Thanks. It's still Heise, but that's good. So how do we get myself up there? Okay, good. Say thank you. I'm Bob Heise, University of Michigan. And so let's see. We're gonna click. Where do we click? Just regular? On the left. Oh, there we go. Okay. I can't go back. I can't go back. So here are my disclosures. I get up-to-date royalties. I write for them. I'm a calculation. I have a Springer textbook. Getting a little old now, but maybe we'll revise that. I'm on the DSMB of a study for Stimdia, and I have a part of our APS consortium, which is U01 grant. It's a phenotyping consortium looking at ARDS, pneumonia, and sepsis. And we're just getting our situation organized. So what's the current word on PEEP? If you want to know the current word, here's a network meta-analysis from the Blue Journal two years ago now, and trying to look at the evidence. And it says, geez, look, here's probably the best evidence we have that maybe high PEEP without a recruitment maneuver is a good idea. Here's the risk ratio, so go and do it, right? Well, the thing is, we've known for some period of time, just going back to Gattinoni's work, New England Journal now, more than 15 years ago, that not all patients recruit equally. So they just say, well, high PEEP probably works, so here's the network meta-analysis. It's probably insufficient. You can see here on the left, at five, and then on the right, 45 centimeters of water PEEP. Above here, you have a low recruitment situation. You don't see much change there. Yet below, the recruiters, you can see that a lot of lung has been opened. This is actually a, you may remember, the Loves and the Express high PEEP, low PEEP trials that were in JAMA 2008. Negative trials, but if you if you tease them out by the PEEP responders, you saw a mortality benefit. So clearly, a broad brush of one-size-fits-all is with PEEP, and recruitment is probably insufficient. This is an editorial from the Junkman article I talked about a little bit ago, but in bold, it says, maybe it's time to step back from further, ever-larger PEEP recruitment trials, and says strive for unified consensus on how to best define recruitment, how to successfully predict it. After all, we have a bunch of new toys, and the toy I'm going to tell you about is electrical impedance tomography. And what that is, is actually a belt with 16 to 32 electrodes placed around the chest, around T4, and the current conductivity of the current relates to images that are generated. You have high impedance and low impedance areas, and because high tissue resistivity leaves a low signal, and air acts as electrical resistor, you have high impedance. So you can contrast air-filled areas, if you will, with water-filled areas, like, oh, I don't know, collapsed lung. And so you generate images from this belt. You can see here on the right, this is just how the display, if you will, from one of those electrodes, but in fact, the electrode belt covers the thorax at that level. So what the setup looks like when you use it is a belt, as I said. You've got the patient hooked up to the ventilator, and a separate machine is employed to give you this data. Now you can do a lot of things with this. This is a review that Victor Jimenez and I wrote with some colleagues two years ago now in critical care medicine. The application I'm going to talk to you about is PEEP, but this has been employed pre-invasive macelebration, if you will, ECMO, and weaning situations, but the application I'll tell you about is with PEEP titration. Here is just a snapshot I took the other day on the plane right here, actually, about from PubMed, and you can see literally an exponential increase in publications related to electrical impedance tomography, dating to just now. So the interest in this is burgeoning, the publications are burgeoning, and maybe it's a coming thing. And there's a bunch of tools that we consider when it comes to the issue of PEEP titration in ARDS. I'll focus today, though, on only one aspect. That's called the over-descent and collapse intercept. Through that impedance issue and resistivity issue, you get areas that are collapsed, areas that are over-descended, and there is the Goldilocks just right. And so this notion of this trade-off between avoiding over-descension and yet opening as much of the lung is called this over-descension collapse intercept. I'll show you that graphically here in just a second. And this is a publication that gives you that sense as you look through slices of lung, and you can see that intercept here. This is from our review. The diagram to the left, this is perhaps a little more explicit. You can see a downward PEEP titration, like a staircase thing. You start at a high PEEP, come down. You can see the images here where you avoid over-descension here at the high PEEPs, and you avoid collapse at the low PEEPs. You find the optimal trade-off with that intercept. So this patient, for example, would be optimized at the intercept about PEEP of 14, whereas this patient on