which make breathing more difficult and changing chest wall compliance. So, confronted with a consolidation on the chest radiograph, or with clinical findings, often the stethoscope is used. And I think, yeah, we should acknowledge nowadays that it's old-fashioned medicine, and we should relegate the stethoscope to the museum. For example, a recent meta-analysis showed that in the ICU, the sensitivity of the stethoscope for pneumonia is around 50% or lower. So, I think you can really use it to make any clinical judgments. And I think the same applies for chest radiograph. It has been nicely shown in critical care medicine, I think in around 2017, also in a meta-analysis where lung ultrasound and chest X-ray were compared with CT scan, and also the sensitivity of the chest radiograph was around 50%. Unfortunately, the same applies a bit to the inflammatory markers, because there are a lot of reasons to have increased levels of your inflammatory markers. So, you see a reactive protein is often increased, and if you have very high values, it's of course suspected to be there an infection, but you don't know where. But also the newer procalcitinin can be used really to start your antibiotics. So, we need something else, and I think this is really important for different reasons. First, it has treatment implications. So, if you have a pneumonia, you want to start first fast with your antibiotics, because we know that pneumonia is associated with increased mortality, but also morbidity and increased length of stay. But on the other hand, we don't want to overuse the antibiotics, because antibiotic stewardship is very important with all the resistant bugs in current medicine. So, we need to guide it carefully. And maybe ultrasound can be a good tool in this aspect, and I think most of you know, and because already one third of you uses this for differentiating between pneumonia and atelectasis, that it's an accurate tool to diagnose pulmonary conditions. However, there are some important limitations. For example, the most used protocols are the Blue Protocol, which was published by Daniel Liechtenstein from Paris in just around 2008. He has a very nice flowchart, but it was a study on the emergency department, and in this flowchart, there are no patients with atelectasis, and those patients with multiple diagnoses, like ICU patients, were excluded. And the other most often used protocol is the Lung Aeration Score, or the Lung Ultrascore, which was firstly published, I think, by Bournemouth in 2010, and what you do, you calculate an aeration score, but it's used for monitoring and see the effect of treatment, also, for example, for ventilator-associated pneumonia, so it's also different from making a diagnosis. And there have been studies with Culler-Doppler imaging, I will come back to that later, and dynamic air bongogram, but they have never been combined in an ICU population, so that is of interest. When we're talking about the Culler-Doppler, it has been published in the 90s, first mostly for pulmonary tumours, but later the same authors used it also in a consolidation in 2000, and what they saw was that there was pulmonary fascial constriction with loss of flow in a consolidation in the lung, and they said it was associated with pneumonia, and because of the loss of aeration, you can see, of course, normally the lung completely blocks your ultrasound signal, but if you have loss of aeration, the ultrasound signal can be transmitted and you can see the flow, and it had quite a high sensitivity for pneumonia with low specificity. So I will show you here two clips of an image, so on the left side, let's start there first, and the flow, you need to adjust the speed of the flow, and then what you see is like a tree like, so we have the consolidation, even through the consolidation, you can see the heart, and with the heart beating, you see a tree-like pulsatile flow in the consolidation, and if you go to the other image, I click this, I stop this one, because for better resolution of this one, this one is a bit more difficult, but at one point, you also will see here in orange, the pulsatile flow, which has a tree-like aspect, so in the beginning, it's a bit difficult, you think maybe there is something here, and later on, you see in orange, there's clearly pulsatile flow, but this is quite sensitive for pneumonia, but not specific. Let me move on to the next slide, hopefully. Sorry, I have to move on back, and then looking at the dynamic air bronchogram, this has also been described by Daniel Liechtenstein, and what it is, is that normally in a consolidation, you have dots of white, which is air trapped in the lung, but if the airways are open, and there is fluid or mucus in the airways, those bubbles of air, so those white dots, can move up and down with respiration, and if you see that, and once you've seen it, it's quite easy to recognise, it's a very high specificity for pneumonia, but a low sensitivity. And in the next clips, I will also show you the dynamic air bronchogram, so first, start on the left, so you see here, the white dots moving up and down with respiration, so that means there's air moving up and down in the airways, in the fluids, and if we go to the other clip, then you will also see at one point, bubbles going up and down here, so it's also positive, so you have here, the spleen or the liver, you have the consolidation, and in the consolidation, you see the flow, and here you see some small neural fusion, and here the diaphragm. If you move on to the next slides, hopefully, we go back one, then let's discuss the study we did. So what we did, was that we included patients, adult patients in our ICU, with a consolidation on the chest radiograph, ultrasound had to be performed within 24 hours, by one of the researchers, and we had as a final diagnosis, a composite reference standard, meaning that the doctor used all this information, except lung ultrasound, to make a diagnosis, so it included lab values, it included cultures, it included CT scans, if available, and the chest x-ray, in the 72 hours after the chest x-ray, so I think that's one of the best ways to look at it, and