Episode 93: Fluid in sepsis and the FloPatch, with Jon-Emile Kenny

Two quick announcements before the show. First of all, don't forget, if you need your continuing education for this year, go check out the Intensive Care Academy at ICU101.com, it's Brian and my other project, tons of educational stuff, a little more formal and robust than the podcast here, and you can claim continuing education. It's probably one of the cheapest places you can get it, and we think it's good stuff. ICU101.com. Good for CME every year, it should be valid as well for nurses and most people. Second, there is a conference coming up that might be of interest to you. It's called Revive, and it's being run by the Center for Recessitative Sciences over at Penn, which if you don't know is a research group, it does a ton of the work in cardiac arrest and similar topics. And they have a new event. I think it's hoped to be an annual thing, but this is the first one. It's called Revive. It's based around cardiac arrest and resuscitation. It's going to be in March of next year, 2026, March 14th. So we wanted to let you know about it because it sounds cool, and we've partnered with them a little bit to try to support it. Listeners of this show are eligible for a discount to the conference. When you check out, use the code Revive Penn, R-E-V-I-V-E-P-E-N-N as in Pennsylvania, you get $75 off, which is a good deal. It's not an expensive conference, there's also a purely virtual option if you don't want to go there, although I would, there's some cool hands-on stuff. So think about it. You've got enough lead time that you should be able to get the opportunity. It's only one day if you're anywhere in the region or willing to travel a little bit cool event. Revive Penn, if you want a little bit of money off, go check it out. Otherwise, onto the show. Hey everyone, I'm Brandon Odo. And I'm Brian Bowling. And this is Critical Care Scenario. It's the podcast where we use clinical cases, narrative storytelling, and expert guests to unpack how critical care is practiced in the real world. All right, welcome back everybody. It is Brandon Otoe. And there is no Brian Bowling for you unfortunately. Brian had a last minute crisis. Maybe he can tell you about it next time. I think his refrigerator exploded or, I don't know, became sentient or something like that. But it's just going to be me and our guest today, which is good because we have a great one. So we're here with John Emil Kenny, who is a pulmonary and critical care physician, who some of you may have heard of. He's been around the educational sphere online for some time. He's written over at a poem CCM, the blog over there. He has an entire online textbook he may have come across at heartlung.org on physiology topics. So a lot of kind of useful stuff that a lot of people encountered. But we wanted to get into some topics of fluid today and how to assess fluid responsiveness. And I want to just lead with a little bit of transparency. Generally when we have guests on the show, it's because we had an idea for a topic and reached out to them. This one is backwards. We were actually gotten a hold of by, not even Dr. Kenny, but someone from his kind of I guess marketing team, to talk about the product that he's involved with, which I'm sure will come up today. We actually get a fair amount of messages like this and generally the answer is no because generally the topic is not of any interest to us. This actually was, which is why we're all here. So I want to just be clear on the nature here. There's no money changed hands here. This is not a sponsored deal. There's no consideration for all of this. Obviously, you should probably presume that Dr. Kenny has some bias if this topic comes up. As far as I know, I do not. It's not like he sent us a product. And as I said, he's not paying for our lunch or anything like that. But just to be kind of clear on what we're talking about here. So that's that. Thank you for joining us, Dr. Kenny. Scott, thank you so much. OK, so let me just pose to you a brief tale about a patient who might be in your ICU. Let's say a 65-year-old female, just a history of hef-pef, hypertension, and she came to the hospital with malaise, abdominal pain, weakness, some vague symptoms, but was found to be hypotensive with a fever, lucocytosis, an elevated lactate. She had some vague abdominal symptoms, and her LFTs were elevated. So someone did an ultrasound, which showed a dilated common bile duct. So everyone thought she had colonitis with resulting sepsis. She's boldest with two liters of isotonic crystalloid, pick your poison. And they start norepinephrine to support a blood pressure. They call IR to try to put a drain in this thing. And in the meanwhile, the resuscitation continues. Something does a quick ultrasound exam, and the heart looks OK, pretty normal bivontricular function. They look at her IVC, it's about 1.9 centimeters. It's collapsing about 40% with spontaneous breathing. The norepinephrine dose is training up a little bit, and so your resident asks, "Should we give more fluid?" Share with me just to your general take on this sort of congenius topic. It's fluid resuscitation, and not just the shocked patient, but let's say specifically the septic patient. This is sort of one of the favorite topics in critical care. Everyone has their own perspective and their own biases, but where do you kind of come from in general? Yeah, I mean, like certainly I'm a big advocate of personalized tailored resuscitation, certainly using point-of-care ultrasound is definitely my academic bias. In a situation like this, there's some hypertension, there's an elevated lactate, don't know how high that lactate is, but if the patient is persistently hypotensive or if that lactate is above four, the CMS essentially mandates a bolus of a recommended 30 ml per kilogram. I know that that's very contentious, especially in a patient like this, because you don't really know kind of how that patient's heart especially is behaving at this moment. These patients with presumably an impacted stone, you see them in the irritors, you see them in the CBT, they just get so cytokiney, so serzy, so hypotensive, and those cytokines they affect the, I think in sepsis, we think of like it affecting the peripheral vasculature and leading to hypotension and venodilation, et cetera, et cetera, but it also affects the myocardium, and you can very commonly have acute septic diastolic dysfunctions, and you told me she has some path to begin with, so these are the kinds of patients where you can look at their heart and see a normal looking ejection fraction, see an LV cavity banging away, but if the LV is actually quite stiff, you could preload these patients, and there's a non-zero risk of flashing somebody like this, so I don't think it's something that you just, you know, take kind of, you know, a super-lippable approach and just do 30 mls per kilogram wide or right away, I think it's something that you do with a little bit of, you know, certain inspection, and certainly with the probe to help you out. When I think about giving fluids, you know, I kind of have this framework that I teach my team, and it's just, we call it the 3Y, the 3S framework, and essentially you've got to answer three questions when you give fluids, there's got to be first an indication, so yes, this patient's hypotensive, on