HFPEF’s diverse features pose management and trial challenges

When the highly-anticipated results from the TOPCAT trial reported last November failed to show a statistically significant efficacy benefit from spironolactone treatment in patients with heart failure and preserved left ventricular ejection fraction for the study’s primary endpoint of death or heart failure hospitalization, the neutral result represented more than another failed treatment trial in this rapidly expanding patient population that remains bereft of treatments with proven efficacy.

TOPCAT’s results also underscored the heterogeneity of patients diagnosed with heart failure with preserved ejection fraction (HFPEF), a signature feature of the syndrome. Heterogeneity is believed to have played a large role in the lack of success in TOPCAT as well as in several other major HFPEF trials.

Although HFPEF is defined as patients who exhibit clear clinical symptoms of heart failure but with a LVEF of at least 45%, and more recently at least 50%, what’s become increasingly clear as neutral trial results accumulate is that HFPEF is a syndrome. Far from a single pathologic entity, it can have different etiologies and present as different subtypes in a spectrum of abnormalities and severities.

HFPEF "is not just a simple disease of diastolic function and a stiff ventricle. It’s a very complicated, highly integrated, multisystem failure of cardiovascular and peripheral reserve," said Dr. Barry A. Borlaug, a cardiologist at the Mayo Clinic in Rochester, Minn. "A lot of us are coming to the conclusion that there are different subphenotypes of HFPEF, each with unique mechanisms and clinical trajectories that probably need unique treatment," he said in an interview.

"This is a very heterogeneous syndrome with no one etiology and a wide range of patients, some with marked ventricular hypertrophy, some with diastolic dysfunction, some who are fairly asymptomatic but become symptomatic with exercise, some who are compensated and stable, and others who are uncompensated. The spectrum of disease is very broad, and it may be many diseases," said Dr. Scott D. Solomon, professor of medicine at Harvard Medical School and director of noninvasive cardiology at Brigham and Women’s Hospital, Boston.

Last year, the heart failure literature featured two editorials spelling out proposed visions of the HFPEF spectrum. One of these proposed three characteristic types of HFPEF patients (J. Am. Coll. Cardiol. 2013;62:1339-42): those with exercise-induced diastolic dysfunction with no symptoms at rest, minimal fluid retention, and never hospitalized for heart failure but with long-standing hypertension and exercise intolerance; patients with volume overload and edema, recently hospitalized heart failure with dyspnea on exertion, and moderately severe heart failure symptoms; and the worst form, patients who have developed pulmonary hypertension and right ventricular failure as a consequence of their HFPEF and now have frequent heart failure hospitalizations. But patients do not necessarily progress from one severity stage to the next, noted Dr. Sanjiv J. Shah, a cardiologist at Northwestern University, Chicago, who wrote the editorial.

"Patients can progress back and forth across the spectrum" or remain in one stage, he said in an interview. "The syndrome is quite heterogeneous, not only in clinical presentation, but in etiology, and pathophysiology. There is still much to learn about the natural history of HFPEF , how it develops, and the trajectory of patients."

The second editorial described several relatively common HFPEF phenotypes that can appear individually or in combination, including filling limitation, ejection limitation, cardioacceleration, and vasoregulation (Eur. Heart J. 2013;34:1393-5). "Patients with heart failure caused by severe mitral insufficiency or aortic stenosis will clearly behave differently and respond to treatments differently from patients with hypertropic cardiomyopathy, constrictive pericarditis, or high-output heart failure. However, despite this heterogeneity these entities continue currently to be lumped together into the category of HFPEF," Dr. Borlaug wrote in this editorial.

The perils of heterogeneity

In addition to providing a better framework for understanding the causes and consequences of HFPEF, the paradigm of HFPEF as a heterogeneous syndrome also helps explain why the major intervention trials that focused on HFPEF patients, starting a decade ago with studies such as CHARM Preserved (Lancet 2003;362:777-81) and I-PRESERVE (N. Engl. J. Med. 2008;359:2456-67), and continuing through to the recent TOPCAT, have all failed to produce a statistically significant benefit for their primary endpoints.

