Refractory Congestion in Heart Failure: A Practical Approach to Diuretic Resistance with Dr. Wilson Tang

Podcast Transcript

Dr. Brateanu: 

Welcome to the Medicine Grand Rounders Podcast, a platform dedicated to exploring key topics in internal medicine. Highly relevant to the medical community. This podcast is made possible through the generous support of a grant from the Cleveland Clinic Education Institute. However, the views and opinions expressed here are those of the speakers and do not necessarily reflect the the official position of the Cleveland Clinic. Each episode brings together world class experts and distinguished physicians from Cleveland Clinic to share their knowledge, experience and perspectives on issues that impact healthcare professionals and patient care. Our discussions aim to promote learning, advance professional development and inspire meaningful conversations with the medical community. Today's episode is hosted by Dr. Nitu Kataria, Internal Medicine Physician at the Cleveland Clinic, and me, Dr. Andrei Brateanu also in Internal medicine. We invite you to join us as we delve into today's thought-provoking topic.

Faysal:

We have with us today, Dr. Wilson Tang. Dr. Tang is an advanced heart failure specialist at the Cleveland Clinic. He's also a professor at the Cleveland Clinic learner College of Medicine and the director of research for heart failure and advanced therapies at the Heart and Vascular Institute. Good morning, Dr. Tang, thanks for being here. 

Dr. Tang: 

Hello. Happy to be here.

Faysal: So, Dr. Tang, we're going to start with a clinical scenario, which we commonly see in practice as internists. A 68-year-old female comes in with heart failure, with reduced ejection fraction, chronic kidney disease and repeated heart failure admissions. She presents with progressive dyspnea, abdominal distension, orthopnea and weight gain. She has received IV loop diuretics but achieved only marginal net negative output over the past 24 hours. Her creatinine rises slightly. Blood pressure is soft and her congestion clearly persists. 

I'd like to use this case as a framework to go deeper into how you conceptualize refractory congestion, therapeutic sequencing and what future therapies might change how we treat these patients. 

So first, how do you personally define a patient as being refractory to diuretic therapy? Is it purely output based or does the definition evolve depending on clinical context? And how do you structure the decision of when to intensify loop diuretics versus adding other agents?

Dr. Tang: 

Yeah, great question. And in fact, I think this is almost on a daily basis we see most patients coming in with heart failure. They are usually congested. I mean, as we know, heart failure used to be called congestive heart failure, and that's because before we have even any measurements people come in, you know, really congested. Now, interestingly, your introduction just said they already tried for 24 hours and had marginal output. I think the question that has always come up with this is, we always think we are giving people a certain amount of diuretics, and then wait until the next day, when we see how much response is. If you actually understand how the drugs work and how what is the half-life of an intravenous diuretic, you would realize that it only lasts for at most 2+ hours. So, in many ways, we really need to have a goal every time when we admit a patient. So, to identify somebody that need diuretics is the first goal, obviously. The most common way of not responding is that you just don't even recognize that they need it that much. So, we usually tell people, at least for a rule of thumb, at least 2 to 2.5 times their home diuretic dose is the minimum starting dose. And so many of the professional society guidance’s have actually started to promote the idea that we really should follow very closely between 2-6 hours after we give the diuretic. So, this is not something that you just write an order and then just, you know, go and take a nap and then wait for next morning, okay? We really have to kind of check what is the response? And there are two ways to do that. You can give adequate dosage, or you can actually put it in a continuous infusion. That means that it's just not drop down for the 2 hours, but continuously get a continuous dosing, or even to repeat dosing. And the repeat dosing is where most people don't do, they don't really recognize that. You know, there's only a limited amount of urine output. Now, the heart failure network has done at least 4 or 5 different clinical trials in exactly this arena, because previously, we really never have a lot of good clinical trials on diuretics and decongestion, which is ironic because it's one of the most common reasons why people come to the hospital. 

