Thank you very much, Ashish, and thank you all for sticking around to hear me. So this talk is going to be on pulmonary physiology at the bedside, but I'll try to weave in a bit of my own personal research journey. So these are my disclosures. And it all started for me at the bedside. So this was a patient that we took care of in my fellowship, multiple trauma, ARDS, very seriously injured, and we found him on these ventilator settings in the morning, so CMV, lowish tidal volume. I don't know how they got to that number exactly, but OK, FiO2 of 1 and a PEEP of 13. And my division director at the time said, why don't we call Steve Loring? He's the smartest guy around here. So we called Steve Loring, and I'll get back to him in a second. The question was, how much PEEP to use? And there's multiple studies in the literature then and subsequently, all of which are very confusing. So this is the ARDSNet alveoli trial. This is the Canadian trial, which was published shortly after. Both were completely flat negative, showed no benefit to higher versus lower levels of PEEP. There's a couple other trials in the literature. So this is the EXPRESS trial from France. Here they tried to be a little smarter. They set the PEEP to push the plateau pressure to 28, a strategy which doesn't make a lot of sense to me, because that means that the most compliant lungs and the healthiest lungs get the highest levels of PEEP. And then this is the ART trial recently published out of Brazil, which actually showed injury from high PEEP. But if you dive into it, it wasn't actually from the higher PEEP. It was from the very aggressive recruitment maneuvers that were delivered with the trial. So a confusing literature and difficult to know what to do. Now why is this literature confusing? Because if you take two research strategies, so a control arm with a lower level of PEEP and a intervention arm with a higher level of PEEP, and you include all comer patients in the trial, you're going to essentially be randomizing two separate groups of patients. You're going to be randomizing patients who are non-recruitable, in whom if you increase the PEEP, you increase the plateau pressure and you increase the driving pressure. Those patients are going to be injured by the PEEP. On the other hand, recruitable patients, you increase the PEEP and the plateau pressure stays the same or goes up somewhat, but the driving pressure will go down. Those are recruitable patients, and in them the benefit of high PEEP will outweigh the harm. The issue is if we include all these patients in our trial, some are injured, some are harmed, and the sum total is no benefit. So we need to be smarter about designing our trials and figuring out who's recruitable and who's not. And there's been a lot of ways in the literature described to do that, but really we have focused on titrating mechanical ventilation to transpulmonary pressure. So when we measure pressures on our ventilator, we're measuring them here at the airway. But airway pressure is not the pressure that distends the lungs. The pressure that distends the lungs is the transpulmonary pressure, and that is the airway pressure minus the pleural pressure. Think of the pleural pressure as the weight of the chest wall pushing down on the lungs. You have to overcome that before you even begin to open up the lungs. And we showed way back in 2006 that the transpulmonary pressure varies widely from airway pressure. So at any given airway pressure, you can have a very different transpulmonary pressure. And if you don't measure the pleural pressure, you can't know what your transpulmonary pressure is. Now the real way to measure pleural pressure is to stick a pressure measuring wafer in the chest wall, transduce that, but it's not really realistic in live patients. So we substitute that with the esophageal balloon catheter. This is a catheter that we place through the esophagus. It sits behind the heart, and it measures the pressure in the esophagus. So the esophagus is an extra pleural organ, more or less at the center of the chest. So it's a good substitute for the pleural pressure. And I mentioned Steve Loring. This is Steve in action here, one of my great mentors. And this is Ray Ritz, a respiratory therapist from whom I learned pretty much everything I know. A great example for this society of how multidisciplinary critical care works. So these two were really instrumental in everything to follow. So the research question for us is if we measure pressure here, and then we calculate the transpulmonary pressure, and then we match the PEEP to maintain a positive transpulmonary pressure, can we prevent that collapse of the lung at end expiration? So would that be the appropriate way to set the PEEP, by the transpulmonary pressure? And there's some evidence to show that this may actually work in actual lungs. So this is an interesting study by Lorenzo Berra at MGH. He measured the transpulmonary pressure here, and he measured collapse on EIT here. And you can see that below a certain level of PEEP, we begin to see collapse of the lung. That by coincidence is the PEEP level where the transpulmonary pressure is zero. So if I maintain a PEEP level of transpulmonary pressure greater than zero, I may be preventing this collapse of the lung. So there's imaging evidence to support this approach. So back to our patient, Steve came and we made measurements. We measured the airway pressure, we measured the pressure in the esophagus, and