the left is recruitable. The patient here on the right, certainly less so. And you can see here around a PEEP of 6. So now this is a research tool at the present time. The EIT devices are commercially available. Only one is currently approved by the FDA. That's the Tempel device, which we're fortunate to now have at my institution. Right now the intercept notion is not FDA-approved, but this is the thing that's most in play academically. And I know that, as I understand it, they're pursuing that approval. So a little bit of this is research, and a little bit of this is clinical. So the biggest, one of the things about personalized medicine is, if you will, upscaling it to do clinical trials to show efficacy. And you can imagine, my little trial was quite small. The biggest trial we have of patients getting electropenis tomography was done by Junkman. It was in the Blue Journal last year. It was 108 patients, and there wasn't really not a clinical outcome. They just compared EIT titration to best compliance. So if you were in here by driving pressure, for example, if you maximize driving pressure, driving pressure being a surrogate for recruitment, you can come up with the best PEEP on that basis. And they divided these 108 COVID-19 patients into low, medium, and high recruiting subgroups. And interestingly, the severity ARDS Berlin criteria, which relates to mild, moderate, severe, based on PDF ratio, i.e. severity of hypoxemia, didn't correlate with recruitability in this data set. Now you could argue, as we did a few years ago, maybe COVID-19 is somehow different. Let's table that for one second. But you do see that this was scalable, at least for in a range where you could have a reasonably sized clinical trial, multiple centers admittedly. And what they tended to find, though, is that actually the EIT PEEP compared to the best compliance PEEP was a little higher than they found with best compliance. So tuck that in the back of your brain, because the issue is, what are we seeing? We have clinical tools we've conventionally used to to look at best PEEP. Best compliance is certainly one justifiable approach, but EIT doesn't necessarily give you the same optimum number. So let me tell you about our research that we did a little trial a few years ago when we had our machine. And to do that, I want to tell a little bit about mechanical power. Everyone knows about low tidal volume ventilation, low plateau pressure. These are static variables and may be more important to look at dynamic variables, how much energy the lung is absorbing with the ventilator and correlating that to injury. And so we've seen an evolution of our concerns from low tidal volume, low plateaus, high PEEPs to low driving pressure to low ventilating power and driving power. And in fact, this is how you calculate mechanical power. I won't delve into it too much with you, but so hasty to say this is a variable in play. It has been looked at writ large. This is the Toronto I-Corps registry of over, I believe, 6,000 patients. And you can see that I highlighted only the mechanical power part, but the time, the number of days above a threshold of driving pressure, that's your right, driving pressure of 15, was associated with mortality. And also the days of mechanical power above 17 joules or some joules a minute associated with mortality. So in other words, you can see in a large database, the more energy the lung is absorbed through the ventilator, the worse it gets. So this could be a useful surrogate endpoint as we try to study EIT or other modalities where it's difficult to be scalable to the size of a large, I don't know, say 1,000 patient RCT. So what we did at my place was a, and this was actually funded by the Chest Foundation, thanks to them, this was a prospective randomized crossover trial where at the time of randomization, we either went with EIT PEEP titration or essentially the LUVs, the Canadian high PEEP table, and then crossed them over and then studied them again. And why did we choose the high PEEP table? I'm going to show you the Youngman trial used best compliance. Well, we chose the high PEEP table of the LUVs trial because, and which I've conveniently reproduced for you here, because there was a study by Gattinoni a few years ago where he compared the best trade-off he could find with his cat scanner at the bedside between the LUVs way of titrating PEEP express, which was the European high PEEP trial, looking at what's called the stress index, which I spoke at SCCM about probably eight years ago, and the esophageal balloon approach, right? And he found the best trade-off between avoiding over-distention recruitment to be the LUVs PEEP table. Okay. Now I'm not saying one size fits all, but we thought that would be an effective standard for comparison. Here's our patient characteristics. It was a small trial. Okay. I mean, the, originally this was supposed to be a cytokine trial. Then the pandemic came and I didn't recruit for a year while we're waiting for the