This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.
Carol Wysham, MD: Hello. I'm Dr Carol Wysham. Welcome to the second season of Medscape's InDiscussion series on type 2 diabetes. Today we'll be discussing heart failure and diabetes. First, let me introduce our guest. Dr Butler is a senior vice president at Baylor Scott and White Health in Dallas, Texas, and distinguished professor of medicine at the University of Mississippi in Jackson, Mississippi. Welcome, Javed.
Javed Butler, MD: Great to be here. Thank you so much for inviting me.
Wysham: Oh, it's my pleasure. One of the questions I've always had for you, Javed, is what led you to seek a fellowship in heart failure when you were looking at your options?
Butler: It's actually a really interesting question. Many folks have much more diffuse answers. I know the exact moment I became a heart failure doctor. I was born to be a hepatologist; I wanted to do GI. All my electives were in GI during internship, applying for my fellowship, and then as a second-year resident, I was sitting in and somebody came from New York to New Haven to give grand rounds, and that was Milton Packer. And remember at that time — we're talking about the 1990s — heart failure was not what it is today. And he talked about neurohormonal theory for heart failure, and I can tell you that one grand round changed my life. I was just like … wow! I changed my electives to do some cardiology electives. Now, many years down the road, here I am, not only as a heart failure doctor, but I work at the same institution with Dr Packer, day in and day out. That's a pretty interesting story, at least from my perspective.
Wysham: Have you actually had this conversation with him? I assume that he knows the large part that he played in your life.
Butler: Absolutely. And he was so gracious when I told him the story.
Wysham: Oh, that's wonderful. Well, you know, I've been in this field for a very long time and I, like you, have been very tuned in to cardiovascular complications of diabetes. It wasn't until the EMPA-REG study, which was published in 2015, that heart failure started to be on my radar screen. Like many, I consider heart failure to be one of the stealth complications of diabetes because, unlike the other things I frequently see, I don't have a clue when and how to be concerned about heart failure. And therefore, I think the first question that I would like you to answer is what are the more subtle signs and symptoms of heart failure that we as diabetologists or primary care providers can look for to increase our suspicion of heart failure?
Butler: Yes. So, this is a critical question, and it's both an easy and a difficult question. It's an easy question because the signs and symptoms for heart failure are very generic. We are used to dealing with these signs and symptoms all the time: shortness of breath, fatigue, tiredness, dyspnea, inability to be able to do things that that you were [previously] able to do. And then if heart failure is not diagnosed for some time, we see the inability to lie flat, waking up in the middle of the night, shortness of breath, and going to the bathroom more frequently at nighttime. These are the common symptoms. The problem here is that, for the longest time, we have been focusing on prevention of cardiovascular disease: looking at the Framingham Heart Study, the lipid revolution, going on from Framingham Risk scores in the 1970s and 1980s, when the whole idea was that cardiovascular disease was equal to atherosclerotic cardiovascular disease. It was all about coronary artery disease or vascular disease, and heart failure was a consequence of vascular disease. Then, by definition, if you would just focus on diagnosing, treating, and preventing vascular disease, you would prevent heart failure. The problem is that while that statement is true, that statement is true only for a proportion of heart failure. What we learned over time is that a large proportion of heart failure, almost half of all heart failure, occurs in the absence of manifest vascular disease. I don't know whether people have subclinical coronary disease or not, but at least they have no history of angina, no history of MI [myocardial infarction], no history of stroke, no history of peripheral vascular disease. In this population, there is no manifest vascular disease and people develop heart failure, especially heart failure with preserved ejection fraction (HFpEF). We need to change our mindset to broaden the concept of prevention and thinking about heart failure. The second issue is that in heart failure with reduced ejection fraction (EF), your heart doesn't contract enough. But when we talk about heart failure with preserved ejection fraction, you're literally trying to diagnose a disease that classically has required a normality in finding. You do an echo and the EF is normal. So that becomes pretty complicated in the primary care setting. What people end up doing is ruling out all the common things in a person complaining of shortness of breath and tiredness: COPD [chronic obstructive pulmonary disease], asthma, lung disease, hyperthyroidism, anemia … and then they stop. What ends up happening is you don't bring up heart failure as one of the things that you ought to be screening for, and obesity or aging is blamed for the patient's symptoms. What's not realized is that 3 or 4 months ago, the patient was the same weight and the same age, and these are new-onset symptoms that need to be looked into. Therefore, the number one place where heart failure gets diagnosed is in the emergency room. Why? Eighty percent of heart failure is diagnosed for the first time in the emergency room because the outpatient diagnosis has been delayed and ignored. And now a person shows up in extremis, with pulmonary edema, and that's where the diagnosis is made. This is a long-winded answer to say the signs and symptoms are really common — shortness of breath, fatigue, edema, inability to do what you once were able to do, your satiety, orthopnea, for example. But the issue is not that the symptoms are rare; the issue is that we don't think about the differential diagnosis of heart failure.