"Heterogeneity confounds our ability to identify effective treatments. It’s the most likely single explanation" for the neutral HFPEF trials, Dr. Borlaug said. "I don’t think it’s as simple as one thing, but my speculation is that it’s the dominant reason."

"Many trials such as CHARM Preserved and TOPCAT seemed to enroll some patients who in retrospect did not have HFPEF," noted Dr. Gregg C. Fonarow, professor of medicine and associate chief of cardiology at UCLA in Los Angeles. "The key to future clinical trials will be better classification of HFPEF" and better matching of HFPEF patients to their treatment, said Dr. Shah.

"In the neutral megatrials some of the patients did not have HFPEF but had other disorders such as deconditioning and obesity – things that cannot be helped by treatments directed at the heart. We need to identify the patients who will respond to the treatment," said Dr. Burkert M. Pieske, professor and director of cardiology at the Medical University of Graz, Austria.

The TOPCAT (Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist) results reported last November exemplified the problem of patient heterogeneity. The study enrolled more than 3,400 patients in six countries: Argentina, Brazil, Canada, Russia, the Republic of Georgia, and the United States. Retrospective analysis showed that among patients in the placebo group, the primary endpoint of death or heart failure hospitalization during follow-up occurred at a rate of 12.6 events/100 patient-years among the 881 patients treated in the four Western Hemisphere countries, and at a rate of 2.3 events/100 patient-years among the 842 patients treated in Russia or Georgia, a greater than fivefold difference between the two subgroups.

"As a whole, the patients in Russia and Georgia had a lower event rate and were certainly less severely ill than the other patients, but they supposedly still had heart failure. Defining this disorder is difficult, and when a patient has signs and symptoms of heart failure and preserved ejection fraction you may not be certain that heart failure is causing the symptoms. That’s why many people think that we should use another criterion" to define HFPEF in trials, such as elevated serum level of some form of natriuretic peptide, said Dr. Solomon, a TOPCAT coinvestigator. By design, patients could enter TOPCAT either because of a recent heart failure hospitalization, which is how 72% of patients got in, or by having a threshold level of natriuretic peptide, the way the remaining 28% entered the study. Within the subgroup enrolled by natriuretic peptide level, spironolactone treatment had a statistically significant effect in reducing the primary endpoint, while in the other 72% the drug produced no discernable benefit over placebo.

"TOPCAT had funding problems, recruitment problems, and monitoring problems. There was not enough money for good monitoring," said Dr. Pieske, who did not participate in TOPCAT. The patients enrolled in Russia and Georgia were "unbelievably stable," with their 2.3/100 patient-years event rate. "Even if the drug works, it can’t exert an effect if there are no events."

Dr. Pieske ran a much smaller study of spironolactone involving 422 patients with HFPEF, the Aldosterone Receptor Blockade in Diastolic Heart Failure (Aldo-DHF) trial, which showed efficacy for the endpoint of improved diastolic function but not for the co-endpoint of improved exercise capacity (JAMA 2013;309:781-91).

"If you look at TOPCAT and Aldo-DHF together and ask, does spironolactone work, the answer is yes if you select the right patients, those with clear evidence of cardiac functional abnormalities and increased natriuretic peptide levels or documented hospitalization for heart failure. The drug clearly works in these patients, but there were enough inconsistencies in these trials so that this will not get into the treatment guidelines," Dr. Pieske said in an interview.

"Spironolactone provides most benefit in patients who are volume overloaded, have right-heart failure, or both. These are also the patients most likely to have elevated natriuretic peptide. Spironolactone does not seem to benefit as much early-stage HFPEF," noted Dr. Shah.

Several experts cautioned that because the TOPCAT results failed to produce a statistically significant benefit for the study’s primary endpoint, any other conclusions from the results must be made very cautiously, and several also said that they wanted to see the full, published results which had not appeared as of late February. They also noted that many patients with HFPEF, most likely a majority, do not have elevated blood levels of natriuretic peptide.

Where HFPEF management stands now

Evidence-based guidelines for the specific treatment of HFPEF are simple. There aren’t any.