norepi, elevated lactate, so certainly there's an indication of hypoprefusion that you'd want to augment stroke volume or cardiac output with some preload, so indication, first, yes, is there, and the second question is, is it safe to give fluids? And that is really a clinical exam supplemented by ultrasound, so you're looking for that volume tolerance kind of picture, you know, are there belines in the lungs? Is the chest X-ray wet? Are they crackly? Do they have a big IVC? Do they have a high CVP? Do they have weeping edema? You know, if it looks unsafe to give fluids, then whether or not their fluid responsive or not becomes a little bit moot because the risk profile is already, you know, it's already been expressed to you that fluids might not be the right option, but if it looks safe to give fluids that they look clear lungs, you know, collapsing IVC, low CVP, a history consistent with volume depletion, so safety, yes, then the next question is are fluids effective? And that is really when you're looking at whether or not the patient is moving upwards on the starling curve or not, say with a dynamic assessment. And I think, you know, as an analogy, I think people have a hard time thinking about fluids that way, and there tends to be the sort of misconceptions that fluid safety and efficacy are kind of one and the same, like if it looks safe, if the patient's volume tolerant, then they're also volume responsive. And, you know, the converse is well, but we think about this commonly in the setting of antibiotics where we could think of it, you know, piptezo or something where a patient could have terrible anaphylaxis to it or AIN to it, so it could be very unsafe to give that patient that particular drug, but the drug could still be effective, it could kill the ecoli that's in the blood that's safe for this patient. And the converse is also true, it can be, it can be, you know, safe to give an antibiotic, they may have no history of anaphylaxis or any adverse reaction to the antibiotic, but the antibiotic could be ineffective, i.e., you know, the ecoli growing in the blood is resistant. So we think about fluids very similarly that you've got to do that safety check first. Do they They look volume overloaded. Well, then, why are you going to pound them? more fluids. And then if it is safe, then you want to make sure it's also effective by doing a dynamic assessment to see if there's no responsive. Well, let me, let me, I guess pose another philosophical question. Would you agree that maybe after an initial resuscitation, the most common phenotype of sepsis, ruling out the one with a concomitant cardiac issue, whether they already had that or it's induced by sepsis, most of them do not have a low cardiac output. If anything, it's often a sort of hyperdynamic state. We're not talking about a pump problem primarily. This is a primarily distributive shock. Would you agree with that? Uh, you know, I think that it's a grab, it's a grab bag. And I think that, you know, like I said, there's a, you know, there's a very high and misunderstood incidents of a septic diastolic dysfunction where patients could have a normal stroke volume or even a high stroke volume or a high LVOTVTI if you're doing an echo. And, you know, it may, it may look like they've got a distributive problem. It may look like they have a high cardiac output and don't necessarily, you know, need more fluids or could be benefit from fluids, but they, they, they could kind of have a cryptic septic diastolic cardiac dysfunction or giving preload to that patient could be harmful. The converse is also true. Like paradoxically, patients with really bad LV dysfunction is a function of sepsis. You can't see, those patients tend to have kind of a better prognosis for kind of their reasons. But, you know, in other words, if I looked in a numbers sense, let's say we put PA catheter as an everyone, which of course we don't need more. If you were to measure all these people's cardiac output or index, some of them may have, quote, normal values or maybe normal high, but even some of those may be, if you increase that number, whether it's with fluid or something else, might be in less shock and have better profusion. Yeah. Yeah. They may, you know, they may have high values and they may or may not respond to fluids. They may have low values and they may or may not respond to fluids. I think that that's where kind of the personalization comes in now. If somebody's, you know, hypotensive and their stroke volume is 110 mls and their LVOTVTI is, you know, 27 centimeters, you know, like you could make a very reasonable argument that giving that patient more fluids is not not the issue in that, you know, pressure is flow times resistance. So, pressure is low and cardiac output is high, then you know, vascular resistance is low and you could be well within your, your rights to start that patient on a, on a normal happy infusion or a major pressure infusion and do really well for that patient, like 100%. But I think that the important point is like, you always want to, you always want to take a look, you want to measure, you want to have some sense of what stroke volume or cardiac output is before you start looking at these patients and and that could be as simple as a pulse pressure, you know, a patient who's septic who's 90 over 70 is very different from the septic if the septic patient who's 90 over 30, you know, like the latter case might have a normal normal is stroke volume or hyper, you know, or or a high stroke volume and that patient may be the one and I think they're advocating for this, the Andromeda Shock 2 study that you start with with press, right, you know, and get that diastolic pressure up. If they're 90 over 70 and they're septic, narrow pulse pressure might be a low stroke volume, then I think you have to kind of get a sense of cardiac function, you know, is it their stroke volume low because, you know, their, their, their tank is empty or is their stroke volume low because the heart's not doing so long. So, you know, I think for a lot of people, maybe who are less on terminally online, their perspective on fluid or cessation and it could be in anything but sepsis is, you know, bread and butter everywhere. It feels like giving fluid compared to other interventions like giving pressers is easier and maybe safer. It feels harmless and in some cases avoiding complications of other therapies, which might be as simple as having to transfer a patient to the ICU or, you know, more serious things that we know can be associated with pressers. And if they think about complications from fluid, they're going to think about one or two things only pulmonary edema, particularly in patients who are susceptible to it. Maybe in some select patients, more specific things like a surgical patient, you're worried about having abdominal compartments syndrome or something like that. You kind of talked about safety. Let's try to make a little more real the idea of the harms or complications of fluid because any risk-benefit consideration, you have to, you know, reflect on both sides. So, what are the downsides of giving more people more fluid? Yeah, you know, like when I was an intern, it was it was in the heyday of Mani Rivers, you know, the early gold directed therapy in sepsis. And so that was the 2001 wingland paper that, you know, every resident had that that