Hi, I'm Steve Hollenberg from Hackensack Meridian University Hospital, and it's a pleasure to be able to give this talk. I have no conflicts of interest to declare. The prognosis may be a little better than it had been in the past. This study comes out of 2001, looking at data in 1991, but the point remains the same. What you see is heart failure mortality in the bright orange, and that is worse than most cancers with the exception of lung cancer. This is the textbook definition of heart failure, a clinical syndrome caused by inability of the heart to supply blood to meet tissue metabolic requirements. If you think about it, this really implies that there's perfusion failure in heart failure. This is a study that looked at perfusion failure in patients with heart failure. What you see is lactate plotted against arteriovenous oxygen difference, and lactate plotted against cardiac index. And what you see is that most of the patients don't have an elevated lactate in heart failure. And over to the right, you have a pie chart. A vast majority of the patients have class 1 to 3 heart failure, and really none of those have perfusion failure. The only ones with perfusion failure are those in class 4 heart failure, and even of those, only about a quarter of the patients in class 4 heart failure, or 1% overall, met the historic definition of a failure of the heart to deliver adequate blood supply to meet metabolic demands. The other thing to think about with the previous staging is that there were stages, and then there was a functional classification. So you had a functional classification. Class 1 had no symptoms. Class 4 had symptoms at rest. Class 3 had symptoms with mild exertion. And there were two categories of class 3, and class 2 had symptoms with moderate exertion. And then there were stages. Stage A was at risk. Stage B was asymptomatic left ventricular dysfunction. Stage C was symptomatic heart failure, and stage D was sort of an end-stage heart failure with symptoms at rest. So the new universal definition goes away from that. And here it is on this slide. Heart failure is a clinical syndrome with symptoms and or signs caused by structural and or functional cardiac abnormalities and corroborated by at least one of the following. Either objective evidence of pulmonary or systemic congestion or elevated natriuretic peptide levels. The stages of heart failure have also been modified at least a little bit. Stage A is still patients at risk for heart failure, and stage B is now called pre-heart failure, but patients with either structural heart disease or abnormal cardiac function or potentially elevated natriuretic peptide levels without symptoms. Stage D is also similarly defined as before, advanced heart failure with severe symptoms and requiring advanced therapies. And the difference here is in heart failure in stage C. It is patients with current or prior symptoms, and they do have to have a structural and or functional cardiac abnormality, but then they go into two slightly different categories. They either have persistent heart failure, but once you get to stage C, you stay in stage C, but you might be in remission. That is, risk factor modification and guideline-directed medical therapy might remove your symptoms, so you're still stage C because you've been symptomatic once, but you're currently in remission. The ejection fraction classification has also been changed. There's reduced ejection fraction, less than 40 percent, and there's preserved ejection fraction greater than 50 percent, and then there's this intermediate category, heart failure with mildly reduced ejection fraction, which is somewhere in between. Finally, there's a new category called heart failure with improved ejection fraction, defined as heart failure with a baseline left ventricular ejection fraction, less than 40 percent, but a subsequent greater than 10 percent increase such that the second measurement is greater than 40 percent. In summary, you have a standardization of the heart failure syndrome definition. You have a new revised classification, heart failure, that's hopefully easier to understand. It talks at patients at at-risk, pre-heart failure, heart failure, and advanced heart failure. You have a different EF classification with an emphasis on improved but not recovered ejection fraction, and then you have trajectories. You have persistent heart failure rather than stable heart failure, and you have the potential for heart failure remission rather than recovered heart failure. Now let's move on to management. Management of heart failure has a number of components. You assess the acuity, acute or chronic, or maybe acute on chronic. You establish the diagnosis. You define the syndrome, heart failure with preserved ejection fraction, reduced ejection fraction, or maybe intermediate, and the etiology. You correct precipitating factors for decompensation, the most important of which is evaluating for and correcting ischemia at present, and then you think about the treatment goals. You want to improve the symptoms. You want to reverse the pathophysiology, and you want to improve the outcomes, and not all of the therapies do all of these things. The historical perspective of the pathogenesis of congestive heart failures is that CHF started as a