Wysham: Do you have a question that you would suggest we ask our patients on a routine basis that might start teasing out some of these issues?
Butler: We routinely do vital signs, right? Somebody does not have to have a complaint of headache or dizziness to do vital signs. You will always do heart rate and blood pressure on everybody coming in. Some sort of a questionnaire as an intake form about tiredness, fatigue, and shortness of breath is also very helpful because that will flag it. Otherwise, unless and until the patient actively complains of these things, we will not necessarily ask these probing questions. And what ends up happening is clinicians do not necessarily take these symptoms seriously when they are mild to moderate, but even the patients don't take it seriously. What patients end up doing is lowering their expectations of themselves; for example, "I used to be able to walk six blocks and now I am able to walk three blocks. I used to take care of everything at home myself, but now my son or my daughter helps me." Those dynamics come in to play. When you don't ask these probing questions, they may come to the clinic and they are "doing fine" and everything goes under the radar screen until it becomes a really big problem. For early screening, having some sort of a questionnaire for these generic diseases would be very helpful.
Wysham: Yes. And I find just asking patients, "Is there something that you were doing last year that you no longer feel like you can do?" A lot of times they say, "Oh, it's just because I'm getting older." They don't really think of it as being a significant change. Hopefully, the audience will take that and put that into their toolbox to help us identify our patients sooner. Speaking of that, as you probably know, the American Diabetes Association had a consensus paper that was published recently where they recommended that we routinely use biomarkers for screening for heart failure in our patients with type 2 diabetes. What's your opinion on that?
Butler: I think it has been long due. This is a great step that the American Diabetes Association has taken, and I would hope that this will multiply and other societies will do the same. The reason I say this is that for most diseases, quantification really matters. I'm a little bit shy about talking about hemoglobin A1c with you, but hemoglobin A1c is not perfect, right? For instance, if you had a threshold of 7, a threshold of 6.5, or many people with 6.4 and you were to do oral glucose tolerance test, maybe it's worse than what you are looking at [based on] hemoglobin A1c. Nevertheless, having some degree of quantification to make clinical decisions really helps; then, for the gray zone around that number, you can manage clinically. We have a number to look at in blood pressure, we have a number in CKD [chronic kidney disease], we have a number in cholesterol, but when it comes to heart failure, it's relegated to signs and symptoms. That leads to all those issues that we just briefly touched upon in the earlier question. I think that first, the data is incredibly robust on natriuretic peptide levels being high as a marker of cardiac stress. Now, cardiac stress does not equal heart failure, but it equals something. Not ignoring high anti-pro-BNP [brain natriuretic peptide] is a great idea, so that it can take you to the next level. Now it may be aortic stenosis, it may be just uncontrolled hypertension, it may be just bad CKD, it may not be heart failure, but at least it rings the alarm bells that there is something going on that you need to be looking into. I think this idea that an annual assessment of natriuretic peptide in type 2 diabetes patients who are, by definition, at high risk for heart failure is a great step. We don't want to wait too long, and they develop florid heart failure. We want to catch them earlier, when they are at risk for developing heart failure, to intervene.
Wysham: I think you answered my next question. The NT-proBNP is the one that you would recommend that we measure. Do you have different cutoffs for going to the next testing level according to age or CKD status? Or would you say any time the level is above 50 or 100, the patient should proceed with the next step?