resuscitation algorithm burned into his or her brain and it was like CVP less than 8, you give fluids, you give fluids, you give fluids. And you know what, and I had a senior resident, you know, say, you give fluids to get their CVP up above 8 if they can't breathe, you anticipate them and you keep giving them fluids to get the CVP above 8, which usually happened because of the peat, but that's another discussion. But what we, what was found in that study, and anecdotally as a, you know, practicing resident in the mid-2000s was these patients would get 10 liters in three days, 13 liters in three days, 15 liters in three days, and they would just, you know, were we killing them? You know, I think you could make an argument that you probably tipped some older people over the edge, but you certainly were keeping them in the ICU for longer than they needed to be. They were, you flood them, you put them into pulmonary edema, you anticipate them, they're in the unit for a week, you know, they get a VAP, etc., etc., they get a CDF, they get a bed sore, etc., etc., like, you know, I think, you know, an analogy that I was, I was told is sort of like, you know, they had this idea, you just give fluids, and then you deal with the consequences is like, it's just like, you know, saying that you've got good car insurance, I'm just going to drive like a maniac and crash into things, and I'll just, my insurance will take care of it. Well, that's one way to approach life. I think the other, I think the other kind of approach is like, don't drive like a maniac to begin with, and just be judicious with your fluids, and rational with your fluids, and you can avoid those downstream consequences. And I think there's pretty good literature that when you, when you are more judicious with your fluids, when you're more thoughtful, more rational, you keep the patient off the vent longer, you more than likely reduce length of stay in the ICU, you probably reduce complications of renal insufficiency, which is paradoxical, because like, hey, like the kid needs like fluids, but there's, you know, Phil Rola, who I, you know, I know very well, we'll tell you that, you know, if you congest the kidneys, if you increase the CVP, then you, you get kidney insufficiency as well. So, not to mention, you know, the gut edema and ilius and anasarca that just keep, prevent patients from getting out of bed and amulating, and you know, if, if, you know, I think if you're just more judicious with your fluids, then, then you can avoid a lot of those downstream implications. Maybe you say, 12 hours in the ICU, maybe you say 24 hours in the ICU, maybe you prevent a VAP, but these are all, I think, righteous approaches when you're, when you're resuscitating a patient with crystalline. IR comes, they put in a drain, they think they got source control. We get to about, let's say, that the 12 hour after they presented, their pressure requirement is sort of stabilized, but it's still ample, they're on kind of a moderate dose of norepinephrine, they're in vasopressin, they're on steroids. Their fluid balance by now is, you know, four and a half liters, some positive, between some fluids, some medications and so on. So, your resident calls again and says, you know, it's still in a fair amount of shock. I think you might still be fluid responsive. What about now? So, you kind of alluded to markers or tools we might use to try to understand if a patient is fluid responsive, and you distinguish that from the question of whether it's safe, or maybe even a good idea to give them fluid, but we'll call it a responsiveness, at least, to prerequisite further being any benefit to fluids. What are some markers that might be useful here? I'll throw out the most basic ones. Guessing, a sort of general gestalt based on the disease, like, oh, they were, they were bleeding or something, they're a decay, they seem like they're dry, that sort of thing. And then a common one is sort of eyeballing the patient's IVC and making some sense of that. Do you think that there's any tools that are better than others, or at least more practical? And I guess I would distinguish practical practicality versus theoretical ones, because I think there are a lot of things that have been talked about that no one is actually doing on a day-to-day basis. Yeah, I mean, I think ultrasound is kind of the key differentiator in this day and age. You know, when we approach a patient like this, one of the, and we published on this framework, it's something that we kind of call the Doppler Starling curve, or we take the old venerable diamond and foraster framework that they achieved from PA catheter's post MI, and you kind of quadrant the patient. There's four options that you can be in when you're hypoperfused. And you essentially dichotomize the x-axis, which is filling pressure, so the diamond and foraster was with a wedge, so it was above a blue 18, which on the x-axis. Y axis your dichotomase based on your forward flow. So in diamond and forster it was in the index of I think two point two or two point one or something like that, but you were either like, you know, normal cardiac output low filling pressure quadrant one so warm and dry whatever that means you are high filling pressure normal cardiac output so quadrant two that's warm wet whatever that means and then quadrant three is, you know, low filling pressure low cardiac output. Cold dry and then cold wet is the bad one that you worry about in medical school and you know post at my patients that have a high filling pressure in a little cardiac output. And so what we have argued for is is that you make that same phenotypic evaluation using the probe you could you could use the the height of the jugular venous pulse and or a CVP or you know or clinical you know plus the pulse pressure. But the important thing is to sort of get a sense of where your patient is in those four quadrants because it does immediately imply therapy. So you know if this patient if you if you looked at your patient in a big dilated IVC it suggests a high filling pressure so you kind of know they're either in quadrant two or quadrant four so warm dry or sorry warm wet or cold wet and then you could and then on the y axis you look to see if they're what their LVOTVTI is as a surrogate for their stroke volume. So if it's above 17 to 18 centimeters you know certainly if it's above 20 it's it's more of a normal stroke volume they'd be in the they'd be in the warm wet profile. And then you know if they had a very low LVOTVTI and and a distended IVC they'd be in that cold wet profile. And so I think that that is a really great starting point now the goal of this video game is to get to quadrant one you and me hopefully right now are sitting here in quadrant one with a normalish LVOTVTI and a collapsing IVC a low CDP of close to probably zero. So you kind of then have to troubleshoot like how do I get from quadrant I'm in to quadrant one and that's where the rational therapy I think comes in. So if you told me this I would look at this patient and I would say well if you tell me they've got a dilated IVC it's not really collapsing and they've got an LVOTVTI of 15 well it sounds like they're they're in that kind of cold wet profile. And I'm not even going to think about giving fluids that's telling you there's there's probably pump dysfunction and I'm thinking like do