hemodynamic model. There was preload. There was afterload. There was contractility, and you addressed those to address heart function. And then we moved on to the neurohormonal modeling, which we'll talk about, looking at the neurohormonal influences that perpetuate heart failure. And now we've moved on to a model that looks at remodeling of the structure of the heart as an indicator of long-term outcome, as you will see. There are compensatory mechanisms in heart failure. As ventricular function deteriorates, the heart relies on a variety of adaptive mechanisms basically designed to maintain adequate cardiac output. There's the Frank-Starling law of the heart, the renin-angiotensin-aldosterone system, and the sympathetic nervous system. So this is the heart failure therapies defined by pathophysiologic rationale. You have myocardial failure. On the right side of the slide, you have decreased cardiac output and increased left ventricular filling pressures. On the left side, you have increased afterload and increased preload with vasoconstriction and systemic vascular resistance. And then you have the consequences of compensatory mechanisms. So you might address the hemodynamics with inotropic agents. You might address the compensatory mechanisms by blockers of the renin-angiotensin-aldosterone system or beta blockers. You might address the afterload with arterial dilators and the preload with venodilators. Let's start with diuretics. This is just to remind you that different diuretics have different sites of action, be they in the ascending loop or the distal tubule or the proximal tubule. Vasodilators are classified by their site of action as well. There are those with predominantly venous vasodilatory effects, such as nitrates. There are those with predominantly arterial effects, such as minoxidil or hydralazine. And then there are the mixed vasodilators in the middle, including the renin-angiotensin system blockers. Nitrates are first-line agents for angina symptoms and also for myocardial infarction with CHF. Because they're venodilators, they reduce the preload and they seem like a good idea given the fact that the prevalence of coronary disease is high in heart failure patients. If you're in the ICU, I think you probably ought to start intravenously at a low dose of something like 5 micrograms and titrate upwards. The side effects are headache and occasionally hypotension. And if you see marked hypotension in response to a small dose of nitrates, a little bell should go off and you should ask yourself whether you have right ventricular ischemia or infarction. There's also a little bit of withdrawal with nitrates. And they can be combined with hydralazine for chronic therapy of heart failure. The first data for that come from the V-HEF trial, the first trial to show a beneficial effect of anything in heart failure. And here you see the mortality of heart failure at 42 months and it's quite substantial, up to 50%. But the combination of hydralazine and isosorbide was better than placebo and in fact better than prazosin, something that was used at the time, which had a hemodynamic effect but no effect on outcome. Further data on hydralazine and isosorbide came from the AHEF trial. This is a trial in self-declared African Americans and these patients were a different cohort. They had New York Heart Association class 3 or 4 with low ejection fraction and they were reasonably well treated. 90% were on diuretics, about 85% on ACE inhibitors or ARBs and 75% on beta blockers. And you have to watch the Y-axis here, it goes from 100 to 85%. But the isosorbide hydralazine was superior to placebo with a 43% reduction in mortality. This will remind you of the mechanism of action of renin-angiotensin system blockers. ACE inhibitors block the conversion of angiotensin 1 to angiotensin 2, but also block the conversion of bradykinin to inactive kinins. That increases nitric oxide, but also leads to potential for cough mediated by bradykinin. Angiotensin receptor blockers work downstream and they block the angiotensin 2 receptors. You get vasodilation and decreased blood pressure with both of them. Here's the effect of ACE inhibitors on congestive heart failure. They decrease mortality and CHF mortality and the combined endpoint of death or heart failure hospitalization. Now one of the things that keeps people from at least titrating ACE inhibitors is renal function. So I want to show you a little. So this is a subset of the SOLVE trial, which is an early study of enalapril for heart failure that shows what happens if you have worsening renal failure after initiation of an ACE inhibitor. If you're in the placebo group in the dotted line, worsening renal function is bad and you do worse. However, if you're in the enalapril group and you have worsening renal function, you actually don't do worse and there's sort of a little suggestion you might do better. So let's talk a little bit about practical points on use of ACE inhibitors in heart failure. You want to titrate steadily to target doses. Your expected decrease in blood pressure is not the same as treating hypertension because when you decrease afterload, the cardiac output goes up and that mitigates the hypotension. So actually in the SAVE trial, the blood pressure decreased, actually increased