So, throughout that learnings, we learn a few things. That diuretics are actually quite safe. You can actually give higher doses without much, you know, penalty here. The second thing is, we really need a goal, and that actually came from the CARRESS-HF study. When we try and ask, how much are we supposed to actually give? Meaning: what is a good response? It turns out that was actually because we were designing a study that matches ultra filtration, which has a set negative, you know, your output, and so we have to match that, and that was up to even 200ccs an hour for 24 hours. So actually, the goal is 3-5 liters negative of urine output, which kind of makes sense, because if you have a 1-2 liter intake, or 1-1.5 liter intake, then you have 1-2 liters negative, or 2-3 liters negative, which is kind of what we are happy when we round the next morning, isn't it? I mean, you got a few pounds off and all that. So, in some ways, you can actually kind of divvy out in your kind of estimated goal, and that's why, after your first dose, at least, you know half a liter or more. And if you have difficulty getting urine and all that, we can actually use something like urine sodium, because after all the loop diuretics is trying to inhibit the sodium co-transporters, and what you're doing is you're actually trying to measure the efficacy, the pharmacodynamics of the drug that is delivered to the luminal side or the urinary side of the auto tubules. And so, in some ways, we are actually able to estimate that if you don't have adequate natriuresis, adequate spot urine sodium, about to 2-6 hours out, about maybe more than it had to be about more than 50meq/L. So that's actually the kind of minimum threshold. So, there are these two ways that the ESC-HFA, you know, consensus statement, and subsequently the ACC expert opinions and consensus has recommended. 

Faysal: 

Moving to our second question would be, do you approach patients with refractory diuresis differently in heart failure with reduced ejection fraction versus those with preserved ejection fraction, and where do the underlying hemodynamics diverge in your mind?

Dr. Tang: 

I think the overall mechanism of congestion and decongestion are actually very similar, because we are really talking about circulatory volume and then the response by the kidneys to relieve salt and volume. The underlying difference between impaired versus preserved ejection fraction comes down to how the salt and water is accumulated and how easy it is to be mobilized into the bloodstream, into the kidneys. 

In general, patients with impaired ejection fraction has a little bit more hemodynamic dependency. They actually little bit more even responsive to vasodilators, for example, but overall, they also have a tendency to accumulate more fluid over time. On the other hand, in HFpEF, the effective circling volume, which is what the nephrologists always used to call it, is a little lower, meaning that it's actually quicker for them to, you know, get intravascular volume depletion, and that's why you actually see a little bit more granny rise and more difficult in mobilizing fluid. But in general, it's really individual. I think, like the patient that you mentioned, there is already a few red flags that you can see that is identifying somebody having a not very good response, as you find out later. The person already has some kidney dysfunction, had multiple hospitalization, blood pressure is not very high. All these are factors that we now actually put together in actually a risk score called BAN-ADHF risk score. You can find it online. It's actually a machine learning derived risk score from the heart failure network program and validated it in various large databases, to actually identify and predict diuretic resistance. The definition of it is a little bit difficult. Usually, we talked about a unit of dosing of diuretics in terms of how much weight loss or how much urine output. But at least by theory, some of these are well known risk factors. You have elevated BNP, home dosing that's elevated. So, all these features are actually in your patient as well. So regardless of preserved or impaired ejection fraction, it’s really how the kidneys are unable to effectively relieve this congestion. That is probably a telltale sign of what's going on.

Faysal: 

Do you think that there are certain patients’ characteristics such as right sided or those with right sided Heart Failure? We talked about renal function, but venous compliance, that immediately change how you classify and manage diuretic resistance.

Dr. Tang:

Yeah, that's a whole story to tell. For a long time, you know, we have always thought that in patients with heart failure, the heart is unable to pump effectively, to perfuse the kidney. So, if you actually pull up the New England Journal review on heart kidney problems, say, 20-30 years ago. You know, we caught the Schrier model, Robert Schrier and Bill Abraham actually wrote all this about arterial under filling that's causing impaired cardio-renal compromise, sodium retention. This is not a new idea. We actually knew that for a long, long time, that's the renal model for heart failure. You know, that actually even precedes the hemodynamic model for heart failure. The congestion is because the kidney is sodium avid. It's actually trying to hold on to water and salt. I fondly tell the patients, it's like the kidney thinks we're in the desert, but we're not. So ended up our basement flooded and we can't get rid of it. So fast forward, when there's a lot of idea about backward failure, that's when the right heart failure and all that came along. We've known for a long time people with right heart failure have a lot of congestion. It is the vein part. They may not have a lot of pulmonary edema, but they certainly have a lot of fluid. You have a lot of belly abdominal pain, abdominal congestion and extension curve, and then you have pitting edema. Fast Forward, when we actually directly measure the pressures in patients with cardio-renal compromise, people with worsening kidney function, despite trying to use diuretics and still have persistent edema, we noticed that it's really not the malperfusion, it's not the forward flow problem. The cardiac index, by measuring directly with a Swan-Ganz catheter, is very intact, and that has actually been proven in even multi standard studies like ESCAPE or single center studies here about 15 years ago by Mullens et al and our group, in our advanced heart failure ICU, that show that the venous congestion actually drives a lot. In fact, this is almost like having a traffic jam. You're actually having a backward failure that goes all the way down into the splanchnic circulation to impede the flow getting out of the organs like the liver and the kidney and so for that, you kind of have a difficulty in those organs to actually maintain perfusion. So, it is almost like the preload and afterload of the kidney here. And so, in many ways, the big kind of advance in the understanding is that the ability for us to relieve the congestion actually sometimes help the kidneys to actually get the flow going. And of course, now we have a lot more noninvasive measures, like vexes and portable ultrasounds to actually show that when they are discontinuous flow in the renal Doppler, we actually see that that's almost like the stop and go traffic jam that we are dealing with in rush hours. And in some ways, it kind of clarified, again, the more importantly, that adequate dosing of diabetics is actually critical to try and get that jump start going.

