DR PUNZI: Hello. My name is Dr Punzi. I’m an internist and hypertension specialist in the Dallas-Fort Worth area. With me are Dr McCullough and Dr Verma and we will be discussing the issue of heart failure (HF). The purpose for this discussion is to educate primary care physicians, cardiologists, and endocrinologists on the key HF risk factors in patients with type 2 diabetes mellitus (T2DM). The issue of HF is pretty significant. In the United States, there are 6.5 million adults with HF. The American Heart Association (AHA) statistics show that 1 of every 8 deaths in 2017 was related to HF.1 Two percent of the population has HF and the cost of HF in 2012 was about $30.7 billion. We have numerous studies that show that, in the treatment of T2DM, there is a significant benefit in the reduction of HF hospitalizations.
With that, I’d like to have Dr McCullough introduce yourself and give a little bit of your background on HF and the clinical aspects. One of the important problems that we see is, as clinicians who have been educated over the past 2 to 3 decades, that diabetes has not been on the list of major risk factors for the incidence of HF. Even in the latest 2019 statistics from the AHA, in their summary, they have diabetes way down on the list when compared with other risk factors.1 Dr McCullough, will you discuss that for us?
DR MCCULLOUGH: Yes, well, thank you for that opening introduction and set of remarks. I’m Peter McCullough. I’m an internist and cardiologist and Vice Chair of Medicine at Baylor University Medical Center in Dallas, TX. As a medical cardiologist, I’ve been particularly interested in the interface between heart and kidney disease and diabetes, of which 90% is T2DM and is the leading cause of chronic kidney disease. In my view, some of the early epidemiologic studies including the Framingham Heart Study2 really miss the connection between HF and diabetes. I think largely because during that time HF was dominated by patients with coronary heart disease and valvular disease and there was a really heavy overlay of untreated hypertension. So, the modern epidemiologic observations are, just as you stated, that T2DM really is a major risk for HF.
DR PUNZI: Dr Verma, would you care to introduce yourself and give us your perspective on the association of T2DM and HF?
DR VERMA: Yes, I’m Subodh Verma. I’m a cardiac surgeon and a Professor at the University of Toronto with an interest in diabetes and cardiovascular disease, specifically diabetes and HF. I’m involved in all of the major large outcome trials with sodium-glucose cotransporter-2 (SGLT-2) inhibitors in the treatment of HF.
I completely agree with what Professor McCullough just said—it is not until recently that we’ve had an appreciation of a problem that is actually quite old and that is that HF is a common occurrence in people with diabetes and one of the most fatal, grievous complications of T2DM. I think the fact that we’re now seeing that rates of ischemic heart disease are declining because of good antiatherosclerotic treatments, the importance of HF as an outcome in patients with diabetes is becoming apparent.
You don’t value what you don’t measure, and it wasn’t until the cardiovascular outcome trials, even the initial ones with dipeptidyl peptidase-4 (DPP-4) inhibitors, which showed a neutral outcome on major adverse cardiovascular events, that we were reminded how important HF was as an outcome. Then, of course, from there came the SGLT-2 inhibitor trials that have shown us not only the importance of this outcome, but that this is something that can be prevented and potentially even treated in people with diabetes.
DR PUNZI: Thank you, Dr Verma. So, I think one of the issues at hand for some of our colleagues is HF—how does this disease affect our patients? Having trained many years ago, the treatment of HF was merely symptomatic, trying to make the patient feel better, but there was no improvement in the mortality. We know that the 30-day mortality for HF is about 10%, at 1 year is between 20% and 30%, and at 5 years is between 45% and 60%.3 So, I think that as clinicians we should be more aware of what the morbidity and mortality implications are when the patient is diagnosed early on with HF. With that said, Dr McCullough, tell us can weidentify these patients earlier in the disease process and get a good handle on what to do or not to do with these patients?
DR MCCULLOUGH: The largest study in HF, REACH,3 demonstrated that roughly half of the patients with HF had HF with reduced ejection fraction (EF), which was the traditional view of HF, and half had HF with preserved EF. One of the important observations, supported by many other studies, is that HF has a very high cause-determined mortality, meaning that a patient with HF has roughly a 90% chance of dying of HF––HF would be the cause of death. The two major divisions in mortality are either pump failure death or sudden arrhythmic death.