going from baseline to captopril three times a day. By the time you got used to the ACE inhibitor, your blood pressure went up and not down. And as I showed you in the last slide, a small increase in serum creatinine is expected and may correlate with better long-term outcomes. If you have a big increase in serum creatinine and you're heading toward dialysis, you need to stop them, but you'd really like to try to ride out that initial small increase because down the road things will get better. How about the angiotensin receptor blockers? Here's the VALHEF trial and something called the CHARM trial with Candesartan, both showing improvements in heart failure hospitalization and the combined endpoint of death and heart failure hospitalization with Valsartan and or Candesartan. Aldosterone antagonists also work. I showed you the EFASYS trial in myocardial infarction before, but these are the RALS trial in class three, four heart failure patients with spironolactone and the EMFASYS heart failure trial in patients with predominantly class two, a couple in class three, LVEF less than 35% with a plurinone. And there's statistically significant reduction in mortality with both of those agents. Now there's a worry about hyperkalemia with aldosterone antagonists. You can hear the data from the high risk patients in EMFASYS and what you see is that the hazard ratio, the reduction in mortality was 37% so it's really good in these patients. And there was a slightly higher increase of potassium greater than 5.5 with a plurinone, a slightly higher, didn't quite make statistical significance of potassium greater than six, but in fact no differences in hospitalization for hyperkalemia or worsening renal function. Now you have to be careful with the potassium, you have to check the potassium every week after you send the patient home from the hospital and make sure it's not going up. But carefully, you can actually initiate and maintain aldosterone antagonism. Your starting dose of spironolactone is low. It's not really a diuretic dose, it's 12.5, 25 milligrams a day, a plurinone is 25 milligrams a day. And if renal function is truly marginal, you can try every other day dosing. You want to stop the potassium supplementation and think about dietary intake. And as I mentioned, you want to monitor potassium levels on day two, three, and seven, and then monthly for the first three months. Sympathetic activation in heart failure is deleterious, it has, as you can see, negative renal effects, negative vascular effects, and negative cardiac effects, all of which contribute to disease progression. And so the use of beta blockers in chronic heart failure has been associated with effects in all-cause mortality across agents and across trials. What you see are three different trials with carvedilol, bisoprolol, and lock-acting metoprolol. The metoprolol succinate, not the tartrate that hasn't been proven to decrease mortality, but about a 30 to 35% reduction in mortality in chronic heart failure with beta blockers. They can worsen congestive heart failure acutely, and so you don't usually start them in acute CHF exacerbations. You wait until you're uvolemic or at least close to uvolemic, but you don't necessarily have to stop them in a patient who is on them and tolerating them. You try the diuresis through the acute exacerbation. If that doesn't work, you cut the dose in half, and if that doesn't work, you have to stop them. And if you need inotropic agents, then you really do have to stop them. It makes no sense to give an inotrope and a beta blocker at the same time. So what happens when you add all these things together? Well, this is the cumulative effects of congestive heart failure therapy, sort of in the chronological order of when they were introduced. So ACE inhibitors showed a 25% risk reduction compared to placebo. When you add a beta blocker to the ACE inhibitor, that goes to 40 some odd percent, and we add a melanocorticoid antagonist to an ACE inhibitor and a beta blocker, your reduction now is greater than 50%. So the usual therapies, beta blockers, renin-angiotensin system blockers, aldosterone antagonist, should be titrated as necessary. There are a couple of new agents, and we'll talk about two of them. Avabridine is relatively new, but actually now not so new that it might not show up on the test. It acts by inhibiting what's called the funny channel, the inward funny sodium channel that's present only in the cardiac SA node and in some parts of the eye. And so it acts only on the SA node, reduces the persistently elevated heart failure, and has a minimal effect on blood pressure. And this was tested in something called the SHIFT trial. The primary endpoint was a combined endpoint of death and hospitalization for heart failure at three years, and it reduced that endpoint. Mortality was not reduced. So the guidelines gave it a 2A recommendation, class B. Remember, 1 means do it, 3 means don't do it, 2A means you probably should do it, 2B means there is some data favoring it, and you could think about it. And so they followed the trial. Avabridine can be beneficial to reduce hospitalization for heart failure in symptomatic patients who are receiving guideline-directed medical therapy that includes a beta blocker and a maximally tolerated