if a patient was diagnosed by ultrasound, on one side with pneumonia, and on the other side, potentially atelectasis, it was classified as pneumonia, because that prevails. So we had 128 patients, eight patients were excluded, for example, because of they had ARDS, ultrasound was not feasible, because dressings, for example, and two patients withdraw consent, and we also excluded COVID-19 patients, because it was, we studied before COVID-19, and we had a break, and the aim of the study was to determine the diagnostic accuracy of extended lung ultrasound assessment, to differentiate between atelectasis and pneumonia, and extended means that we used color Doppler, and dynamic airborne program, in addition to the normal ultrasound, and our second aim was to compare this approach to other scores, like the simplified CPS score, the LUSCPS, and the LUSCPS, you exclude the chest x-ray, you include lung ultrasound findings, and then they used the Doppler findings, and also we compared it with the previously mentioned blue protocol. Yeah, this is also, as in the previous presentation, quite a big table, and you don't have to have a complete overview of it, but what it shows is that there are several differences in clinical characteristics between patients with atelectasis and with pneumonia, so this is the overall, on the left, the column, and then pneumonia and atelectasis, and mostly the difference were notably in inflammatory markers, for example, C-reactive protein was higher in a patient with a pneumonia, which can be expected, but as you can see, there's quite a lot of large overlap, and more often a positive culture, of course, but also there were differences in vital signs, for example, the PF ratio was lower in the pneumonia group, as can maybe be expected, because we see more flow in a consolidation, so less hypophagic constriction. If we then go to the, to how the population distribution was, so we had 120 patients, and then we first looked at the static airborne program, or dynamic airborne program, and then if we look at the left, there were 96 people who had a static airborne program, and almost half there was absent flow, and if flow was absent, they were considered to have atelectasis. On the other hand, if they had a static airborne program with present flow, we also included the leucippus, so we included some clinical available parameters, like fever, like if there was a sputum available, and if it was more than four, which was because of in the previous study, then it was considered pneumonia, and if the leucippus was low, then it was considered atelectasis, and on the other hand, if there was a dynamic airborne program, then you were, it was clear they were diagnosed with pneumonia, and this flow chart, or this decision tree, had a diagnostic accuracy of 86 percent, which is comparable almost to the original blue protocol of Liechtenstein. If we then go into more detail in the diagnostic accuracy of the different aspects of this decision tree, then we first look at the dynamic airborne program, and what you can see is that, as was already known from previous studies, that the specificity was very good, but the sensitivity was quite low, meaning that your positive predictive value was high, and your negative predictive value was much lower, and maybe it's preferable to look at likelihood ratios, because they are not affected by your prevalence of the disease. Of course, this was a bit of a strange cohort of patients, because either way, they had atelectasis or pneumonia, which you can also see with the likelihood ratios, that it has clear effect on them, so it's quite good. Then looking at the color doppler, you see it's a bit the other way around, so sensitivity is high, specificity is lower, affecting the positive and negative predictive value, and then you see that the decision tree has the highest accuracy, so 86 with sensitivity are quite good, and then the Leucistis has a lower score, with a sensitivity of around 70, and specificity around 80, and the blue protocol doesn't include atelectasis, so you can include pneumonia, but you cannot include atelectasis, so we can't really calculate with that. So, how do we interpret these results? I think, first of all, some things are in line with previous literature, so the color doppler is highly sensitive, we already knew, and the dynamic air bronchogram is highly specific, and it's also important to validate those results, and they are also in line with two recent studies, one published by Duro in the European Journal of Anesthesiology in cardiac surgery patients, and one recently published this year in Annals of Intensive Care by the same group from France in VA ECMO patients, so that's comparable, but there also are new findings that the combination of those signs, so of color doppler and dynamic air bronchogram, has a very high diagnostic accuracy, and I think that's really good, and especially because it outperforms other scores used currently, and I think thereby ultrasound is a very valuable additional tool to your history, physical examination, and lab findings, so you never use it alone, you use it in concert with your other findings. Of course, our study had some limitations, the operators performing the lung ultrasound were experienced at one point, so we had one medical student doing the examination, so we teached her 40 exams supervised, and afterwards she did it on her own, so not really experienced, but she was well trained, and the other operators had around two years of ultrasound experience or more. A limitation can be considered the composite reference test, although I not really agree, it's always that reviewers say it's a limitation, but I think it's the best we have. The gold standard will be pathological examination, but you can do that in a live patient, so I think this is the best using all the information you have, and it's used in most of the studies, also the previous studies I mentioned by Duro and that group, so I think on the other hand it's maybe even a strong point, and of course the underlying physiology of the ultrasound findings, especially for the flow, we can't fully comprehend, so we think