then our computer integrated that and measured transpulmonary pressure. So this patient had a PEEP of 13 and an airway plateau pressure of 40. The esophageal pressure was very high due to his injuries. And the transpulmonary pressure was negative eight at end expiration. So there wasn't an actual vacuum in the lungs. There was no negative pressure. It just meant that the weight of the chest wall was higher than the pressure in the lung. And that led to collapse. So what did we do? We increased the PEEP from 13 to 26. That increased the plateau pressure from 40 to 46. How many people put a PEEP of 26 routinely? I don't see any hands. But this is what that looked like. So the transpulmonary pressure at end expiration was now four. So these airway pressures are scary. These transpulmonary pressures are entirely normal. And we wouldn't know that unless we measured. If you calculate the driving pressure, so it went down from 27 to 20. So this patient clearly fell into that recruitable category. And the driving pressure of 20 is not great, but it's sure better than 27, right? So together, Steve and I designed this first randomized control trial of using esophageal guided PEEP versus empiric low PEEP. That trial just scraped statistical significance. We were lucky to publish it in the New England Journal. And I think the reason wasn't the significance, but the novelty. And it was entirely new at the time. So that was the EPVENT1 trial showing a superiority of esophageal pressure guided measure PEEP versus standard low PEEP. We followed that with the EPVENT2 trial. Here we compared esophageal guided PEEP versus empiric high PEEP. And unsurprisingly, the follow-on trial was negative. And so far, so bad. This is a typical story in critical care medicine. But when you do a trial, you really collect a lot of data and you have to look behind the bottom line results. So looking behind the bottom line results, we had for various reasons decided to compare to empiric high PEEP. And at the time we designed the trial, it seemed like that was the direction people were going. So these were the levels of PEEP delivered. Even at an FiO2 of 0.4, people were getting 10 to 18 of PEEP. When you compare that to what the LungSafe study shows is actually given in the community, it's nothing like what people give in the community. Nobody gives these levels of empiric high PEEP. So we made a mistake in our control arm. And because we made that mistake, sorry, I skipped ahead. Because we made that mistake, there was no separation in transpulmonary pressure between the two groups. So we took a look at the data and we found that if we re-divided the patients based on whether they had a transpulmonary pressure around 0 or it was higher or lower than 0, we found that those around 0 had a better mortality. So the goal of the intervention seemed to work. It's just that we had too many crossovers in our two groups. So this seemed to support actually using PEEP set to a transpulmonary pressure near 0. We looked at it a little deeper. So we then looked at the combination of a transpulmonary pressure at or near 0 and a driving pressure less than 12. So if you had a driving pressure less than 12, this was your mortality. If you had a transpulmonary pressure around 0, this was your mortality. If you had neither, you did much worse. If you had both, you did much better. So here we begin to see in the data the opening for an integrated strategy of both PEEP set to a transpulmonary pressure of 0 and driving pressure limitation. And that that is associated with improved mortality. So we wrote a grant around that. And we were lucky to get it funded by NHLBI. My co-investigators, Elias Beydorf-Cassis, Jeremy Beichler, and Tim Houle. So there's a data coordinating center and a clinical coordinating center. This will be a 1,100 patient trial, 20 centers around the United States. And if anybody here would like to talk about joining, please let me know. In the precision ventilation arm, we will set PEEP to a transpulmonary pressure of 0. And then we will reduce tidal volume to get a driving pressure of 12 or less. And there's various ways in the protocol to achieve that. And we know that not everybody is going to achieve this driving pressure. But we will try to get there on everybody. In the intervention arm, instead of specifying high or low PEEP, we're calling it guided usual care. And we're giving very broad ranges to PEEP. And these broad ranges encompass the levels of PEEP seen in multiple clinical trials. We went back and looked at what was seen. And these ranges basically allow clinicians to do pretty much anything they feel is reasonable. Full tidal volume will be 6 to 8 mLs and not looking at driving pressure. So we're very excited to start this. I have to thank my lung injury group, Max Schaefer, Elias, Jeremy, Brian O'Gara here in the audience, and Valerie Banner-Goodspeed, our research director, without whom nothing is possible. So I'd like to thank you for allowing me to share my journey a little bit. I'd like to thank the Society for giving me this award. I accept it on behalf of all of these people who really made me what I am. Thank you very much. And these are my lessons for future investigators, okay? Research begins at the bedside. Mentorship that I receive plenty of is important. Build your team. Try and give back and mostly have fun. Thank you.

pulmonary physiology

mechanical ventilation

ARDS

transpulmonary pressure

personalized ventilation