Faysal:

This brings us to the point, whether there are specific physiologic or biomarker signatures that push you toward the treatment strategy or one therapeutic combination over another?

Dr. Tang: 

There isn't a specific biomarker per se, but a combination of biomarkers that we commonly use certainly give us a lot of clues that the kidney are certainly working hard. And we all know those. We have known that for a long time. That's actually part of internship training almost you know, like, if you know your sodium is low, your creatinine and BUN start to go up, your BNP or NT pro BNP is high. Those are all features that show that your neurohormonal activation is sky high. And the kidney is somewhat dysfunctional, not just dysfunction in terms of measuring of glomerular filtration rate, but ability to maintain electrolyte homeostasis. Okay, so that's something that is often overlooked. You know, most of us are able to maintain that with good balance, despite changes, but if you actually start to have a lot of fluctuation, and more recently, for the last 10 years, our group here also described hypochloremia. Your sodium and your chloride are both drifting down. That means that you are actually not only having a neurohormonal regulation, but also depletion. And so, both those together makes it a lot more refractory. 

And so there has been a lot of really creative ways to even think about not just using loop diuretics and using different types of drugs upstream or downstream from the nephron to try and, you know, overcome the compensatory efforts that even, you know, kind of out of the box idea of intravascularly repleting salt. Because if the sodium avidity is caused by sensors that telling the kidney to not let go the salt. The response has been mixed. We certainly see some improvement in our heart failure ICU. We do have a hypertonic saline protocol that actually works sometimes to stabilize patients, but it doesn't really cure the patient. It really, you know, correct some of the abnormalities, and it may actually allow the direct response to be at least maintained, if not restored, but usually is really dosing, duration and delivery of the drug to the right places to try and get that going.

Faysal:

A lot of the clinicians who take care of heart failure patients fear worsening renal function. How do you balance congestion relief versus protecting kidney function? Which markers do you trust the most?

Dr. Tang:

Almost two decades ago, there's this whole notion that worsening renal function is bad and is always associated with worse outcomes. In fact, if you think about the original ADHERE registry, Dr Gregg Fonarow, actually, you know, outlined two of the three parameters that look at in house mortality is BUN and creatinine being elevated, as well as, systolic blood pressure being low. All these we know are poor prognostic factors. When we actually move forward to try to use different medicines, what we found out is sometimes the drugs work very well, and sometimes the drugs don't work very well, and sometimes the drugs work very well with increasing creatinine, but yet they are still diuresing. So, there's the seminal work, at least by Dr. Jeff Testani, that actually identified that if you actually produce very good diuresis in response to diuretics, despite the rising creatinine, you actually fare much better than if you actually don't respond to diuretics, meaning that your urine output is not adequate. So that rise in creatinine is almost like an indicator of hemoconcentration. It actually increases a biomarker, but it's not really reflective of true, true injury. And in fact, I think about maybe 15 years ago, we actually directly measured, actually, it's done both here and at Fairview hospital, when we actually look at patients coming into general medicine wards, many of them are elderly, many of them congested. Both have HFpEF and have HFrEF, we directly measured a lot of really high-end acute kidney injury markers, KIM-1 and NGAL. These are the ones that our nephrology colleagues truly associate them with direct injury of the kidney. And actually, none of these people who have rise in creatinine show any evidence of injury by these markers, and this has been validated in the samples collected by the heart failure network and the ROSE study and other studies. So ultimately, we are actually witnessing many times in patients who get aggressive diuresis, the rise in creatinine is really a marker of adequate decongestion, rather than true, true injury. Now, I think that's a different discussion when you drop off in terms of your urine output, and that happens in some cases, particularly when you have somebody that came out from the emergency department after rule-out PE CT scan and have acute, you know, tubular necrosis, from contrast, for example. When you can have concomitant injury too. So that's really kind of where to sort it out. And that's really general, you know, internal medicine diagnostic to actually look at how they respond to diuretics and how congested somebody is. In general, it is still true after all these years. If you have fluctuations of your kidney measures, in this case, BUN and creatinine, you do have a worse outcome. We actually, I was actually the first author of AHA scientific statement, almost a year ago now, that is describing kidney dysfunction and advanced heart failure and we were highlighting the fact that there are different profiles. If you have rock steady BUN and creatinine that really doesn't move despite you giving, loop diuretics, you're in a really good situation. Your heart and your kidneys are actually able to relieve the congestion without problems. If you start having fluctuations of either creatinine, electrolytes or whatnot, basically, it tells you a little bit about the kidneys not really handling it very well. And then, of course, the most common ones is the one that, when you, when you actually give people diuretics, and then their creatinine goes up, and then they stay some. Those are the ones that you know, that you know, there's some really significant compromise. And sometimes we do have to back it off, you know, we call it intravascular depletion. Sometimes we call it metabolic alkalosis, contraction alkalosis and all those. They are different ways to basically describe the same thing that, you know, we have to give the vasculature and the perfusion some time so that we could balance the ins and outs to make sure that the kidney is still happy.