dose. And to get into the SHIFT trial, you have to have a heart rate greater than 70 at rest on beta blockers. So the guidelines follow those indications. Now there are the new neprilysin inhibitors. Neprilysin is something that degrades endogenous vasoactive peptides, and these are peptides that do good things. They decrease neurohormonal activation and decrease vascular tone and sodium retention. And so neprilysin metabolizes them. If you inhibit neprilysin, you have higher levels of these beneficial peptides. So this was combined in the compound LCZ-696, otherwise known as secupertylvalsartan. For the record, the pill is not a combination of the two medicines. The pill is one molecule that, when degraded, gets degraded into valsartan and secupertyl in slightly disproportional proportions. So it's not 50-50, it's really 49-51. But in any case, you get the effects of neprilysin causing vasodilation, and you get the effects of valsartan inhibiting vasoconstriction. And together, you get a combined effect in heart failure. This was tested in the Paradigm Heart Failure Study, 10,000 patients in 1,043 centers, the largest trial that's ever been done in heart failure. There was a run-in period showing toleration of either enalapril 10 times a day and then LCZ-696. So you proved that you tolerated the drug, and then 8,400 patients were randomized either to secupertyl valsartan or to enalapril, with a primary endpoint of cardiovascular death or CHF hospitalization. Here's the primary endpoint of death or heart failure hospitalization. With 8,000 patients, there was a reduction of 20% and a p-value with a bunch of zeros in front of it with a number needed to treat of 21. If you then look simply at death, you get a 20% reduction. And in fact, if you look at all of the subgroups for primary endpoint and cardiovascular death, I know you can't see the subgroups, but you can sort of see the point, a very consistent effect for both the primary endpoint and the endpoint of cardiovascular death across subgroups. So the 2016 guidelines talked about that. So the first recommendation is that a clinical strategy of inhibition of the renin-angiotensin system with either ACE inhibitors or ARBs or ARNI, otherwise known as angiotensin receptor neprolysin inhibitor, in conjunction with evidence-based beta blockers and aldosterone antagonists is recommended in everybody, class 1A, level of evidence A for ACE inhibitors, B for angiotensin receptor blockers, and BR, be randomized for ARNI. It's an enormous trial, but only one trial. If you broke it up in two trials, you'd have meta-analysis, but those aren't the rules. And then a class 1, again, B recommendation for ARNI, patients who tolerate an ACE inhibitor or an ARB replacement by an ARNI is recommended to further reduce morbidity and mortality. Now let's look at the cause of death by congestive heart failure severity. In class 4 heart failure patients, congestive heart failure or pump failure is the most common cause of death. But in class 2 or 3, sudden death is actually more common, and presumably some, if not much of that, is arrhythmic. So we come to the SCUD-HEF trial. That was a trial in patients with heart failure at EF less than 35%, and there were three arms, amiodarone, placebo, and implantable defibrillator. And what you can see is in this trial, amiodarone wasn't much different than placebo with respect to mortality, but an ICD reduced mortality compared to either amiodarone or placebo. So, it's also become recognized that some patients with left bundle branch block have ventricular dyssynchrony, which is shown in this graphic. So on the left you have asynchronous ventricle, on the right ventricle you have a depiction of asynchrony where first one wall contracts and then the other wall contracts. And this is bad. This is an inefficient way for the heart to pump, and it increases mitral regurgitation and deleterious remodeling. Cardiac resynchronization is atrial synchronous biventricular pacing. You put a left ventricular lead in the coronary sinus. You can see that graphically, and you can see this LV lead going out here to the LV, and you have a lead in the atrium and a lead in the right ventricular apex. And you can see that increases EF, decreases mitral regurgitation, decreases left ventricular volumes and thus remodeling, and decreases left atrial pressure. So cardiac resynchronization for eligible candidates, those with decreased ejection fraction, wide QRS, and left bundle branch block is a good therapy that reduces remodeling and improves mortality. And there's another drug that you need to think about, although it works in the kidneys and my renal colleagues and my endocrinology colleagues know all about it. This is inhibition of SGLT2, soluble glucose transporter. And what you see on the left is that this transporter transports glucose out of the nephron and back into the blood. So when you inhibit this, you get glucose urea, and you also get a diuresis because with the glucose goes more fluid. As you can see on the right, this glucose urea has effects on the kidneys and the liver and the heart, and all of them are beneficial. This is a meta-analysis of a bunch of trials. There are a number of large trials of all of the SGLT2 inhibitors. It really appears to be a class effect. And