The first question is, in settings when there's no thermodilution or swan gas catheter, can I use the FIC method? Jim, I'm going to give it a first try and see what you think about this. The answer is yes, sort of, but. The physiology says you can measure a venous saturation out of a central venous catheter. Central venous saturation correlates with mixed venous saturation, but isn't quite the same thing. You can't do it from peripheral catheter, and to me, the variability of that is much higher. So, you could argue that you're just taking a trend, that that trend ought to go in the same direction, but remember, mixed venous oxygen saturation is mixed, and the central venous number depends on how much is coming from the IVC, and how much is coming from the SVC, and where is the catheter, and then you have that same variability with oxygen consumption that you have. So, in my experience, you can do it, but it's not nearly as good as trying to do it with an appropriately placed mixed venous catheter, which is usually in the pulmonary artery. Jim, your take? Yeah, no, I completely agree with your comments, Steve. It's really only useful when they're really crazy numbers. Like if you were to get an SVC or IVC SAT being a SAT of 20, I think that's meaningful. Anything from arguably 50 or 60 and higher, it's again, very difficult to know how to interpret that. Your comment about trends, I think, is also relevant, and you always have to figure out when the nursing staff drew the sample, because if it was drawn when they were brushing their teeth, you know, versus completely zonked out after their morphine, you know, it's tough. But in patients in whom there's a diagnostic challenge, and they're not fully instrumented for invasive monitoring, a sample like that can give you some insight. Yeah, I agree with that. I would also just inject a little more personal bias. I mean, yes, if your mixed venous saturation in your central venous catheter is 20, you're in big trouble, but that's the time when you want accurate information. So if your mixed venous saturation from the central venous catheter is totally normal and you just kind of got it because, fine, I can go with that number, I don't necessarily need to bury a PA catheter, but if my patient is so sick that the mixed venous is 20, yeah, I know I'm in big trouble and I need to start doing stuff, and maybe I want to do that based on an accurate measurement. We have one, we have another question about the use, and David, do you want, you have a take on this? I'll give you a shot. I guess the question is, you know, how much do you care about the accuracy of the absolute number of cardiac output at a given time, right? So if you just know that there's bias, and this is true of any monitoring device or really any test that you do in medicine, where you just have to know what the bias is, and is the bias going to be a lot or a little, right? Is the number that you're going to generate close to or far away from the true value, assuming you can imagine what the true value is, and is that bias going to be always in one direction, or is it going to vary, and is it going to move around the true value over time? I agree with James. I think that, you know, if you get a central venous oxygen saturation that's 42%, right, I don't think you need to know what the cardiac output is per se, right? You know that this person has a mismatch of oxygen delivery, or rather oxygen demand and oxygen delivery, right? And then, you know, you can make some assumptions based on your clinical exam or other tests about whether the right next step is to try to improve cardiac output in some way and what you're going to do to improve cardiac output and oxygen delivery. So I guess it's a long way of saying you could try to use it, but I don't know how valuable getting a number of 3.2 or 4.1 is when you're unsure of how much that's going to deviate from some true thermodilution value that you've gotten with a pulmonary artery catheter. Excellent, thank you. All right, so second question has to do with the use of MRIs in the ICU with LV diastolic heart failure. That actually goes to Lakshmi's talk, and I'll start and see if you guys have additions. The data with mineralocorticoid antagonists does not come from the ICU. It comes from the outpatient facility, and strictly speaking, the trials in the outpatient are currently negative, although the field is kind of moving. And if you look at the range of people with what would be considered diastolic heart failure, that is EF, say, 40 and up, and you look at the patients in the lower end of that range as opposed to the patients in the higher end of that range, then you get a little more benefit in the lower range, and we're talking about, and the term is mildly reduced. In the ICU, kind of all bets are off. It seems reasonable to use an MRI if you don't have anything else. There are data on SGLT2 inhibitors, but I think it's reasonable but not currently data driven. Welcome, Lakshmi. Lakshmi was doing an M&M. So the question is, what do you think about the use of MRAs in diastolic heart failure in the ICU? Great, and thank you. Yeah, so in the cardiac ICU, we frequently use MRAs in diastolic heart failure, and even in patients who are hypertensive, we're supporting them with inotropes and other agents, and we typically do that in the context of the fact that we're aggressively diuresing them, and it sort of helps us also with their hypokalemia, quite frankly. And so oftentimes, you'll see many of our patients on higher doses of MRAs than you might consider in other circumstances when they're hypertensive, but really, it's because we're supporting their blood pressures and we're diuresing them pretty aggressively. I think, to Steve's earlier point, it can be used as long as you're using it judiciously and you're monitoring their, you're very close about monitoring what's going on with them from a blood pressure perspective. To be aware that, of course, the spironolactone at lower doses doesn't have a significant sort of consequence immediately, of course, to their blood pressures. Excellent. So, David, I have a question for me, and this one is philosophical. So without being specific about any particular monitoring device, it's clear that different monitoring devices have different strengths and weaknesses. So if you're deciding how, if you decide you're going to start doing this in your unit and you're not currently doing, do you think that a range of devices with different strengths and weaknesses is a good idea, or do you think that you probably ought to pick a winner and get good at it and kind of really develop a lot of experience with one device before you decide whether to go with more? And this is kind of a generic question. It's not restricted to monitoring, but I'll let you take a crack at it. Yeah, that's a great question. I think, in my experience, the amount of time and effort required to learn to use any of the monitors currently available on the market well is small. They're all virtually plug and play. Now, that has