Well, thank you for that introduction, and thank you for having me attend today. It's nice to have a heart failure cardiologist in the room. I've learned a lot already during the day. And I've been charged with talking about how to implement changes in therapy, but really three aspects to this. I was asked to talk about risk assessment, a concept called clinical trajectories, and then implementing therapies during hospitalization. A lot of what we discuss comes from this expert consensus decision pathway that's written in conjunction with the ACC, led by Dr. Hollenberg as the chair and Dr. Stevenson as the vice chair. I had the pleasure of being on the writing committee with a number of other people in the heart failure field. And you'll see that within this ECDP, they talk about risk assessment, implementing therapies or management, and clinical trajectory, which is really how the outline of the talk will go. So let's first start with risk assessment. What I wanted to do is really focus on an estimate of hemodynamics. You've already heard a lot about that in this session, but really focus in on clinical examination. And this is what's been called the two-minute classification, or by Lynn Stevenson, where you apply the estimates of the hemodynamics to the bedside, and specifically whether the patient's congested or whether they have a low output or not. So for example, in the top right-hand quadrant, patients are congested with adequate perfusion. They're the warm and wet group. Whereas in the bottom right-hand quadrant, they're the cold and wet group, elevated filling pressures but low cardiac output. So the question is, how do you do this if you don't have a Swann-Gans catheter? How can you see your patient and place them into one of these quadrants? On the top right-hand, you'll see a number of the signs and symptoms that we use to decide whether a patient's congested or not. The first I'd like to focus on is the jugular venous distension. This is a quote from Jay Cohen, who was the founding father of the HFSA. And he said that the JVP is the single most important marker of the status of intravascular volume. And really, you cannot care for patients with heart failure unless you can estimate their jugular venous pressure, in my opinion. If you think about the history and physical exam as a way of gathering a number of variables, whether it's 20, 30, or 40 variables, it turned out that there were only two variables that were associated with an elevated wedge pressure, here defined as greater than 30. And those two variables were an elevated JVP and the presence of orthopnea. You can see the odds ratio of the JVP was about 4.5 and orthopnea about 3.5. So if you really want to boil it down when you're seeing a patient and quickly decide whether you think the filling presses are high or not, you can zoom right in on the JVP and whether they have orthopnea or not. Now, one question to consider is, I know a lot of you are interested in hemodynamics, why do we care about the jugular venous pressure if that is an estimate of the right atrial pressure, while what we're really trying to get at is what is the left ventricle filling pressure or the wedge pressure? And I think the reason that is is shown here. These are 1,000 patients who had undergone right heart catheterization. And then we just stratified them based on whether the wedge pressure is high or not and whether the right atrial pressure is high or not. And what you can see in those quadrants in green, these are concordant groups. That is, the right atrial pressure is mirroring the left atrial pressure. If the wedge is high, the right atrial pressure is high. If the wedge is low, the right atrial pressure is low. And you can see about 780 of the 1,000 or about 78% of the time, the right atrial pressure does mirror the left side of filling pressure. If you estimate the JVP to estimate the right atrial pressure, that should give you an accurate estimate of the left side of filling pressure. On the other hand, in about 20% of the time, there's discordance. That is, the right atrial pressure is not a mirror of the left side of filling pressure. And so if you look at the neck veins, and for example, you see high neck veins, and they happen to be in the bottom right-hand quadrant, you'll think the wedge pressure is really high, but the wedge pressure is not as high as you might anticipate. And so you have to remember that when you're looking at the neck veins, it is not an absolute direct measure of the wedge pressure. You're going for this relationship between the right and left atrial pressure. Now another symptom that has been relatively recently defined that some of you may not be aware of is something that's called bendopnea. And this really emanated in our research group, where we noticed that patients were coming to us telling us that they were short of breath when they were tying their shoes. And this led to this heated debate in our research group about, could this actually be a symptom of heart failure? Now I live in Texas, and we happen to have a lot of people who are obese, and there was a lot of us who thought maybe this was just central obesity causing them to be short of breath. But my colleague, Dr. Thibodeau, who was a fellow then, now a close colleague of mine, set out to study this. And we defined this as dyspnea within 30 seconds when bending forward, for example, when putting on shoes. What she found was that this actually was surprisingly common. Almost 30% of patients with heart failure had this symptom. It happens very quickly. The median time to onset is about eight seconds, all the actions over by about 11 seconds. And when she tested it in the holding area and then tested it again during the catheterization itself on the same day, there was 100% agreement. So at least on the same day, it was reproducible. If the patient reported bendopnea in the holding area, they had it when they were undergoing the catheterization. If they didn't have bendopnea in the holding area, they didn't have bendopnea when they were catheterization. So at least it's a reproducible