there is flow in the pneumonia because the airways are open and there's inflammation, so the hypoxic facial constriction is less, but we don't know for sure, and not all patients of course with absence of flow have an ectalectasis, and we had one patient with a dynamic airborne neuroma who was diagnosed with something else, so always remember not to say that the results are absolute. So for example, I think this is a case we all see in our ICU, so it's for example a 50 year old man, previously healthy, and he was admitted to the ICU after a neurotrauma, so a car versus a bicycle, or current days versus an e-bike, he was operated, he's recovering, but he has some weaning difficulties, so for example on the ventilator for two weeks, and then his C-erective protein starts to rise, his white blood cells start to rise, but not really dramatically, he changes his central line, you take cultures, but they remain negative, and then because of you change the line, you have a chest radiograph, and it reveals bilateral consolidations, but they were already a bit present, so you don't know if they really changed, and then in those patients, I think it's really good to perform the lung ultrasound to look for flow and for the dynamic airborne program, and then based on that findings, you can start your antibiotics, or maybe withhold them, of course also taking the clinical picture into account, if the patient's really septic, of course you won't withhold antibiotics, but maybe you should look further for another focus, and then in this patient after three days, the sputum cultures were positive with the pseudomonas, and you can also use it with the current studies in patients after cardiac surgery, or patients on VA ECMO, for example. So in summary, the pearls and pitfalls, I think the pearls are that the dynamic bronchogram is highly specific, flow is highly sensitive, use it always in the clinical context, as I previously mentioned, and understand the underlying concepts of lung ultrasound, of course, and also the limitations of lung ultrasound. Things that make it difficult are, for example, large tidal volumes, because you need to have the consolidation when the patient is breathing in and out in one view, so if he's breathing heavily, sometimes the whole consolidation is moving out of your view, so you can't really look for flow and for the dynamic airborne program. You need a machine, so an ultrasound machine with some computational power, so you need to, of course, can use flow, so with some of the, for example, now the handheld devices, that's difficult, but they are becoming better and better and cheaper and cheaper, so I think that will be solved, and you have to adjust the flow of the Doppler to quite a low flow, because the flow in the consolidation is lower, so to 0.25 meters per second. So let's finish with a poll, and then we can go to the questions, so after this talk, would you consider using lung ultrasound, including Doppler and dynamic airborne program to diagnose pneumonia or to differentiate between pneumonia and atelectasis? Now, this is an overwhelming result, so that's very nice, and also for me, if you have any questions about how to do it, please contact me by email, and I can explain some of the things which maybe are not really clear from the presentation or from the manuscript. I'm happy to do so, and I'm also happy to take any questions from the audience. Thank you very much both of you for the very interesting presentations. I'm Thomas Zagmani and I've managed to catch up with the webcast. So I will take over moderating the question and answer portion of the webcast. The first one, I think, if I may ask from you, Dr. Toeman, first, is that how many hours do you reckon a provider should put in to learn the technique? So how difficult it is? It's not really difficult. Like I mentioned, we taught one of our medical students with 40 supervised examinations. And most of the research we do is by training students and they start without any ultrasound experience. So I think that also makes a difference. But the problem is you have to have seen it a couple of times, of course. And if you're in doubt, you need a colleague who you can ask, what do you think? And that's in the beginning a bit of a problem. But luckily now on the Internet, you can find a lot of images from lung ultrasound and that will really help. So I think if you are already an ultrasound quite experienced, I think you can learn it in, for example, 10 exams or something. But you need to look on the Internet for some examples because it's pattern recognition. So you have to have seen it a couple of times, especially the right clip I showed. It was a bit more difficult than the left because it's really clear. Then you see it. And sometimes you're a bit in doubt. Thank you very much. If I if I can just expand on that. Do you think that this this technique could be could be used in in the setting of tele ICU where a remote provider tries to to direct somebody at the bedside to devaluate the lungs? Yeah, I think so. And as I started with, we teach our nurses now lung ultrasound and also some basic cardiac ultrasound. And that would really go with the with with your question that if someone else can do the lung ultrasound, you can have someone outside of the room or in another hospital look at the images. And, for example, also in the COVID-19 areas, we did some ultrasounds on the ward. And if one of the residents was in doubt, we looked at the images. If we're not we're not on the same ward remotes. So I think it definitely can be used that way. Yeah. Thank you very much. The next question goes to Dr. Hedberg. And. You know, listening to the to the presentation and reading the paper, is it. Do you think that the the rate is of of ventilator associated infectious events is higher simply because the COVID-19 patients were ventilated for longer so they had more time to develop these these secondary infections? Or do you think there is something else as well? Thank you so much for that question, Thomas. I think I think that's a great question. And that's, of course, something that all of us, I would say so. So what you