Experts discuss urgent need for strategies, opportunities to improve HF care
Deaths related to other major CVDs continue to decline, but the number of patients with HF is increasing. Possible reasons for this rise include worsening CVD risk factors such as obesity and diabetes, as well as improved care and outcomes for ACS and other CVDs such as valvular heart disease. Perhaps the most important reason for this trend is the aging of the population, as HF remains primarily a disease of the elderly.
The incidence of HF has exploded. An estimated 1 million HF hospitalizations occur each year in the United States. Post-discharge recurrent readmission and mortality rates remain unacceptably high despite improved signs and symptoms during index hospitalization, and hospital reimbursement is now tied to 30-day readmission rates.
Cardiology Today conducted a round table discussion with experts in this field to discuss current and future care of patients with HF, with particular focus on biomarkers, novel drugs and treatment methods.
Round Table Participants
Carl J. Pepine, MD
Javed Butler, MD, MPH
James L. Januzzi Jr., MD
Bertram Pitt, MD
Randall C. Starling, MD, MPH
John R. Teerlink, MD
Current issues in management
Carl J. Pepine, MD: Gentlemen, welcome to Cardiology Today. Let’s begin: Where are we today?
James L. Januzzi Jr., MD: HF is a problem that every physician in modern medicine will contend with. We have seen dramatic increases in HF even in just the past few years. Based on current trends, the issue will likely grow moving into the future.
John R. Teerlink, MD: We have learned that the management of HF patients in hospital may be different than out of hospital. Recently, we have focused more attention on the application of new therapies to inpatients and the development of new therapies for acute HF. But, mortality and rehospitalization rates have not changed during the last few decades despite tremendous advances in chronic HF.
Pepine: What about the issue of repeat hospitalizations for HF?
Javed Butler, MD, MPH: This is a large clinical and economic issue. Although a recent paper by Chen et al published in JAMA in 2011 suggested that relative rates of readmission may have leveled off and started to decline in the last decade, the absolute number of hospitalizations has remained at more than 1 million, since the prevalence of HF continues to increase. The costs are enormous: $40 billion in 2010, with more than 50% of this cost related to hospitalizations. Once hospitalized, a patient has an approximate 20% to 25% risk for readmission within 1 month and 50% risk within 6 months. More importantly, these patients have an approximate 30% risk for dying within 1 year of discharge, a figure that has not changed in the past 2 decades.
Identification of high-risk patients
Pepine: Using HF risk factors beyond age and hypertension, is it possible to reliably identify a cohort with a very high probability of HF hospitalization in the near future?
Teerlink: A number of predictive tools have been developed to determine risk for rehospitalization, upon hospital discharge and in the outpatient setting. The models focus primarily on things that we believe make physiologic sense. For example, patients with advanced renal dysfunction, those who are not completely decongested and have persistent signs or symptoms of HF, and those with other comorbidities are at increased risk for readmission. So there are many options available to attempt to predict readmission risk. What’s emerging is not just biomarkers or the clinical impression or the patient’s individual history; it is a combination of all those factors that empowers us to see a patient’s future.
Januzzi: This is the problem that we face in HF: Patients are incredibly complicated. The comorbidities are considerably more numerous than in patients presenting with a first MI, for example. The challenge of managing these patients is not just about preventing a HF hospitalization, but preventing the many non–HF-related hospitalizations that also arise. Any model to predict risk must consider all the comorbidities.
Importance of prevention
Butler: With the growing elderly population in the United States, it is important to focus on prevention.
Pepine: Dr. Pitt, you were a pioneer in the first HF prevention trial, SOLVD. What can you tell us from your experience?
Bertram Pitt, MD: We have made great progress, but there are now opportunities to go further in acute HF, diastolic HF and HF prevention. However, with the exception of the SOLVD Prevention Trial, we haven’t made great strides. The clue would be identifying high-risk individuals. We are now trying to plan a prevention trial, but it is not easy. We are starting with patients with diabetic nephropathy.
Then, the question is: Should we use biomarkers? They would add to the trial, but may create a burden. It becomes a trade-off between identifying high-risk individuals, applicability, sample size, budget and so on. It is a challenge, but there are unbelievable opportunities.
Randall C. Starling, MD, MPH: Importantly, we need to think about the whole continuum of care for the HF patient. We have focused on hospital and outpatient settings, but are learning more about the importance of the process of discharge from the hospital to the outpatient setting. That is where the question related to prevention comes in. If we use the model developed by the American College of Cardiology and the American Heart Association with a continuum from prevention to end-stage care, it really sets the platform as we move into accountable care and the responsibility to not deal with “an episode,” but rather deal with “the patient.” It is going to force cardiologists, in a good way, to begin thinking in those terms.