they need an Ina trope do they need. You know they're on norapy vaso is that is that they actually need a little more beta do they need some epi or dog or you know something along those lights to pull that starling curve from that low right Q4 up into that high left Q1. So I mean just thinking it through the only person who would benefit from fluid would be the cold dry patient because if they're already well perfused and if they're wet regardless of whether they're in shock they're not going to respond to fluid. I like the way that you're thinking about this I think yes but I think that this is the asterisk and this is where my bias comes in that when you look at the cold dry paradigm that Q4 so the collapsing IVC low cardiac output low stroke volume phenotype. That in the septic population there's still probably a 20 to 30% fluid unresponsiveness rate that you need to account for to not just push that patient out a flat starling curve from quadrant 3 out of 4. And that is kind of where the many rivers paradigm broke down because implicit in the many rivers paradigm was that if you had a low CVP so you were in that you know Q3 then you would respond to fluids you would have an upright starling curve. But in sepsis all bets are off and there's a fair amount of patients that have a depressed starling curve but who also or vino dilated or maybe a little volume down so they have actually a low ish filling pressure they may have a CVP of 5. You know it could be a collapsing IVC low ish CVP and they may have a stroke volume or LVOTVTI of you know 13 centimeters or you know 40 ml or something like that and you're thinking oh they just need fluids they've you know tank is empty but there's probably one in five to one in three patients in that Doppler phenotype that you're actually just going to shoot them out along the starling curve further you're going to shoot them out to quadrant 4. And if you start to see that repeatedly then that patient probably needs more beta is kind of like how I would approach this issue. Right and that seems like it would apply to any of the tools you're using to try to place these patients because you're relying on some kind of cut off within some kind of device but whatever you're calling a high low or normal VTI that's what you're using whatever you're calling a high low or normal CVP or size of the IVC or Vexis or whatever you're using those may or may not map to the question that we're trying to ask which is will the patient actually not even just increase their cardiac output but clinically have a meaningful response to more fluids. I mean you yourself said whatever we call a normal cardiac output I mean who's to say that more might not be better in someone like this or maybe low is okay. Yeah, you know I think I kind of the more is better I think was kind of litigated in the 80s with the studies of Schumacher like the super normal oxygen delivery it didn't really pan out. But I think you know what intrinsically what I like about quadrupling the patients is that you know exactly what it's like at the bedside you got a zillion things going on. You got to look at you got three other patients that are crashing it's like how do I just break it down quickly for this patient right now it's the human brain can dichotomize on the x axis and dichotomize on the y axis pretty well it's like one of four. And then it kind of suggests where you need to go. If you are thinking about giving fluid so exactly as you say if you're in that quadrant three you know my my bias and I think that the literature generally supports this is to do a dynamic assessment to see if they're actually going up to quadrant one or am I just shooting them out to quadrant one. So rather than just trying to eyeball one thing give them some kind of a challenge to actually see what happens where they're going. And what do you think works well for that because people again have talked about many things everyone talks about passive leg raises and then no one wants to bother to do it. There's various devices realistically what I think most people are doing is giving a little fluid and then seeing if the blood pressure gets better. There's no I would say to that latter point there's no no fault in doing kind of a fluid challenge and seeing directionally which way they go. The problem with blood pressure is that it has a very poor sensitivity for knowing if you've actually augmented cardiac output or not the specificity of pulse pressure at least is reasonable but it's not perfect. But there's a very large fraction of septic patients in whom you give fluids who cardiac output actually increases significantly but map or pulse pressure doesn't really. And in those situations you're sort of left with this conundrum of like did the fluids not actually help. Or did they actually help and maybe they need a little bit more I'm going in the right direction and I don't know and that's why I think flow monitoring is is so important to actually know that you. Augmented what's coming out of the left heart with an LVOT VTI or you know other other Doppler systems you know my commercial bias notwithstanding that you're at least using arterial Doppler specifically to infer an increase in in life in particular output because I'm sure you've been in multiple situations where you've had a map of 52 and you've given you could imagine two patients with a map of 52. One patient you give fluids to if you measured flow and you knew that flow didn't change and their maps went to 53 versus another patient with a map of 52 and their map went to 53 but they're flowing up by 30% after that preload challenge like I guess the question is like would that change how you think about that patient with that change how you manage the patient. Some people would say I'll just go and trend their LVOT VTI usually my answer is but will you really I mean it's burden is that the main is at the main downside I know there was at least one device floating around that would try to do this continuously but it's I mean it's challenging some people even just getting a spot view of this measurement is hard and then doing it whatever I don't know what five times a day I don't know what you would have to do for this to be a meaningful monitoring tool. It's hard I mean it's it was the genesis of of you know the commercial device that that my company is working on for sure like I just got tired of doing LVOT VTIs so one of the device that you might be talking about is called the us come device I think it's a supersonic notch continuous wave Doppler it's based out of Australia and I really like that idea. You just sort of sit it in the supersonic notch that looks down actually gets an ascending a Arctic VTI but it helps you kind of noninvasively determine whether or not the VTI is going up or not. The device that we build sits on the neck it makes an inference from the duration of sisterly from the carotid artery but it's the same idea like I want to give fluids and I want to know if map isn't changing is flow going up or not. You know because that helps me think about the patient and if the patient is map isn't changing and their flow is just spot on. like delta zero and it's like us patient needs either dog or you know beta alpha or some combination of the two If I'm seeing their map is going 52 to 54 55, but they're having these big augmentations and volumetric output