Faysal: 

Basically, Dr Tang the ADVOR trial brought acetazolamide back into focus. I know that you personally were involved in a lot of work looking into the role of acetazolamide in heart failure. Do you think acetazolamide has been underused in practice? If so, why and how can we make better use of it?

Dr. Tang: 

Yeah, this is a journey in his making the primary PI you know, his name is Wilfried Mullens. He's the upcoming president of HFA, European Society of heart failure president and used to be a fellow here, and very studious and very productive. I always say that, you know, if your mentees are more bright than you are, then you've done a good job. I was happy to see all the events, and he actually had this idea that a lot of people frowned upon, that upfront use of Acetazolamide in blocking the sodium reabsorption is actually very important, and the rationale is actually quite clear. In fact, acetazolamide has been used in the 50s, and it didn't work very well because the compensation mechanism was very good distally, particularly at the loop of Henle and so, and that's even before we have loop diuretics. And so, it's kind of came and gone and didn't really gone too much. We use it for like mountain sickness and all that. What we have actually noticed that it does protect the sodium depletion that the distal nephron senses. And so what we actually recognize that is, by enhancing that delivery to the loop of Henle, we can actually reduce the neurohormonal activation, particularly the renin release. That is one of the reasons why we had this, you know, blood breaking phenomenon at this kind of neuro hormonal response when we actually give loop diuretics. And so, in some ways, it's a little bit of a preventive strategy for aggressive loop diuretics that we attenuate the neurohormonal response. And so logically, what it would do is, for the same amount of diuretic, loop diuretics we give, we will have more effective naturesis and decongestion. And so, they set out and did a double-blinded, randomized control trial. They were not intending to do double blinded. I actually convinced them to do double blinded because I told them that that's the only way you can actually show that it's not the clinician favoring one drug or the other, you know. So, if you assume that both groups are having it, would this still beat placebo. Lo and behold, this is the one of the biggest randomized control trials that was done, and they did see that there are more people achieving successful decongestion at 72 hours and at the end of study, without any significant cost of rising creatinine, worsening organ dysfunction or adverse clinical outcomes. And so that's reduced the length of stay by almost a day or so, and that's actually quite impressive, considering that this is actually both the clinicians didn't really know which they are. 

And so that tells us two very, very obvious things. Number one, in the best-case scenario, when somebody was admitted and was enrolled in a study, we really inadequately treat patients. There's still a lot of residual congestion, and that add on upfront Acetazolamide, which is basically 500 milligrams for three days upfront, at the same time, within 24 hours of admission actually can produce incremental diuresis, naturesis and achieve successful decongestion. And so, it actually shows at least a proof of concept that this segmental blockade is actually an effective strategy. And of course, it's very economically favorable. It's very cheap. It's an old drug. We have it since the 50s. In many ways, we could see the effect directly by the so-called contraction alkalosis, the chloride and then the bicarbonate changes, and then much less drop in chloride when we actually in the acetazolamide group. It is underused, because unfortunately, I think even now, what two or three years out after this is published, it's about 5% of people still getting/receiving this. So, there's a lot of at least uncertainty still out there. I think we are so eager to look at mortality benefits and all that, and most people say, oh, there's no mortality benefits. Firstly, it's not powered by this. Number two is that majority of the people are late 70s and early 80s is actually quite old elderly patient population. In fact, most of these drugs here now really for stabilization purposes. I would add to the point that you need adequate decongestion to start guideline directed medical therapy, which is the mortality benefit part. So residual congestion is an issue, still is an issue. Even in clinical trials, the latest clinical trials, by report, about half of the patients remain inadequately decongested; still have edema, still have JVD, still have abdominal swelling. And so, this is something that when we treat patients as they leave the hospital, we really need to try and make sure, even if they don't have, fully decongestion, to have very, very close follow up to make sure they have adequate decongestion.