pandemic to end. It didn't end. We said, oh heck, let's just recruit, let's recruit COVID and we'll forget the cytokines and put an amendment with mechanical power. So you can study that if you want. It was in critical care last year, about a year ago. And basically what we found was this, that mechanical power, when we titrated compared to high PEEP table, went down as did, and then most of this was in what's called the elastic dynamic power. So in other words, is it the energy that, well, the vent is pushing air into the lung and driving pressure and static lung compliance also went down as we moved one way or the other, right? Either EIT over to high PEEP table or high PEEP table over to EIT. We saw a different optimal PEEP and, and I'll show you the delta in just a second, but lower variables, which I think we would generally consider to be reasonable surrogates for doing good, right? I showed you the I-Core data about driving pressure, high mechanical power, they were lower. Okay. And so in fact, in our setting, as opposed to the Youngman study, where they seem to say that, gosh, EIT told us to go up two centimeters of water, we found you generally from the high PEEP table of LUVs went down two centimeters of water to find the optimal PEEP using the intercept I shared with you, where the best trade-off between collapse and voiding over distension. So we went in the opposite direction. So that's kind of interesting, isn't it? Because it depends on what your gold standard is, right? If you titrate with EIT, it kind of, you say, well, it's compared to what? Well, in one study compliance, we did high PEEP table and things went off in a different direction. So when the Youngman article came out, we had a, you always want to get a little publication, we had a little letter to the editor of the Book Journal, so we call it latent heterogeneity treatment effect. In other words, EIT seems to tell us something. And in many of these little small studies, admittedly, the EIT titration gives you lower mechanical power driving pressure compared to what compared to the other standard. Okay, here's me with a high PEEP table. Here's one using the low PEEP table of ARMA, also static pressure. And so titrating with EIT may give you a more optimal way because all these studies with a different gold standard, if you will, show low mechanical power. Now there is one small series, this was Frontier Medicine 2021, I don't know if you can read this, but they actually looked at them all, right? A low PEEP table like ARMA, the old 6 versus 12, New England Journal 2000, high PEEP table, kind of like gloves that we did. Let's see, oh, this is pleural pressure over here, like a, you know, a Danny Telmer EpiVent. And here's the EIT. As you can see, the optimal EIT here in this small series was 13, where it was 15 with pleural pressure measurement, 17 on a high PEEP thing. Remember, we dialed it down. And 10 with ARMA, which they had to dial up. So EIT can give you something that maybe other modalities can't. And, oh, by the way, read the fine print here. In many of these instances, the mechanical power was lower. Now, not with the low PEEP table, and in fact, it was above the threshold seen in I-Corps, but as proof of concept, driving pressure was about the same, but mechanical power seemed to diminish with the EIT approach. Okay, so I know I've only got 15 minutes, hopefully that's not too much information, but let's talk about what we just told you. So the ability to recruit lung varies in ARDS, and I think, hopefully you agree, not everyone can be treated the same broad brush. And in fact, one-size-fits-all may be misguided, and a personalized approach may be justified. EIT offers a non-invasive means to examine the optimal trade-off. Right, FDA use right now would be best compliance approach, like they did in Junkman, but again, you may not get the same optimal approach when you use the intercept. The distension-collapse interface, that crossing over, if you will, to optimize the two ends of the spectrum, it might be the best approach, and hopefully that will come off the research notion into the bedside soon. So compared with high PEEP tables, EIT decreased mechanical power, driving pressure set of compliance in my series, and so, and because persistent elevations are associated with high mortality, and it's difficult to scale up a big trial with a device like this, you can, I consider them to be reasonable surrogates as a clinical endpoint. So best PEEP by the over a distension-collapse interface can be at variance, depends on what your gold standard is. It seems to be almost at variance in one way or the other. So here's some recent reviews. This is the one we had with Victor Jimenez. Here's a Blue Journal one. Stay tuned. There, I think, is something also coming out in a European Journal under review soon. So thank you for attention.

personalized respiratory medicine

positive end-expiratory pressure

Electrical Impedance Tomography

lung recruitability

patient outcomes