difficulties, right? Because then they kind of are black boxes that generate numbers, and then people just kind of act on numbers without sort of understanding what the physiology being measured is. I think the monitors have different niches along the continuum of care. I mean, I think that for the very sickest patients where they have multiple organ failure, they have persistent shock, they're accumulating fluid and going into ARDS, I think monitors that let you know what's going on with vascular permeability and how much fluid is accumulating in the lungs are really helpful in making clinical decisions. And they give you continuous information about cardiac output and stroke volume, and they give you information about the permeability of the pulmonary vasculature several times a day, as often as you want to do thermodilution. If you're recovering patients from complex surgery, and they're all going to have an A-line anyway, then maybe one of the radial A-line based devices, which helps you intraoperatively and immediately postoperatively, just sort of know, you know, there are good data from the perioperative world that suggests that keeping people sort of topped up on that fluid, on that knuckle between fluid responsiveness and fluid unresponsiveness, probably helps prevent renal failure. You know, so that's a suitable niche there. And then looking at what you can do, say, in the hospital wards in the emergency department, when you're triaging patients, trying to understand if giving that third bolus of of crystalloid for somebody who has persistent hyperlactateemia or persistent hypotension in the emergency department is helpful. I think one of the highly non-invasive devices that can be placed on the patient and calibrated in three minutes are probably helpful, because you can do a passive leg raise virtually anywhere. And knowing, is my patient's cardiac output likely to be low? And is my patient going to respond to more fluid when you sort of don't know the patient's likely volume history? I think those are useful, because you can, like I said, you can set it up and it's plug and play in three minutes. And so in a busy environment, like a rapid response team or an emergency department, that's a suitable alternative. So I think that, you know, like buying a car, buying a bicycle, there's just lots of different models out there that are going to fit different people's needs at different times. Yeah. Other thoughts or should we move on? Now we can move on. Okay, I got one more question. Nothing like an easy one. When should we start to transition from fluid resuscitation to minimizing fluids, planning for extubation, waking up the patient, rehabilitating? Is this part of the A to F bundle? Who wants this one? At the right time, huh? Yeah, exactly. I mean, that's the, that's, you know, there's that old, trying to, the name of the, there was a traumatologist back in the 30s and 40s who sort of coined the ebb and flow of, you know, sort of acute care, right? There's this, there's this transition that people tend to make when they're getting better when they move from a catabolic state to an anabolic state. And I wish we had the ability to identify that moment, but that would be the moment to start de-resuscitating, right? And probably there isn't one moment. There's probably a series of moments and different organ systems switch from ebb to flow at different times, depending on the patient's background, what medicines they're receiving, what medicines they came in on and so forth and so on. Do they have hypertension? Do they have chronic kidney disease? Nephrotic syndrome and so on and so forth. So it's a very, very, like the person who figures this problem out gets a call from, you know, Stockholm one evening in October, right? That kind of level of thinking. I think the most commonsensical way to think about it though is we should always be thinking about whether we should be careful to give more fluid or even start to take fluid off. And I think that there's a strong argument to be made and folks in our cardiac ICU here at NYU are thinking about adopting some of these continuous cardiac output monitors to help understand when it's safe to start diuresing or when it's potentially unsafe to continue diuresing, right? Because you don't want to take a ventricle that's performing poorly and then push it from the flat part of the Frank-Stolling curve down onto the steep part. So an answer is kind of the 180 degree question of the one we typically think of trying to answer with these monitoring devices. But I think that, you know, just knowing physiology and where you want your patient's stroke volume to be at any given time and whether you want to give diuretics or try to increase preload at any given time, it's clinical judgment and it's using physiology, it's using physical exam and using lab tests and monitoring devices. Yeah. And to give, I mean, that's a great answer. And to give my take, I mean, it's easy to say at the right time, but when the patient stops getting worse and starts getting better is the time, you're absolutely right. The key, if you're lucky, you might anticipate that. And, you know, maybe when they stop getting worse, we can try and maybe make them get better a little faster by taking away some of that fluid. And, again, you might get it wrong and have to give some fluid back. It's not the worst thing to give it a try and see what happens. But to your point, Steve, it's very common in an ICU as the intensivist changes, right? That the nurse is like, we gave fluids for the last 24 hours and now Dr. Fang comes on and now we're diuresing for the next 24 hours. I have to admit, it drives the nursing staff crazy because all of us, you know, view the same patient and trajectory in different ways. And the way we basically assess that is, you know, we fluid resuscitate it until you think that the risk of that is greater than the benefit. And if the patient's getting better and there is no, and frankly, there is a little risk to stopping that than you do. But if a patient's still tenuous and you're giving volume, the risk of giving volume outweighs, I think, any benefit. So that's how we generally do it. But I think these patients are so heterogeneous, I think that's really the bottom line, right? There's hard, it's really hard to have one size fits all. I mean, I can tell you, even as a cardiologist, I'm struck by the fact that we've been unable to prove that, you know, invasively hemodynamic managed patients improves overall outcomes, although I still do it. Because our MICU attending... Little observational data in cardiogenic shock, that for what that's worth, it says, if you use it, patients do better. Maybe that just means that you pick the right patients or they're smarter docs or whatever that means. Doctors are pretty good at figuring out who's not going to do well. And you're like, oh, what's the whole point of putting a line in? So anyway, I think good comment. All right. Excellent.

ICU monitoring

fluid management

central venous saturation

ScvO2

mineralocorticoid antagonists