I might try and squeeze a little bit more out of them because they're part their pump as the ability to Effectively deal with it before I before I jump to press to prevent the downstream potential complications of overpress that makes sense Okay, why I don't want to tease people too much. So I mean tell me about your device. What's it called? What's it look like? What's the principle? Sure, so it's thanks It's a it's a wireless wearable continuous wave Doppler ultrasound. It's it sticks over the neck. It shoots to an iPad It measures the so it's a patch literally sticking on the skin attaches to nothing It's literally an ultrasound patch attaches to nothing but the patient shoots to an iPad and then The more for more sophisticated users the nice thing about the device the neurovascular bundle there is that it's insinuating the jugular vein and So you get a mini Vexis. I know it's part of the Vexis, but I bug Corbin and fill about this Why the jugular wasn't part of the Vexis? But the physiology is exactly the same as the hepatic vein intramural vein Portal vein that the more pulsatile the jugular is the more chunky the the right atrium is the more congested the right atrium is So the device kind of gives you where you are on that X axis that that you know low-filling high-filling it points you in that direction And then we're looking at the jugular vein or the carotid artery So jugular vein internal jugular vein and the carotid arm a carotid artery Simultaneously, so the spectrogam gives you both Because it's a continuous wave system and then it measures actually delta changes in the carotid artery So the neat thing about the device is that you can sort of think of it as it sits on the neck And it's simultaneously looking down at the right heart and the left heart You know the right heart through the jugular vein the left heart through the common carotid artery concurrently so you could Give a preload challenge and you could see like where they're moving on that X axis or is there jugular just getting very congested and the carotid is having very little delta on the Y So then you know you've got a flat starling curve slow versus you give a preload challenge and the internal jugular changes Very little in its morphology doesn't get pulsed out all and you have this big delta in the Y axis from the carotid Well, then you know you've got a very efficient heart and and it's liking the preloader at least handling the preloads safely and effectively That's a genesis To connect this to applications people might be more familiar with if you were to do something similar Using normal tools. This would be applying probably a high-frequency probe to the neck in like a like a cross-sectional Transverse view where you saw both the jugular and the carotid and then putting probably pulsed wave Doppler over each one getting a spectral tracing and then measuring the Change in the peak velocities Over time. Yeah, so in the jugular it gives you it just gives you the the the spectacular spectrograms There's no image. I just want to be clear So you don't see the jugular you don't see the carotid you just see the wave forms in the jugular and the carotid Concurrently so if you did a preload challenge you would see the S in the d waves evolve or the S get less than the d wave or the Become a monophasic d wave as the right atrium fills up and then that is a Volotative clinician assessment like we don't this is like watching a CVP tracing a center. Yeah Exactly, but you watch it in real time and when you do a leg raise or a trendellanburg or you have a rapid preload challenge You actually it's quite neat. You see the the venous Doppler change in real time as the right atrium dills and then but what the device actually does Is it measures a delta output from the carotid artery that is an inference of that strobe volume? So it's telling you if that strobe volume is going up or not So it's driving a volume by using the velocity over time So so good question we measure the velocity time integral of the carotid But the measure that we defer to because it has a much better evidence base is actually just the duration of systole the flow time and probably Spend a number of papers on the flow time actually over 80 years or so, but probably the best paper in the last In the last 10 years came out of UCLA by The intensive is group there that used handheld carotid Measurements to look at the flow time and compared it to bioreactance But like for people that are familiar with Doppler the time of systole is like the bait You know when you get that VTI out of the out of the left ventricle It's like how long from the start to the end of ejection exactly, but it's like the Stroke distance the VTI is the area of that triangle and the area of a triangle is Based on types about it by two, right? So what we're telling you is the base and how that base is changing in response to Because the the height the velocity doesn't seem to be as important or at least not it is not as well evidence-based Phenomenal question when you give preload the Increase in the stroke distance tends to express itself in the time domain rather than the velocity domain So this the flow time is really germane for that preload change Changes in afterload and contractility will express it in the velocity domain the height of that triangle So predicated on time as a surrogate for distance You need to infer that Afterload and contractility are pretty constant, which is a fair assumption when you're giving a rapid preload challenge or doing a leg raise or trend ellenburg Like afterload and contractility are pretty constant and so that increase in the time domain the base of the LVOT triangle Can be used to infer the increase in the distance now if I were to like start a patient on dog Or epi when you start increasing that contractility then the time and the distance the area of that triangle can dissociate So it is not You know recommended or fully validated in situations where you trade it you're significantly changing Chrono trope or or or or afterload Okay, but within that's fixed conditions of the heart the duration is something you can trend if if in the interim something else changed like somebody added on a nine atrope Maybe you can't trend that but now maybe you have a new baseline. Yes, so yeah, the goal is sort of like you know The you activate the device on the iPad you collect a 30-second baseline You know the ventilators the same the phase of active infusions are the same you do a trend ellenburg right you recruit that Venus blood Oh for two minutes maybe maybe maybe two and a half minutes you look at that change in the jugular You look at the change in the carotid artery you get a sense of what the slope of the starling curve is at that moment And then you make your decision and then the device goes to sleep. Oh, okay Yeah, so wait let's operationalize it. So somebody walks in the room. What data is on the screen? You walk into the room the the the device is on the patient's neck and it's actually asleep. Okay, so it's it's a It's not a continuous monitoring tool you tell it when you want to me. It's like shooting numbers on a swan Yeah, I mean I kind of Call it like you know we call it wearable point of care ultrasound. It's like walking in with the probe to do an assessment But it's supposed to like Allow anyone to do it and it and it's it's supposed to give