phenomenon. This is the key data slide from that publication. What you're seeing is the sitting and bending pulmonocapillary wedge pressures. The red are the participants who have bendopnea. Blue are participants who did not have bendopnea. And you'll see that even in the sitting position, the patients who had bendopnea had higher pulmonocapillary wedge pressure. When the participants bent forward, the wedge pressure elevated similarly in both groups. But since the patients who had bendopnea started higher, they ended higher. And we postulated that perhaps they crossed a dyspnea threshold. And that's why they reported this symptom of shortness of breath when they bent forward. It is irrefutable at this point, however you estimate filling pressures, whether you look at the neck veins, bendopnea, peripheral edema, ascites, however you decide the patient's clinically congested. It is irrefutable at this point that that is a marker of increased risk. These are data looking at the SOLVE database. This was looking just at the jugular venous pressure on study entry, comparing subjects who had that or who did not have that, followed them for four years. And you can see that those who had JVP on study entry were at a much higher risk of death or heart failure hospitalization over the next four years than those individuals who didn't have JVP. These are data from the late Mihai Georgiadi from the Everest trial. Here they calculated a congestion score on discharge or day 7. They gave 0 to 3 points based on orthotomy, a JVD, or edema. And then they just looked at the subsequent risk of these clinical endpoints over the next 10 months. This is about 2,000 subjects. And if you focus here, for example, you can see that as congestion score increases from discharge to day 7, there's over a twofold increased risk of death over the next 10 months and almost a 50% increased risk of the composite of heart failure hospitalization or death, again, over the next 10 months. Clinical congestion identifies a patient with increased risk. These are the best data, in my opinion, that solidify this thesis. These are data from the paradigm investigators. What you're seeing here are four different outcomes. The top left is cardiovascular death or heart failure hospitalization, then cardiovascular death on the right, heart failure hospitalization on the left, and all-cause mortality in the bottom right. In blue is the analopril, and in red are the ARNI patients. And what you can see is that as you increase the number of signs of clinical congestion, this is JVD, estuarals, or edema, you can see that the risk of each of these endpoints increases. And what was remarkable is that this clinical congestion marker of increased risk held up independent even after you adjusted for natriuretic peptide levels. So these associations persisted despite adjustment for natriuretic peptide levels. So we really don't need any further data on this. Any of your patients who are congested are at increased risk. Now what about this axis? What about the low perfusion or low cardiac index? Well, when we looked in the ESCAPE database, it turned out that the only variable that was associated with the reduced cardiac index was the overall gestalt by the investigators that the patient or the participant was cold. If they were cold, the cardiac index was 1.75. And if they were warm, it was 2. There was no other variable in the HNP that identified a low cardiac index in the ESCAPE database. But it's very important we go back and we look a little deeper and look at the sensitivity of the clinical exam in this experience. If you look, for example, at cool extremities or a narrow pulse pressure or even this overall assessment of a cold profile, the sensitivity is about 20% to 30%, very poor. And so on a routine basis in the hospital, a patient whom I think is warm, I send down for a right heart catheter and they have a low cardiac index. It happens over and over again. In our hands, we cannot detect a low cardiac index on physical exam. On the other hand, when the patient does seem like they're in a low index, they are in a low index. It's good specificity. So I've come up with this saying which I actually think is very useful for clinical purposes. If they are cold, they are cold. But if they are warm, they may still be cold. If they are cold, they are cold. But if they are warm, they may still be cold. As you see patients with heart failure, if the kidneys are falling apart in you, you think they're well perfused, you may not even recognize that they're in a low output state. You just have to recognize this is a limitation of the clinical exam and patients who you think are well perfused still can have a low cardiac index. Now these are the best data that are fairly recent from the European Society of Cardiology Heart Failure Association where they took 8,000 hospitalized patients and then they just looked at the admission profile. And what you see in green are the patients who are warm and dry on admission. The red line are the warm and wet group and the blue line are the cold and wet group, the low output and congested patients. And you can see that the patients who are both congested and then low output were at the highest risk. And this held up in multivariable analysis where you're seeing in the top bar here, these are the patients who are warm and wet. They had an increased risk. The second line is the cold and wet versus the dry and warm group. They had increased risk. And the cold and wet even was at increased risk compared to the warm and wet group, meaning that the patients who the investigator thought was low output and congested had a worse outcome than the patients who were congested but had a perceived adequate output. All right, so risk assessment, congestion, and you can quickly ascertain the volume status and the state of perfusion using the clinical exam. Of course, the right heart catheter gives us the definitive answer, but many times you don't have that right heart catheter available and this is useful information. The next concept I wanted to discuss was this idea of trajectories that