There is an urgency to identify patients with HF because there really are disease-modifying mortality-reducing therapies in HF. One of the things we showed in REACH years ago was that about a third of patients were diagnosed in an ambulatory setting alone and about two thirds were diagnosed with an incident hospitalization. I think a real goal would be to identify more patients before they’re ever hospitalized in the ambulatory setting. So, it really takes a clinical suspicion, the use of blood testing for natriuretic peptides—brain natriuretic peptide or N-terminal pro-brain natriuretic peptide—and the use of echocardiography for the detection of HF in the ambulatory setting.
DR PUNZI: Dr Verma, would you care to comment on that?
DR VERMA: Yes, so again, I would agree with my colleague and maybe add a couple of comments to say that we now have data that in people with diabetes of all of the known complications, having diabetes plus HF carries the worst mortality compared with any other complication in diabetes, whether it’s peripheral arterial disease or chronic kidney disease, prior myocardial infarction, or history of retinopathy, etc. So, just recognizing the fact that this combination is one of the worst with respect to mortality is important, but it also begs the question on why we are waiting until people fall off the cliff and develop this bad diagnosis. We should be really investing in trying to understand how to identify these patients earlier.
There are various ways of doing that. Some have already been articulated nicely by Professor McCullough. They can either be based on clinical suspicion, duration of diabetes, looking at biomarkers, looking at people who have multiple risk factors, and really having a heightened sense of suspicion of occult HF with preserved or reduced EF in these patients early, because when we say diabetes and HF to primary care physicians or to other clinicians, immediately they think about patients with established HF who have classic signs and symptoms of HF. You have therapies, you can modify their prognosis to some extent––but, really, by then, the horse has left the barn.
DR PUNZI: The damage is done.
DR VERMA: The damage has been done and the opportunity to prevent that has been lost. So, I think that is a key feature. I’ll also add that this link between HF and renal disease is a very important one because both of these are complications that occur not only coincident with each other in people with diabetes but really serve to accelerate, in a bidirectional fashion, the outcome of each entity such that renal disease is a very strong determinant of HF and as the pump gets worse so does the kidney. So this bad relationship between the heart and the kidney is also something that inevitably happens in this population and requires us to think about the prevention not just of HF but the prevention of the cardiorenal or the HF renal continuum in this patient population.
DR PUNZI: Right. Thank you, Dr Verma. Looking at the context of risk factors such as diabetes, hypertension, and coronary heart disease, Dr Haffner of the San Antonio Heart Study4 felt that in his patient population, a diagnosis of diabetes carried a similar cardiovascular risk to coronary heart disease. This coincides with what we’re seeing today, where 50% of people with T2DM could end up developing HF.
There are some European studies that evaluated the prevalence of HF in diabetic patients. Dr Shah and colleagues5 in the United Kingdom studied a large prospective cohort using linked electronic health records. Of the 1,921,260 patients, 34,198 had diabetes (1.8%). Of these diabetic patients, 6137 (17.9%) had a first cardiovascular presentation, the most common of which were peripheral vascular disease (992 [16.2%] of 6137 patients) and HF (866 [14.1%] of the 6137 patients).
Dr Boonman-de Winter et al6 evaluated 605 patients aged 60 and older with T2DM. The standardized diagnostic workup included medical history, physical examination, ECG, and echocardiography. Of the 581 patients studied, 161 (27.7%) were found to have previously unknown HF: 28 (4.8%) with reduced EF and 133 (22.9%) with preserved EF. Left ventricular dysfunction was also diagnosed in 150 patients (25.8%); 146 (25.1%) had diastolic dysfunction. This implies that in the diabetic population we’re starting to see HF manifesting earlier in the disease process and without the physician’s knowledge. How do we predict which diabetic patients are at greater risk? What tools can clinicians use at the bedside to evaluate these patients? There is also a paradigm shift on when we should start treatment of hypertension, and also in the hypertensive diabetics. I think that when you look at data from numerous studies, especially our SPRINT study,7 the underlying theme is that lower is better to decrease the risk of cardiovascular disease including HF. What do you see as a feature for clinicians to be able to identify these patients earlier and start treatment?