you a faxus and a BTI at the same time Okay, yeah, so then you tell it you want to you want to do a measurement and then you you put you do like a leg raise you put them in trend ellenburg Yeah, or you could do a rapid fluid challenge, you know you you know 5300 ml over a couple minutes like it's got to be a good a good squeeze Usually requires a pressure bag or or a large bore with gravity, but it it's not this kind of like 999 kind of nonsense like you really got to do a Stretch the myocytes and see what develops. Okay, and then what what is it what is it reporting to you? So it the qualitatively you it shows you that you know the pre-post and then it shows you the jugular changes And it shows you the delta flow time so that shape that you you're looking for the augmentation in the base of that LVOT Like a percentage change So it can do that, but what we quote is that that paper from from my my former colleague Egor's group at UCLA where they found that An increase in 7 milliseconds was the optimal threshold for detecting a 10% stroke volume change We we've recently published comparing the same thing to TEE LVOT BTI and we found very very similar thresholds to Egor in terms of an increase in absolute flow time As a as a very good marker of an increase in LVOT BTI Okay, so if if your goal is to increase the cardiac out or the stroke volume and hence the cardiac output by about 10% And I I would imagine that cut off just because that's often what the literature is used as the definition of fluid responsiveness Then that seems to be indicated by an increase in that Sistolic that ejection time of what point seven seconds you said, okay. Yeah, so seven milliseconds And yeah, like we could talk about the definition of fluid response that I'd love to talk about that I mean, it's just like that 10 to 15% augmentation is just the precision of essentially thermodilution. I'll say it just was like the precision of the gold standard utensil. So like a change of more than 10 to 15% meant that it was kind of real and less than that was just kind of wobble around the baseline. So that's, there's no like mammalian cardiac truth in that 10 to 15% number. It's just like, oh, there was a significant change, you know, based on thermodilution. - And so that's from the arterial side. And then from the venous side, it's just showing you the waveform before and after if that's of use to you and all the people who can't interpret CVP waveforms will ignore it. - Yeah, we're working on automating and interpretation using AI, et cetera, et cetera. But for now, it's just for the Corbin Heycox and the Phil role is of the world to look at it and say, well, like the D wave is really big now. The right atrium is probably angry. - Yeah, 'cause this is much like if you had looked at the, I'd be see or the like the hepatic veins and got a tracing from there. I mean, you're above the heart and up below it, but the same sort of vibe. - 100%. - 100%. - What are the caveats here? I would imagine it depends on the patient's position. Like if you're intentionally putting them into nowhere, but if you, let's say, challenge them with fluid while they're sitting up versus laying flat, I feel like they would have different performance characteristics. - Yeah, for sure, I mean, we see it regularly is that when you give a 300 ml crystalloid infusion with the patient in the upright position, the changes in the jugular morphology are very subtle. And I think that's a consequence of it being, as you mentioned, gravitationally above the heart, but also just like small crystalloid challenges lead to relatively small changes in right atrial filling. When you do a passive laygraze or a trendellenberg, you really see the morphological changes in the right atrial. So when the device is, as you've scanned a jugular vein with a B mode, you know that there's a collapse point that's so-called wine bottle. And when the device is above the collapse point of the jugular vein, you get this very kind of high velocity washed out Doppler morphology that is kind of lightly undulating, but it's-- - The cardly any flow. - It's more like kind of like, imagine putting your thumb over the end of a hose, and it just says it's high velocity wash. Like that's kind of the pattern you see in the jugular when the device is above that collapse point. So when you see that pattern, you know that the CVP is like less than the height of the device above the right atrial. I'm not advocating that people actually measure it and calculate it, but you could if you're kind of a dork. When you lay flat and you get that rush of venous blood into the right atrial, you immediately see the CVP trace in the jugular. You see the X descent creating the S wave, you see the V wave, and then you see the Y descent creating the D wave. And in healthy people in S greater than D in the supine position is totally lower. - Any other caveats or country indications? I mean, a lot of these non-invasive devices will have a laundry list of times and they don't think you should use it, or at least the company doesn't recommend that hasn't been validated or whatever. - Yeah, I mean, like any ultrasound device, it's if it's garbage in, it's garbage out, right? So it's not a monitor in that sense. It's you're like a mini sonographer. So if you have a crummy window and you've got a crummy spectrogram and you've got crummy, the device would say, like it's like the bars that come up that represent every heartbeat will be missing. You'll say, oh, this is bad, like it's not good. So a patient who's thrashing, screaming, you have to imagine ultrasound as being like a super, super sensitive set the scope. It's literally listening for the movement of PRBCs. If you could imagine putting a set the scope on somebody and they're screaming 'cause they're delirious, you're not gonna hear long sounds, you're not definitely not gonna hear red blood cells moving in the carotid. So that's like the key patient population where it doesn't work. Where time diverges from distance, it's changed as a measure of Delta SV. Kind of mentioned some of them already, like significant changes in an initropy or afterload during that intervention or across time, we'll make that less true. And then certainly like patients with like really severe mitral regurgitation or severe aortic stenosis, where you could have more of that LV ejection going back up into the left atrium rather than out the aortic valve. Then that relationship between time and output is decreed. You could say, well, the patient's acting as their own baseline 'cause you get a baseline in that Delta. But if more of that ejection is going back out into the left atrium 'cause the severe MR, then that time will be a problem as a LVOTV guy. - I assume it doesn't work if their flow is not pulsatile. Like they're on ECMO, they're not doing any of their own work. - It's a phenomenal question. We have a case series in LVOT patients and we are able to detect changes in RPMs. - I guess it's the flow, right? So it's, if there's an increase in your baseline, like laminar flow, it should be detectable. Just no time is not. - Yeah, when you turn up the, when the, you know, RPMs are high, you just see this wash in the carotid and as you give the LV more of the work, you start to see these little shark fins like up here over the wash and it sort of becomes more pulsatile, but in terms of