Butler: Yes, so at least my bias would be that if your BNP is greater than 50 or your NT-proBNP is greater than 125, then proceeding with the next step is a good idea. Now, what is the next step and who should be doing that? One can argue from a cardiologist perspective, the obvious next step is an echocardiogram. Echocardiogram, remember, is one thing we do in medicine that has absolutely no risk. I mean, there's no controversy, no risk, no radiation, no nothing. The issue that then comes up is cost. You don't want to do unnecessary echocardiograms because you will multiply the health care costs. That's why having some thresholds is a good idea if a diagnosis of heart failure is made. The next question is if somebody has shortness of breath, tiredness, some lower extremity edema, something is different, and you do natriuretic peptide and are thinking about diagnosis of heart failure. In primary care, you can do other things like rule out hyperthyroidism, rule out anemia, and then send it to cardiology and echocardiogram for a one-time assessment to rule out pericardial disease, valvular disease, and to rule out heart failure. If there is no heart failure, then the person goes back into the primary care setting. If a person does get a diagnosis of heart failure, then that requires a whole new evaluation for the cause of heart failure, and that can occur in the cardiology practice. If you're doing what you just said, which is screening NT-proBNP in the absence of symptoms, now you're not talking about diagnosis of heart failure in a symptomatic patient, but potential strategies for prevention of heart failure. In that case, it's a little bit more questionable whether everybody should get an echocardiogram or not. So a person is completely asymptomatic going on with their life, but their natriuretic peptide levels are a little bit high … in that case, I think I would individualize it. If somebody is older, bad family history, multiple risk factors, smoker or whatever, you might want to do an echo. I would at least do a strong risk factor modification and a strong consideration of SGLT2 [sodium-glucose cotransporter-2] inhibitor in a type 2 diabetes patient with high natriuretic peptide levels who is really set up for developing heart failure. With so much data on prevention of heart failure, with SGLT2 inhibitors, that would be a good idea. But then one can argue whether these are hard numbers, or whether we should alter these numbers depending on the patient phenotype.
Wysham: I think that's really helpful. What I'm hearing is, if in doubt, it would probably be appropriate to get a cardiology consult if you're on the fence about whether or not your patient has a diagnosis. Well, that is incredibly helpful for me in my practice. Let's talk about the medications and the continuum of care for patients with type 2 diabetes and heart failure. Obviously, there exists now guideline-directed medical care for patients, particularly those with HFrEF. Obviously, there are four to five classes of medications that are recommended, which of course include the SGLT2 inhibitors with or without regard to type 2 diabetes diagnosis. Starting with HFrEF, which I think, as you indicated, accounts for probably about half or maybe slightly less than half of the total number of patients. Can you tell me how you go about staging the medications, starting them? Do you start them in the hospital? Do you have a specific guideline for how they are started and in which order and you know all at once? Or what's the timeline between the patient diagnosis of heart failure and the time you have them on the appropriate therapies?
Butler: Great question. And there are several principles here. First is that if you look at the last set of guidelines, both American and European guidelines, they were very algorithmic, you know … do this for to do this. Second, wait this long … watch for this. And this iteration of guidelines is distinctly different. All of those things have just simply gone away, and there are several reasons for that. One is that, remember, sequencing is a historical construct, not a biological construct. It's history of medicine that ACE inhibitors were tested in the 1990s and beta-blockers and angiotensin receptor/neprilysin inhibitors (ARNIs) and SGLT2 inhibitor. But there's no scientific rationale that you have to prep the heart with one medication first before the second medication will work. In fact, there is plenty of data that if a medication works, it works irrespective of baseline therapy. For example, when the RALES trial was done with MRAs (mineralocorticoid receptor antagonists), beta-blockers were not standard of care; only 10% of the patients were on beta-blockers. When the EMPHASIS-HF trial was done, 90% of the patients were on beta-blockers, but beta-blocker presence did not matter — MRAs work regardless. We have the same data with ARNIs and MRAs, with SGLT2 inhibitors and ARNIs. The sequencing concept is completely disfavored right now. Also, how can we have the same algorithm when all our patients are different, right? We see congestion, no congestion, atrial fibrillation, high heart rate, creatinine, potassium — you have to individualize it to a given patient. Third is that the benefit with all these therapies is seen within weeks of starting the therapy. And therefore, following a strict regimen and waiting for months on end puts patients at high risk and inertia sets in, and then you don't necessarily give the next therapy. And then the fourth, and last, is that doses are important but of secondary importance overall. The primary importance is covering all the necessary pathways with some dose because with introduction of medications at low doses you get the biggest bang for the buck. This doesn't mean don't go up on the doses, definitely go up on the doses, but it's a bad idea to go up on the dose so much of one medication that you have no blood pressure or creatinine left for the other medication. Or you go up on one medication and it takes such a long time that the other medications are never introduced. Now, with SGLT2 inhibitors, dose is not an issue, right? It's just one pill, 10 mg once a day. With MRAs, dose is really not an issue: 25 mg spironolactone is good enough. With beta-blockers and with ARNI or ACE inhibitors, dose is an issue. Also, barring accessibility issues, the data are incredibly clear that ARNIs are the preferred drug, not ACE inhibitors. So it's really ARNI, beta-blocker, MRAs, and SGLT2 inhibitor. I have absolutely no preference on which one you start first, but there is some edge that some drugs can help facilitate other therapies. So, for instance, if you are congested, you're not going to tolerate beta-blockers. But if you get an SGLT2 inhibitor or ARNI that has diuretic properties, your beta-blockers will be better tolerated. If your GFR [glomerular filtration rate] is borderline or potassium is borderline, you're not going to tolerate MRA. But reasonable data suggest that SGLT2 inhibitors will help facilitate MRA because they lower the risk of hyperkalemia. Right now, the focus is whether you do it in an in-person visit, telephone visit, or virtual visit, and then on a weekly basis or something else. Just get all four medications on board in any sequence and at least low doses in 4 weeks. And then you can take your time going up on the doses as needed and as tolerated, but cover all bases in about 4 weeks or so. I am not a fan of starting more than one medication at the same time because if the patient has some side effect, you have no idea which one did it.