What I’m learning is that relationships with primary care physicians, hospitalists, referring physicians and the like will become increasingly important to create the awareness to treat all risk factors. The net effect, we would all agree and hope, is to see fewer patients with advanced HF in the hospital.
Pitt: There is reluctance by the pharmaceutical industry to get involved because the pricing of prevention and new strategies is very low, at least in Europe and other areas. The interest in terms of pricing seems to be more in the treatment of advanced HF.
Butler: If you look at the CHD paradigm over the past 40 to 50 years, we individualize the person’s risk and then administer therapy according to an individual person’s risk. In looking at current clinical prevention guidelines and risk prediction models, the focus is on either coronary disease specifically or vascular disease in general, including peripheral vascular disease and stroke, but there is no focus on HF, per se. One can argue that if you manage CHD, perhaps you take care of HF as well. However, as more and more HF in the future will be seen in elderly patients who are particularly at risk for HF with preserved ejection fraction (HFpEF), the current data suggest that up to 30% to 40% of HF in this population is not related to clinically manifest CHD. It is important to begin thinking in such terms of trying to identify and then target the individual’s risk.
Teerlink: Another challenge is exemplified by the atherosclerosis paradigm. With an MI, there is a thrombus to target, a distinct pathophysiology that, in general, is consistent across the board. HF is the final common pathway, if you will, for multiple different disease processes. We need to focus on specific HF patient subgroups by individualizing therapies, biomarkers and other approaches, but also should be open to large trials that target this general pathway, although, it presents conflicting requirements.
Usefulness of biomarkers
Pepine: It sounds like we need earlier detection and intensified management before patients even get to the hospital. Dr. Januzzi, as a pioneer in biomarkers, what might help with early identification of a cohort that we could address before they need to be hospitalized?
Januzzi: A number of advances in the past several years have looked specifically at biomarker testing to identify individuals at the highest risk for HF — those most likely to require near-term hospitalization. The goal in the creation of these tools is to develop strategies around more intensive management to try to prevent such near-term events as hospitalization. It is complicated, of course, because many biomarkers have varying reasons for a rise or fall. It is not as simple as looking at a blood test. It is a nuanced process, and that’s where the importance of being a good clinician also comes in.
Pitt: Has a specific test emerged as a favorite?
Januzzi: Recent experience is largely focused on natriuretic peptides — BNP and NT-proBNP. We have conducted studies looking at NT-proBNP surveillance and management in chronic HF patients; we manage patients with aggressive clinical care, but have a supplementary goal of suppressing NT-proBNP below 1,000 pg/mL or as close to that goal as possible.
We have data from a prospective, randomized trial showing reduced HF hospitalization by 50%. It makes sense that this approach — utilizing an objective tool to identify risk — would be expected to be accompanied by better outcomes, given the high risk of these patients. And, that’s exactly what we found.
In the longer term, newer biomarkers seem even more prognostically powerful than natriuretic peptides. Specifically, we have focused on ST2, a biomarker of remodeling and risk for progressive HF. That seems to have interesting response to therapy. We are also researching galectin-3, another novel marker.
Pepine: Do we have clues into why patients get into the cycle of recurring admissions for HF?
Januzzi: Yes. The observation in patients who have become refractory is that on a clinical level they seem recompensated, but if you look at the profile of not just cardiac biomarkers but the wide palette of markers, including renal dysfunction markers and others, we see that these patients are hardly recompensated, at least biochemically. In a way, it links back to prevention, which is you can detect things before patients realize they have underlying structural heart disease. The same happens in patients who have had a recent decompensation. Put another way, we’ve all had examples of the patient in whom diuresis of maybe 5 lb occurs after admission. You see them; they swear they’re feeling better. By all rights they seem improved; their neck veins are down, they seem more compensated, we send them home, but in a short time they are back in the ED. It has been recognized over the last 10 years that if you measure certain markers in these patients, including the natriuretic peptides, you see substantial improvement in elevations of these markers in patients at lower risk for rehospitalization, whereas those patients without a decline are at significant risk for rehospitalization.
Yet to be proven is whether changing therapies in these patients will prevent that recurrence. That is a trial that we’d love to do, but such trials are challenging in terms of getting funding and, by definition, would involve prolongation of hospitalization, which hospital administrators are not thrilled about.
Novel therapy development
Pepine: Let’s move to novel therapies on the horizon. What is your perspective relative to some new areas for management of HF that appear to be particularly promising?
Teerlink: There was the initial rise of nesiritide (Natrecor, Scios), which opened a new arena for novel therapy development. Within the vasodilators, it has been clear that vasodilator therapies in early treatment of acute HF patients improve symptoms. That is clear with nitroglycerine, as well as nitroprusside. The challenge with those therapies is they have not been studied to demonstrate this improvement in a contemporary, clinical context.
Januzzi: Even furosemide, we are only now learning about how to use it.