like a marker of delta stroke volume, no it won't work, for sure. - So I mean, is this imagine error? I mean, slash, it's typically being marketed for the septic patient like I described, or could it have a role in various other shock states or, you know, fluid management situations? - Yeah, another really good question. I think that the, you know, I think that most of the questions come from these kind of surzy septic patients or the undifferentiated patients where you don't really know what's, like if it's like those rare times when it's just clearly cardiogenic, like when it's fluid's really being debated or if it's like they're clearly having a GI bleed, you know, like we're not gonna check for fluid responsiveness if you're sanguine eating, you just need source control and you need blood. You know, really severe DKA is another situation that we see where patients come in so dry, so dry, and their electrolytes are so bonkers, their pH is so low, their glucose is so high, that these patients can be fluid, unresponsive at the very beginning, and you give them liters of crystalloid, you fill up their venous beds, you actually start to prime the pump, and you change their electrolyte milieu, you get their pH up, their potassium right, and their glucose down, and then the heart starts becoming fluid responsive. So, you know, it's one of those situations where, you know, when it is a known etiology with a clear therapy, you know, fluid's insulin, blood, you know, thrombolytics, if it's a big PE, like it's not exactly the right thing to use, per se. - Well, even the sanguine eating patient, I mean, a not uncommon situation, and a big bleed is that they're obviously hypovalemic, and at some point they start bleeding, and then it's not always obvious when they're no longer hypovalemic, and you need to stop pumping blood into them. So, I guess I can imagine pop-up monitor, and then at some point it says, "Hey, this is not doing anything anymore, "maybe if you're hypotensive, it's just distributive, "or some other whole thing you're missing." - For sure, 100%, like if you've got, if you've got, you know, source control, if you know the bleeding has stopped, and you're sort of in this in between undifferentiated state, then it's, you know, it's the device to use. Like I kind of say, like if you're thinking about, if you're a, if you're a focused person, you're thinking about picking up the probe, well, you know, this is kind of like, you could think about picking up this device as well. - It's those situations when you're just not sure. - Couple practical things. What does it cost? How long do you leave it on? - Cost, that's a, you know. - I know cost in healthcare is all sort of made up and the rules don't matter, but I don't, can you give us some rough sense? Like, yeah, I mean, I would say that, you know, we are, we are priced very comparably to the other reusable stick-on devices, whether it's the finger cuff device or the buyer reactants pads, you know, we are right in line with that. And so that between different institutions, that's kind of anywhere between, you know, three and $500, you know, I think central lines. - So it is a reusable device between patients. - It is, it is a repurposeable. So it's a one-page, one-device use, and then you send it back to us. We give you actually a rebate if you send it back, so we can repurpose it, recycle it and stay green. So we're not just throwing away ultra-sounds. But, you know, like, it is, this, you know, the diplomatic part of me is, it's the same price as other reusable stick-ons, and it's actually cheaper than most invasive catheters for patients. - How long can you leave it on there? - So the adhesive, it's stuck on there, should be taken off and replaced every 24 hours. And that's just to look for skin breakdown, et cetera, et cetera. The device itself gets you about 100 assessments with the battery life. So you can keep it vulnerable once the battery is out. It's done. - Correct, correct. So the, you know, you could keep it on the patient for two weeks if you wanted. If you were, you know, doing that daily skin check and you were below that kind of 100 assessment mark. - Switch side to the neck every day or something like that. - Yeah, yeah, exactly. - How do you see this? this in application and pros and cons compared to other non-invasive cardiac output monitors and I feel like the most common that are out there are the ones that use an arterial line tracing whether a stroke volume variation or pulse pressure variation to say something much like you're describing here of course a whole different technology but this is just kind of going about a different way do you think there are upsides this it's just kind of a different approach yeah I mean like you know the other technologies are great and I you know I love it when I walk into hospital and I see them using some sort of flow guidance for their resuscitation that to me suggests you know progressive you know thoughtful institution and if you know I'll bless you if you're if you're if you're using you know an a-line based device you know I you know the kind of obvious answer is like if you're using Doppler ultrasound with a probe our device it is a direct measure of flow velocity if you're using a pressure waveform then you are using anthropometric or you could calibrate it if you've got you know the pico device but you're using some sort of you know transformation to go from pressure to flow and that can become problematic when there's significant changes in purple resistance or purple compliance so some of the devices that measure pressure and in fur flow do have problems when there's you know significant vasoactive infusions whereas you know a direct measure of flow over a major artery is much more robust to that you know and then there's just the the other obvious I think answer is that it's just it's just a it's a stick on ultrasound you know like ambulance ED like you don't need to put an a-line in you know if you got a if you've got a good BP cuff and you've got you know a way to measure changes in in Doppler ultrasound in an artery then you kind of have you know many ICU at your fingertips and then you could do that almost anywhere you know the nice thing is that it has other kind of applications of like rapid response you could very easily bring it to the bedside on the floor you know we've got at least one case report in a data set for for rust detection and CPR so it's kind of got like these other peripheral applications that are really nice but that's kind of you know my biased it's but that's grounded in my bias in in point of care ultrasound which is which is how I trained so let me as we kind of wrap up let me I don't want to be difficult but let me kind of pose that the devil's advocate position because I think a lot of the audience may have a similar thought and say look maybe there are other niche applications but imagine the septic patient yes it's not great when they get too much fluid but let's say a patient who 20 years ago would have gotten ten liters of fluid in the first day now it's twenty twenty five times have changed maybe they get five liters you go and give them a fifteen minute spiel trying to reify the implications of volume overload like we just did and they say okay now this resident is only given give that patient three liters the the room for for benefit