Wysham: That's really good. So the guidelines differ for HFpEF (preserved ejection fraction). Can you just briefly comment how that changes your practice and how you manage those patients?
Butler: Honestly, I mean, if you look at the details of the guidelines, the recommendations in HFrEF and HFpEF are actually almost the same. The difference being that the level of evidence is different because in HFpEF, a lot of it is based on secondary analysis of the trials and not primary analysis. If you look at the recommendations and don't worry about the level of recommendation (whether it's 1 or 2a or 2b), the real difference is in the medicines that are being recommended. The only difference is that for EF less than 50%, it's quadruple therapy, and for EF more than 50% it's triple therapy, and beta-blockers are not recommended in patients with EF more than 50%. I want to make it clear that this is for treating HFpEF. A lot of these patients have coronary disease and ischemic heart disease, so if you have some other indication to give beta-blockers, by all means give it to them. There's no contraindication per se, but for HFpEF, there's no indication for beta-blockers.
Wysham: I think this is an important message because I have a number of patients who have heart failure and who are not on all the therapies. I speak to them about getting a heart failure specialist on board because their general cardiologist is not really doing the guideline-directed medical care. So it is something that we as primary care and diabetologists can help our patients get to the right place with and get on the right therapies, if we are aware of what the guidelines are. That is very helpful. As our last question today, I'd like to ask if the audience takes away one point from the discussion, what would you say that should be?
Butler: If you have a person with blood pressure 180 or LDL [low-density lipoprotein] of 180 or whatever, hemoglobin A1c of 9, you would never leave that person alone saying "I will treat this patient appropriately, aggressively when the person develops some symptoms." Remember, these diseases are by definition asymptomatic, and the whole idea is to treat to prevent complications. Unfortunately, for whatever reason, we think of heart failure differently, and multiple surveys show that patients are not on optimal therapy. And when you ask the clinicians why, the first answer is that the patient is, quote unquote, doing okay. I mean, it's like saying your PET scan has lit up for cancer recurrence, but you don't have symptoms, so we're not going to treat, right? That's not the mindset we should have. So please treat patients early on to prevent worsening heart failure and sudden cardiac death. Don't wait for hospitalization. Now, of course, if somebody is hospitalized, they are at high risk and given aggressive therapy, but do it as early as possible upstream.
Wysham: Javed, thank you so much for a great conversation today. We've talked about heart failure and diabetes, recognizing early symptoms, the importance of screening, and the importance of treating aggressively once the diagnosis is made. I hope that this has been helpful for the audience. Thank you so much for joining us. This is Dr Carol Wysham for InDiscussion.
Listen to additional seasons of this podcast.
Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes
Heart Failure: An Underappreciated Complication of Diabetes. A Consensus Report of the American Diabetes Association
Brain-Type Natriuretic Peptide (BNP)
The Effect of Spironolactone on Morbidity and Mortality in Patients With Severe Heart Failure
Eplerenone in Patients With Systolic Heart Failure and Mild Symptoms
2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines
Follow Medscape on Facebook, Twitter, Instagram, and YouTube
Medscape © 2023 WebMD, LLC
Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.
Cite this: Recognizing the Signs: Heart Failure and Type 2 Diabetes - Medscape - Feb 01, 2023.