so this shows that the risk ratio for all of these trials, ranging from 2017 up until 2020 with subgroup analysis coming out every day, shows about a 25% reduction in death with and without diabetes mellitus. So it's not just the effects on diabetes, it also appears to be some of the neurohormonal effects of the SGLT2 inhibitors, perhaps on the liver and kidneys. Throughout this talk, I've talked about remodeling. So negative remodeling is dilation with decreased ejection fraction, and positive remodeling is when the left ventricles get smaller and the ejection fraction improves. And what you see here are the mortality ratio and the change in left ventricular volume from baseline. So up here are the ones that didn't remodel, the LV gets bigger, and mortality goes up. Here in the yellow are the neutral trials, LV doesn't really change, and the mortality risk doesn't really change. But all these green dots where there was a favorable effect on LV diastolic volume, and the ventricles shrink, there was a decrease in mortality risk. So the thinking has become that it's really remodeling that counts for a long-term outcome. If you have a beneficial effect on remodeling, it's good. If you have an adverse effect on remodeling, such as it happens with the catecholamines, then long-term mortality is worse. Here's another way of looking at that. These are therapies of demonstrated benefit across the cardiovascular continuum, from post-MI left ventricular dysfunction, to mild, to moderate, to severe heart failure. And what you see is the ACE and ARBs have trials across the spectrum showing benefit, the beta blockers have trials across the spectrum showing benefit, and the aldosterone antagonists now have trials. In addition, ICDs and remodeling and resynchronization therapy in the miracle and companion trials also benefit remodeling and also improve mortality. Now I'd like to move on to what we have been calling diastolic heart failure, but that's now an outdated term. So diastolic heart failure was best defined as a clinical syndrome with signs and symptoms of heart failure and a preserved ejection fraction. And that's distinct from diastolic dysfunction, a physiologic term denoting an upward shift of the diastolic pressure volume curve. Because this was confusing, the preferred term now is simply a descriptive heart failure, that clinical syndrome, with preserved ejection fraction. And because cardiologists like definitions because we want to make everything hard, now we call it HEF-PEF, as opposed to HEF-REF, reduced ejection fraction, and then HEF-MREF, heart failure with moderately reduced ejection fraction. So I'm going to take you back to your physiology and show you a pressure volume loop just to remind you of what happens as the heart contracts. So you start here, your mitral valve opens, you fill, your volume increases, and in diastole your pressure doesn't go very much, go up very much. Then your mitral valve closes, your aortic valve closes, you have isovolumic contraction. When the pressure reaches a certain level, the aortic valve's open, and you have ejection, and then the aortic valve closes. Now all the valves are closed and you have isovolumic relaxation. So for diastole, the key here is this compliance curve, the diastolic filling curve. So in diastolic dysfunction, what happens is that the ventricle gets stiffer, the pressure volume curve increases, and what you see is the same stroke volume but an elevated left ventricular pressure, and that left ventricular pressure translates back into the left atrium. With the potential for pulmonary congestion. Now diastology is fairly complicated, more complicated the deeper you look into it, and I don't really want to go into all of that, but I do want to talk about just a little bit. So you have three periods, you have a period of rapid filling when left atrial pressure exceeds left ventricular pressure, and then you have a period of diastasis when they're roughly the same and there isn't much filling, and then the left atrium contracts and causes a gradient and there is increased filling of the ventricle. However, early on, if you look here, what you have is falling pressure in the left ventricle and rising volume. So if you think like a physicist, if you have falling pressure and rising volume, the ventricle has to be sucking. So that sucking results from a combination of elastic recoil, the heart actually twists when it contracts and it untwists, and active relaxation. And then filling at the end of the cycle is passive and has to do with stiffness. As I mentioned, a story for another day. Now let's talk about management. You want to reduce pulmonary congestion by decreasing left ventricular volume, you want to maintain AV synchrony, you want to optimize blood pressure, and you want to treat ischemia when it's present. Heart rate is a little tricky. Tachycardia is poorly tolerated, particularly in patients with the potential for ischemia, and so in that context, beta blockers and calcium channel blockers can be useful to slow the heart down, but they do not alter diastolic compliance curves. They don't make the heart relax any better. So you have to be a little careful about lowering resting heart rate too much. Remember, these are small ventricles with left ventricular