was developed in the ECDP. And there were three trajectories described. The first was what's called improving towards target. These are patients who are getting decongested. Things are going well. You'll be able to discharge them on GDMT. Everything is going well. But there were two other categories where things are not going the way you want. The orange bar is initial improvement and then stalled. And then the red box is not improved, worsening. And so most of the patients who are going to end up in the ICUs are probably going to be in the stalled or the worsening group. The ones who are improving towards target, things are going well. They may never make it to the unit. And when we say improving towards target, what we really mean is that patients are decongesting. So there's freedom from peripheral edema, no RALs, no JVD, no orthopnea bendopnea, and no hematopoietic reflex. These are all patients who are decongesting. Things are going well. But if you can't decongest your patient, the signs and symptoms are not improving, the weight's not going down, the natural peptides are not going down, this is when they're stalled or potentially even worsening. And in this publication, we listed a number of reasons why patients may fall into those categories, including a low output state, or maybe this dominant right-sided congestion pattern where the right-sided filling pressures are higher than they should be relative to the left-sided filling pressures, or advanced renal disease, or other potential causes of why patients cannot be successfully decongested. Now this is within the ECDP. These are the strategies that were advocated for how to address these patients who either stalled or are worsening. And really, it boiled down to these four strategies. You can escalate your decongestive strategies, which I'll talk about in a minute. You can consider alternative diagnoses. For example, not uncommonly, I'll be caring for a patient with amyloid, can't decongest him, and only then recognize that they have bladder outlet obstruction, and that's why we can't decongest them. And so you want to consider alternative diagnoses as to why you can't decongest your patient. Consider assessment of invasive hemodynamics. If the patient has not improved or is stalled or worsening, that's when a right heart catheter would be reasonable, and you may need to add inopressors or even temporary circulatory support. And then call for help. Consult cardiology if you have access to advanced heart failure cardiologists. If the goals of care align with palliative care, involve the palliative care team. But it's this group of patients that's at risk of having adverse events. All right, the third aspect, then, is implementing therapies during the hospitalization. And really, there's three different advances that are relevant for the discussion today. The first is diuretic strategies. The next is SGLT2 inhibitors. And then something that my colleague, Bikam Bozkart, has poetically named the recipe. How do we put all this data together now? How do we put all the medical therapy together for our patients? So the first one is diuretic strategies. What to do when your patient's on loop diuretics, and you're not adequately decongesting your patient. Here, there are two trials worth mentioning. The first is the ADVOR trial. This was published in 2022, I believe, in the New England Journal, about 500 patients. This is a randomized placebo-controlled trial. Patients were randomized to loop diuretic alone, or loop diuretic with IV acetazolamide 500 milligrams for three days. And this is the primary endpoint of the trial on the left. And the primary endpoint was successful decongestion within three days of randomization. And you can see here that those who got the acetazolamide were more successfully decongested than those who did not get acetazolamide. The top right is looking at the congestion score, showing that the congestion score is falling to a greater degree with acetazolamide. And then the bottom right is looking at discharge, showing that those who had gotten acetazolamide were more likely to be successfully decongested at discharge than those who had not gotten acetazolamide. But there are some cautionary notes. The primary endpoint was just looking at a congestion status at three days. If you go look at the hard endpoints within the trial, death at three months, there was no benefit. And numerically, actually, it was slightly higher than the acetazolamide group. And then there was no benefit on hospitalization. This is a small trial. It was not a trial for clinical outcome endpoint data. But at least we are not even seeing a signal there. Just something to keep in mind, even though you're successfully decongesting the patient, we don't really have the outcome data showing that routine addition of acetazolamide is going to improve hard clinical outcomes. The other trial that was published recently was a chlorotic trial. And this was in the European Heart Journal from last year. And the bottom line in this trial is that they added hydrochlorothiazide, oral hydrochlorothiazide, from 25 to 100 milligrams to IV in addition to loop diuretics, randomized fashion. And although it improved the diuretic response, there was a much higher risk of worsening renal failure, almost threefold higher risk. And so this is really pretty worrisome in terms of incorporating this strategy into routine clinical care when you have a threefold increased risk of important worsening of renal function. Well, the second update, then, is on the sodium glucose transport inhibitors. And here we have a couple trials worth discussing. The first is soloist worsening heart failure. This was in the New England Journal in 2021. So citagliflozin inhibits not only SGLT2, which is in the kidney, but also inhibits SGLT1, which is in the GI tract. This was a trial of type 2 diabetics who were hospitalized. They all were getting IV diuretics. And unfortunately, the sponsor pulled the funding. And they were expecting to have 4,000 subjects. But they only had 1,200 subjects when the trial was terminated. They were followed