if we're talking about fluid stewardship is sort of on the slim side in that kind of middle of the bell curve person when we're talking about sepsis a disease that is not primarily a volume problem now you can say this person I can increase their cardiac output by ten percent that person I couldn't but the the you know professor therapeutic nihilism says I'm not convinced that increasing this septic person's cardiac output by ten percent matters I don't think their cardiac output was their problem they were raised a dilated even if I double their cardiac output I think most of their problem after their initial resuscitation when they're not like overtly having poor macros circulatory flow is more in their micro circulation and it's more kind of cellular and cytokines and whatnot so if a change that's so small that you need a doppler to detect it I don't think is going to be clinically significant it was already kind of a few steps away from what I think the central problem is here so you know it's day one they've had a handful of leaders most of these people I think we're playing in a realm of you know are they going to end up with another leader on board or another half a leader or or not and maybe that matters but how much what would you say to an argument like that where a lot of this is sort of I don't want to say academic but it's the the room for for changing care and moving the needle it doesn't seem all that much you know assuming you're coming from a a pre-test probability and kind of a modern sense of being somewhat somewhat reasonable about fluid management yeah I mean it's a very valid and appropriate question I think you're being difficult I think it's you know I agree with your your proposition that you know augmenting cardiac output is is that like is that going to make a difference for me I see the real value in the earlier breaks it's it's stopping the patient from getting crystalloid who is unresponsive but cryptically unresponsive because it's that patient that whose map goes from 52 to 54 on the first leader and then 53 to 55 on the second leader and then 52 to 56 on the third leader overnight and then it's you know the 70-year-old guy who's got a little bit of a brewing you know half-half from septic cytokines and also a smoke former smoker and now is on by-pad and in the step-down because he got that extra leader leader in a half that he didn't need and then he's on the by-pad and he's got a dry mouth and he gets delirious and he's screaming he pulls it off and he falls and he's got a head CT and a break like I think you know that is an anecdote but I think that you've probably encountered patients like that and so it's I don't think it's an unusual anecdote and I think at the population level like when you think about the arguments between balanced and and isotonic saline there was lots of you know person out the salt trial and it was these small effects but if we're giving two million liters of crystalloid every year then at the population level it matters and I I think you could make a similar argument like that like you know by doing this are you are you saving every patient's life no for sure not but are you saving maybe 12 hours in the ICU or are you saving maybe eight hours in the ICU if you if you rationally withhold that additional one and a half to two liters that they weren't responding to if you did that in 10% 20 percent of septic patients you're resuscitating just as you said sepsis is so common at the population level you know a really rational approach to fluids could save you know many days of precious ICU beds and ostensibly if you're reducing those sort of add-on downstream consequences that we see frequently in the elderly population you know like there is a benefit to it I think I think that there's good literature that that supports that it is mixed I will be the first to fully acknowledge that but at least you know one well well done randomized trial that showed that you know when you look for fluid responsiveness you give less fluids and you have fewer complications and I think it's it's those patients that have the low CVP the collapsing IVC they're unresponsive that that kind of get that extra two liters that they might not have needed that that maybe prolongs their their care in the hospital for an extra hours or days that that could be quite beneficial in in aggregate yeah in I just to hedge my own skepticism the times when I think about something like a cardiac output modern device in a septic patient which is certainly not every day is generally as we kind of mentioned more complex pictures mixed shock states they're septic but they're also in cardiogenic shock or something along those lines and then having some way to give you data and that's even you know even occasionally maybe a pa catheter those situations may may even make more sense because it can be so hard to parse out what is purely a matter of you know vasomotor tone versus maybe you can augment the flow in some other way even bread and butter sepsis there's there's very frequently a mixed picture you know even in young healthy people you can get these that these subtle diastolic dysfunctions now am I saying like you know are do we need to be super aggressive with you know flow monitoring in 35 year olds with hero sepsis like I've seen cases where where it would have mattered there's probably lots of cases where you don't need to be you know super precise with your fluid management but it's just like it's it's kind of common that that shock is especially in sepsis is frequently some kind of mixed picture like the myocardium is never perfect and in a really bad like this case that you describe but they're just gonna be so cytokine and and the heart's gonna be pissed off and in some way shape performs yeah and perhaps we should emphasize also that the fact that somebody's stroke volume increased by 10%, doesn't mean that you should give them more fluid. It is just probably a prerequisite for it, such that if it doesn't, then you probably should not. Doesn't mean that increasing at 10% is gonna help them or help them in a meaningful way. Doesn't mean that that increase will sustain over any amount of time. Doesn't mean that the harm is worth a benefit and so on and so forth. But if it doesn't increase at all, it's hard to imagine how it'd be good unless you have some whole other goal like increasing renal filtration or something like that. - Yeah. Yeah, that's why, like I said, at the outset, we have this very structured approach. Indication, yes, there's gotta be an indication to augment stroke volume first. If there is no indication, then who cares? If they're for the responsive. Indication, yes, safety, yes. They're not overloaded clinically or ultrasonographically, low-vexas, et cetera, et cetera. Safety, yes. And then the third question is, is it effective? And that's what you wanna see if you're engaging your sterling curve or not. - And if anyone's listening to this and they're like, my God, this is exactly what I need. Give me a hundred, just Google Flow Patch and they'll find your website. - I'm a bad salesman. Yeah, I guess it would be flowsonicsmedical.com or you can find me at heart-long.org and send me an email through that site and I'll direct you to the report pretty soon. - All right, well, thank you so much. Rest to you guys, we'll talk to you next time. (upbeat music)