Faysal: 

Dr. Tang, in addition to closely following up with patients, cardiac rehab remained dramatically underutilized. What do we underestimate about its benefits, especially in patients without ischemic disease?

Dr. Tang: 

Yeah, I think cardiac rehab is a challenge. And I always tell people if we have a pill that would help, that the same things, we will be in a lot better shape. And that's the problem. We don't know exactly what are the many benefits that exercise entails. And so, that's the only way that they could help. Now, there are some challenges. 

The first challenge is the logistics and commitment to undergo cardiac rehab. Most of them ended up doing two or three, three or four times a week. You really have to be in a center. And unfortunately, right now, it's still restricted to a set number of indications, particularly with LV ejection fraction 35% or less. So, the main challenge that we have is a lot of people are not even eligible for this. So, there's a lot of people coming in without these indications and doesn't have access to it. There has been some advance over this area too. There has been some suggestion that kind of a low level, you know, physical therapy, having a set protocol can help. There's actually a rehab Heart Failure study that was done at the time of discharge, when they actually get discharged, to undergo some exercise regimen. That seemed to work, and that may end up being the more scalable approach. I think nowadays, with particularly after covid, when we have all these programs that has kind of options for virtual rehab or wearables that are available, I think this is a real growth area, because, clearly, mobilization is one of the key factors after hospitalization, not just heart failure, but any other disease. I mean, after surgery, you know, what is the effective you know, recovery, you know, protocol is really, you know, prehab and rehab. So, any form of disease, then somebody actually get debilitated and need to be stabilized, when they actually start movements are increasing physical activity has all the benefits.

Faysal: 

All right. And as an expert in the field, when do you begin thinking that this patient may be heading towards advanced heart failure strategies such as LVAD or transplant?

Dr. Tang: 

This is an occupational hazard. I guess I always think about somebody may need somebody unless they don't. So clearly, any form of organ dysfunction, we always need to ask the question, is it directly from the heart getting weaker? Now, certainly there are biomarkers, of course, end organ biomarkers, but also natriuretic peptides. We are actually under testing them all the time. In the ACC/AHA 2022 guidelines, there is an indication for pre-discharge or even early follow up natriuretic peptide testing to look at the degree of severity. And I think we often overlooked that many people who self-reported they're feeling fine and doing okay, they're actually sicker than we think. I mentioned about not only the creatinine, but also even electrolyte abnormalities and different, you know, facets that show that somebody is not very stable. It's actually an important indicator. But overall, I think beyond symptoms, really, end organ, you know, perfusion, and sometimes it's very subtle. I mean, they don't have textbook presentation. In fact, a lot of people even minimize their disease or minimize their symptoms. Not intentionally, sometimes they live with it for so long they don't even know and sometimes it actually is intentional. They don't want to look at what the scale tells them, because they know what the consequences are. So, a lot of it is really trusting the clinician and having close follow up and objective measurements. Notice, we haven't talked anything about echoes and all that. You know, it's really the biomarkers that are giving us a lot of clues. Somebody can come in and out of hospital and the ejection fraction looks exactly the same. They may have difference in diastology, they may have difference in strain, but those are also highly variable. So clearly, when we talk about tracking disease severity, end organ biomarkers, cardiac biomarkers, and particularly natriuretic peptides, give us a lot.

Faysal: 

Dr. Tang, this brings us to the end of our talk for today. Please allow me to thank you on my behalf and our production team and our internal medicine residency program for being with us here today, thanks a lot. 

Dr. Tang: 

Thank you, thank you for having me!

Dr. Kataria: 

To our listeners, thank you for joining us on this deep dive into this important topic. We hope you found this episode both educational and engaging. On behalf of the team, thank you to our special guests who joined us today. Thank you also to the Cleveland Clinic Education Institute for the educational support of this project. Until next time, please enjoy this and future podcasts from the Cleveland Clinic Medicine Grand Rounders.