hypertrophy for the most part, and they can have a stroke volume that's as low as large ventricles with low EF. So the idea is to use negative chronotropes to prevent tachycardia, but not to reduce resting heart rate too much. How about reducing afterload or reducing remodeling with angiotensin receptor blockers? You would think this was a good idea, but here's the i-Preserve trial with irbisartan compared to placebo, no difference, combined endpoint of death or CV hospitalization or any of the other endpoints. Aldosterone antagonists also decrease remodeling. And so you have the TopCat trial. They tested spironolactone. This was funded by the NIH because spironolactone is cheap, so this is your tax dollars at work, and they looked at patients greater than 50 with an EF greater than 45% who had either been hospitalized within a year or had an elevated BNP. The primary endpoint was a combined endpoint of death, hospitalization, or cardiac arrest, and if you look carefully, that they did not meet statistical significance, 11% reduction with a P value of 0.14. However, there's a very interesting subgroup analysis of the TopCat results by region. So what they noticed was that the patients in Russia and the Republic of Georgia over near Russia looked very different. Their rates of events were one quarter those of the other population. Their rates of side effects were one quarter those, and spironolactone had no effect, almost as if they either weren't taking the medicine or something in their genetic makeup made the medicine ineffective. If you throw those out and you look at the patients in North America, U.S., Canada, Argentina, and Brazil, and you look at that, now you do get a significant reduction with spironolactone. This is an exploratory post-doc analysis with all of the caveats that pertain, but it is interesting. I would argue you want to take a strategic approach in congestive heart failure. In this context, strategic is pertaining to large-scale planning, long-term planning, and directing of operations, whereas tactical are planning for short-term outcomes. There's nothing wrong with being a good tactician, but you'd also like to be a good strategist. So how do you take a strategic approach? You define the CHF syndrome. You find reversible factors and treat them, but your ultimate target is ventricular remodeling since that correlates with long-term mortality. You want to use evidence-based mortalities, ACE inhibitors, or angiotensin receptor blockers in ACE intolerant patients, beta blockers, aldosterone antagonists, RNE 20% reduction compared to renin-angiotensin inhibitors, biventricular pacing in appropriate candidates, ivabredine in appropriate candidates who remain tachycardic despite beta blockers, and now SGL2 inhibitors may be in everybody. And more important than that, more important than anything perhaps, is use of a multidisciplinary care team with an emphasis on care management rather than episodic care, and the fact that education and follow-up are really the keys to long-term compliance. So one way to look at that, and I thought we'd talk about this, is because this is a document that the ACC put out looking at inpatient heart failure admissions and how to try to manage patients as inpatients with a particular view to long-term care management and the transition from the inpatient to long-term care management. So what you see here is the focus of care, and early on you have clinical decompensation, but as that gets better you start doing discharge coordination, but notice discharge coordination starts very, very early. Guideline-directed medical therapy starts early and ramps up. And at various points you have a trajectory check. You look at the long-term trajectory, but you also look at the short-term trajectory, and you see whether the patients are not improved or worsening in the hospital, improving towards target, or stalled. That is, they might have improved initially, but now they're stalled and they're not really getting very far. And those trajectories can be useful to guide patient management. So here's what you might do with a trajectory check. If you're improving towards target, now you continue toward decongestion, and if you're on track to reach decongestion, it's time to consider optimization of guideline-directed medical therapy and look toward discharge. If you're not improved and worsening, and in fact if you have initial improvement installed, then you have a different course. You want to escalate the current therapy. You want to consider additional diagnoses. Maybe your heart failure therapy isn't working because heart failure isn't the predominant problem. If you're really going in the wrong direction, you want to consider an invasive hemodynamic assessment. You want to think about advanced therapies, and regardless of what you do, you want to assess prognosis with the patient and reassess goals of care. Maybe this is as good as they're going to get, and maybe they have to accept that and move on to a different sort of therapy. And finally, the last word. Heart failure management isn't an individual sport. It's a team sport, and we're all in this together. Thank you very much.

heart failure

pathophysiology

perfusion failure

classification system

ejection fraction

management

medications