do see, if you look at the absolute numbers or the absolute occurrence, it's substantially higher in COVID-19 patients compared to to basically most of the investigated non-COVID-19 patient populations. However, if you take time into account, risk time into account, that difference is kind of disappearing. And I mean, what we see, interestingly enough, in our data is, as I mentioned during the presentation, is the occurrence of these infections differing between the different waves. I mean, of course, one question is how much are potentially the pharmaceutical and non-pharmaceutical interventions we provide to COVID-19 patients affecting this. For example, I have seen papers on on dexamethasone increasing the risk of ventilator associated pneumonia in COVID patients. There has been previous studies before the COVID pandemic, if I'm if I'm not remembering this wrong, that looked at prone positioning, for example, and the risk of developing pneumonia in the ventilator. So to be honest, I cannot give you a very definite answer on this. But yeah, I mean, it, of course, has important implications that these patients are treated so long in the ventilator. I could just briefly mention that now when I look on the updated preliminary data, we do not see any difference in mortality, both acute mortality and long term mortality in the COVID patients that are diagnosed, according to our criteria with these infections compared to patients without the infection, which is quite interesting. Thank you. And again, just a follow on question on that. This is this is my own question because I'm interested in this. Did you use any biomarkers to try to quantify or try to use to predict the development of these infections? No, that's a great question. We have done that for when it comes to bacterial community acquired pneumonia in COVID patients. We have done exactly that, which many groups have done, actually. And our findings there is basically in line with other studies. In this case, that is unfortunately precluded by the fact that at least in a Stockholm or a Swedish setting, the ICU electronic health records are distinct systems compared to the other hospital EHR systems. And that means that we actually did not have access to the entire ICU trajectory of patients. So basically, we did not have access on, for example, inflammatory biomarkers and their dynamics during the ICU. We only had access to that during the first 24 hours of the ICU stay, as well as the entire hospitalisation prior to that. So, yeah, we simply couldn't do that kind of prediction in this case. Thank you. That's very interesting. Going back to the lung ultrasound study, I must say that I'm convinced and I will try to use this. And obviously it has got the many benefits that it is available next to the bedside. What would a series of CT scans add to this question, if anything, in your opinion? It's a bit of a difficult question. I think CT scans would not add really much to it. But maybe I have a bit of a cognitive bias. I do a lot of ultrasound research. A lot of times we see also, for example, the COVID pandemic, we did a lot of CT scans for patients with COVID-19 to see if they had any secondary pulmonary infection. And we also did a lot of ultrasounds. And the scans never pointed CT scans to a secondary infection. Never. Because there were always consolidations and difficulties to see. So I think maybe a CT scan for pneumonia is a bit overrated, except for things like aspergillus, which you can see with ultrasound. And also, of course, you can see things in the lungs, so abscesses or things like that. So I think if ultrasound doesn't help and you're in doubt what's the problem, then you should, if the patient is not improving, go further and use a CT scan. But I think to start with a lung ultrasound is perfect and use it in this aspect. And only if the patient is not improving or you can't figure out what it is. If you have an immunocompromised patient, you think of aspergillus, things like that, or an ILD, you should step up to a CT scan, I think. Thank you very much. And I think you wanted to ask a question regarding this aspergillus about our co-presenter. Yeah, because for the COVID-19 patients, we were quite afraid of the Kappa, the COVID-associated pulmonary aspergillus. And I was wondering if you also looked at the incidence of that in your patients and if it was different with non-COVID and also if it affected the outcome, any results? I mean, that's an extremely good question and an important question. However, what I do find in the data is that in the Swedish setting, that is not being tested very frequently, actually. And that has been the case in COVID-19 patients as well. I think I've seen previous proportions of around 10% or something of COVID-19 patients in the ICU settings. But I can unfortunately not expand on that on Swedish data because our testing is often lacking. But it's a very important question, yes. If I may add, again, just our own sort of local Welsh experiences that aspergillus, although we tested for it, the incidence was extremely low. I think we only had one patient out of the 300 that we have treated in our hospital who had that. Thank you. Thank you. Now I'm just saying very preliminary data. Actually, when I looked at that, I could confirm that what Thomas is saying was what I saw in the data as well. That out of the patients where this testing was performed, the positivity was actually very low. I cannot give you a number, but it was completely in line with what Dr. Thomas just said. Thank you very much. Thank you both of you for the wonderful presentations and for the discussion. This concludes our Q&A session. Thank you very much for joining us today. Please do send us your feedback from the email which will be sent out after this webcast. We would like to invite you to our next Journal Club Critical Care Medicine webcast on the Thursday, June 23rd, 2022, 1pm Central Time. Please stay tuned. Thank you and stay safe. Bye. Thank you. Bye bye.

Critical Care Medicine

ventilator-associated infections

COVID-19

ICU

mechanical ventilation

incidence rates

lung ultrasound

pneumonia

atelectasis