for nine months. And even though it was terminated early, this is a primary endpoint of cardiovascular death visits for urgent visits or hospitalization. And there's a 33% risk reduction when there was additional citagliflozin to routine therapy for the hospitalized patient. This was then mirrored by the MPULSE trial. This was a trial that came out after soloist worsening heart failure. Again, a fairly small trial, 500 patients hospitalized with heart failure, who were randomized this time to impagliflozin versus placebo, followed for 90 days. And the investigators used what's called a stratified win ratio, where you look at a hierarchical variety of endpoints. And you first look, for example, at time to death. If someone wins, if impagliflozin wins, it goes in that category. If placebo wins, it goes in that category. If there's a tie, then you look at the second endpoint. In this case, this was heart failure hospitalization. If there's a tie, you go to the third endpoint, which is a Kansas City Q score. It's a way to increase the statistical power of a primary trial. And in MPULSE, impagliflozin won. The win ratio was significantly favorable, 35% increase. And if you look at each of the individual components, it improved upon death, heart failure event frequency, as well as the KCCQ. So fairly impressive data, even though it's a relatively small trial, suggesting that we should be using sodium glucose transport inhibitors in our patients with decompensated heart failure. So how do we put all this together? What is the recipe? Well, the trial here that really guides us, at least as of now, is this trial in Lancet called the STRONG-HF trial. This was 1,000 patients who were hospitalized. And then they got randomized to usual or high intensity care. High intensity care was triple therapy, GDMT. This was done before SGLT2 inhibitors came on the scene. Participants who got randomized to high intensity care were made sure they were on half dose of triple therapy, GDMT, before discharge. And then they were rapidly up titrated within two weeks after discharge. And then they had several safety visits over the next two months. And so as opposed to a routine strategy, where you kind of slowly implement GDMT, and then you take your time in the outpatient setting, the high intensity group got half dose GDMT before they left, and then rapidly up titrated, hopefully within two weeks. Primary endpoint here was a composite of readmission for heart failure or all-cause mortality at 180 days. And this trial was terminated early by the DSMB for efficacy. And this is the favorable impact shown here of rapid up titration of GDMT. What you're seeing in the red line is the high intensity group. And in the blue line is the usual care group. And you can see there was a significant improvement, an 8% absolute risk difference, and a 180 day endpoint of all-cause hospitalization or death. Very important data to be aware of. There were other benefits seen within this trial. This was including reducing heart failure readmission, improving quality of life, reducing clinical congestion, reducing natriuretic peptides. There was some more hypotension, renal impairment, and hyperkalemia. But overall, the heart clinical outcomes clearly swamped those concerns, really suggesting this is the strategy we should be pursuing. It really doesn't come as a surprise that therapy implementing GDMT should benefit. This is a decision analytic model that was published a few years ago, looking at the estimated benefit of sequentially stacking quadruple therapy GDMT. And what you can see is that as you stack GDMT, ultimately you get down to quadruple therapy leading to a 26 absolute reduction in all-cause mortality, which for medical therapy is huge. So really, it's incumbent upon all of us to get our patients on GDMT and then up titrate them in the outpatient setting. So I would conclude here, the risk assessment, clinical congestion, clinical assessment of hemodynamic status provides important prognostic information. Clinical congestion is a marker of increased risk. This is really irrefutable at this point. JVP is useful, probably the most useful. But orthopneum and dopamine are helpful. Remember that the sensitivity for our estimates of low output is poor. So if they're cold, they're cold. But if they're warm, they may still be cold. And then the cold and wet profile, probably a number of the patients that you're seeing in the ICU, these are the ones that are at the highest risk. In the ECDP, we talked about three trajectories. Two of them when things are not going as well as we hope, the stalled category or the non-improved worsening. And we discussed four strategies to address these groups. Escalate your decongestive strategies, consider alternative diagnoses, define your hemodynamics and basically enact on that, and then call for help, consult cardiology, advanced heart failure, or palliative care. And then in terms of implementing therapy, there's been recent data now with augmenting loop diuretics. It looks like acetazolamide may be reasonable, but oral hydrochlorothiazide probably not. Really the most dramatic advance is with SGLT inhibitors now. It improves outcomes. And I suspect, if not already, very shortly, when the patient's seen in the ER and gets their dose of IV diuretic, they're going to get an oral SGLT2 inhibitor, or a sodium glucose strontium inhibitor. That's not yet state of the art, but I suspect that will ultimately be in guidelines fairly shortly. The recipe, there really is a paucity of data telling us how and when to do all this, but I think it's now emerging. It's safe to put SGLT2 inhibitors in the acute setting in decompensated patients. And the goal is to get your patient at least on half dose therapy, if you can, before discharge, and then have your outpatient colleagues rapidly up titrate that, so we can accrue those dramatic benefits of quadruple therapy for our patients with heart failure. And with that, I'll end. Thank you so much. Thank you.

heart failure

risk assessment

SGLT2 inhibitors

clinical trajectories

diuretic methods

GDMT titration