Teerlink: I’m still not sure we know how to use it. Within that arena, a number of additional vasodilator therapies emerged. Early trials with endothelin receptor antagonists showed promise and then unfortunately failed. Other early natriuretic peptides also emerged and failed because of hypotension and related adverse events.
Recently, we reported results of the 1,161-patient RELAX-AHF trial, which investigated recombinant human relaxin-2 (serelaxin, Novartis) in patients with acute HF. Patients had improvements in dyspnea, liver and renal function tests, worsening HF events and, interestingly, a 37% reduction in both CV and all-cause mortality. More than 50% of the patients had positive troponins, and those treated with serelaxin had less troponin increase during hospitalization and also experienced improvements in renal function and functional markers.
There is a possibility that if we intervene with a medicine that can rapidly affect organ damage we might be able to obtain later improvements in survival.
Butler: We are excited about a number of drugs. No single therapy can affect all the causes why people come into the hospital and have poor outcomes. One very important concern is the issue of congestion. The patients look OK clinically, but about 50% or so lose minimal weight by the time of discharge. Looking at a comprehensive congestion score, even among patients who seem well at discharge and have a congestion score of 0 or 1, if you draw a BNP level it is often still in the 600 pg/mL to 700 pg/mL range, which indicates subclinical vascular congestion, regardless of whether it is related to total volume or redistribution. At least a greater part of the readmission or poor outcome among these patients post-discharge may be related to persistent congestion.
The question is: How can we better decongest patients?
Growing interest in ultrafiltration
Butler: Diuretic therapy continues to be the standard care, but as we all know there are limitations to this approach. One alternate strategy is ultrafiltration. This therapy remains controversial, as it has been from the beginning. Theoretically, there are benefits of ultrafiltration, in terms of the amount of sodium lost and its effect on neurohormonal activation or electrolyte imbalance and so on. There are two potential ways by which we can use ultrafiltration — as an initial strategy or limited to patients resistant to diuretics.
For the first approach, we have a clinical trial of about 200 patients in which ultrafiltration use was associated with an approximate 40% reduction in readmission risk compared with diuretics. However, we need a larger trial to confirm these results. One such trial — AVOID-HF — is ongoing right now.
Currently, ultrafiltration is perhaps used more for patients with diuretic resistance, and for that use we have no good randomized clinical trial data. The CARRESS-HF study was conducted in cardiorenal syndrome patients, defined as a rise in creatinine level, but these patients had approximately 12-lb weight loss in 4 days. Hence, these patients may have fulfilled the criteria for worsening renal function, but not diuretic resistance, per se. So, here is another population where we can potentially look at this application.
I would be even more interested in a trial comparing ultrafiltration with diuretic therapy to a goal directed to decongest patients, whether targeting natriuretic peptides or weight or clinical congestion score, or any other hemodynamic or clinical measure, rather than comparing these approaches to an arbitrarily decided time point. Unfortunately, we don’t yet have that trial.
Pepine: Ultrafiltration sounds and looks attractive. What has been the major barrier to the HF community embracing ultrafiltration?
Starling: It is like religion to some degree in that some strongly believe in it and others don’t believe in it at all. We’ve published a few papers, albeit small cohorts of patients from the Cleveland Clinic, and reported high mortality in patients who received ultrafiltration. As Dr. Butler alluded, one must be careful about patients with diuretic resistance and high creatinines, whether that is a marker for high mortality in the patient population and/or there’s something about the therapy that’s adding to that. That was one question raised in the CARRESS-HF study.
It seems to be a matter of logistics, an issue of cost and of evidence; until there is more evidence, this divergence of opinions will continue.
Pitt: We are still in a very early stage of research. I don’t think we yet understand patient selection or endpoints.
We are starting the STAR AHF trial, which will involve patients who have undergone ultrafiltration and will be randomly assigned to a high-dose aldosterone-blocking strategy over 30 days. We will compare various endpoints. In past trials, we may have over-diuresed patients and dropped them down too much; in STAR AHF, we are going to use a radioisotope blood volume-guided strategy and look at several noninvasive strategies to correlate to see if we can learn for the future.
Part two of this round table will appear in the August issue.
Expect more expert discussion on phenotyping, vagus nerve stimulation, education and training, and other current issues in HF management.
Disclosures: Butler reports consulting for Bayer, BH Medicine, Gambro, Medtronic, Novartis, Ono and Trevena. Januzzi reports receiving grant support from BG Medicine, Critical Diagnostics, Roche Diagnostics, Siemens, Singulex and Thermo-Fisher. Pepine reports no relevant financial disclosures. Pitt reports no relevant financial disclosures. Starling reports no relevant financial disclosures. Teerlink reports research grants/honoraria from Amgen, Corthera, Cytokinetics, Novartis, Servier and Trevena.