The American Heart Association/American College of Cardiology heart failure management guidelines released last October made no disease-specific treatment recommendations (J. Amer. Coll. Cardiol. 2013;62:e147-e239). They call for controlling hypertension in patients who need intervention by prevailing standards, treatment with a diuretic in patients with volume overload, coronary revascularization if coronary disease is present, and management of atrial fibrillation if that’s present. The European Society of Cardiology’s guidelines issued in 2012 say pretty much the same (Eur. J. Cardiol. 2012;33:1787-1847), except they add this statement: "No treatment has yet been shown, convincingly, to reduce morbidity and mortality in patients with HFPEF."

If nothing else, TOPCAT’s results further solidified spironolactone’s role as a reasonably safe drug for blood pressure lowering in HFPEF patients who are also hypertensive. Hypertension control is a must for patients with HFPEF as it’s believed to significantly contribute to HFPEF in most patients. "Hypertension is the most common underlying cause of HFPEF," said Dr. John R. Teerlink, professor of medicine and director of the heart failure program at the San Francisco Veterans Affairs Medical Center.

When it comes to treating hypertension in HFPEF patients, "unless there is a contraindication, spironolactone should be one of the first drugs to try, but I’m not sure you can say it’s the first choice" based on current evidence, Dr. Borlaug said.

A diuretic is an obvious antihypertensive for HFPEF patients with fluid overload. And there were suggestions of benefit from the ACE inhibitor perindopril in elderly patients in the PEP-CHF trial (Eur. Heart J. 2006;27:2338-45), another large HFPEF-treatment trial that failed to show significant benefit for the primary endpoint with perindopril treatment but had positive results in some secondary endpoints.

Patients "get these drugs to treat their hypertension, but we believe they may also help their heart failure. It’s a belief system. My colleagues and I already use a lot of spironolactone to treat hypertension, and the TOPCAT results won’t change my practice. I’m not comfortable telling people that you should use an aldosterone antagonist because of TOPCAT," Dr. Teerlink said in an interview.

"An angiotensin-converting enzyme inhibitor will probably work; spironolactone probably works if you get the diagnosis right; we don’t know about beta-blockers; and there is some evidence for using digoxin," said Dr. John G.F. Cleland, professor of cardiology at the University of Hull, Kingston-upon-Hull, England. "There must also be good treatment of hypertension, and judicious use of diuretics."

"I’m careful with beta-blockers; I’ve had some patients who felt miserable on them," said Dr. Borlaug. "Some HFPEF patients don’t have a stroke volume, they don’t have diastole, all they have is heart rate, and if you take that away they are left with no cardiac output."

Aside from controlling blood pressure, experts advise good management of other comorbidities such as coronary disease, atrial fibrillation, diabetes, sleep disordered breathing, and renal disease. "Clinicians should make sure that they are not missing severe coronary artery disease, infiltrative cardiomyopathy, constrictive pericarditis, or other causes of HFPEF that have specific treatments," said Dr. Shah. Some drugs have shown promise in early-phase studies – such as ivabradine and neprilysin – but phase III trials are needed. "My advice on how to manage patients with HFPEF is to make every effort to enroll them in a randomized clinical trial," said Dr. Fonarow.

Another key is making the diagnosis, ideally to prevent development of irreversible cardiovascular damage. "HFPEF is difficult to diagnose with certainty unless you do a cardiac catheterization to measure filling pressures, but in patients with early-stage HFPEF, even their invasive hemodynamics look pretty normal," said Dr. Borlaug. Another approach is an exercise echo, which is noninvasive and can identify stress-induced diastolic dysfunction, but the sensitivity and specificity of this approach remains uncertain, he said. And an elevated natriuretic peptide level can help nail the HFPEF diagnosis in some patients but many other patients have levels within the normal range.

To find HFPEF patients, apply a low index of suspicion and look for breathlessness, loss of functional capacity, signs of congestion, lung crackles, echocardiographic signs, pulmonary artery hypertension, and an enlarged left atrium, he suggested.

"Finding patients is a big challenge," said Dr. Pieske. HFPEF patients tend to be elderly, women, and people who are obese who have hypertension and perhaps diabetes. "They complain of dyspnea and fatigue and many physicians think this is just how it is. They will not consider that there is a true diagnosis behind these symptoms and complaints," especially if the patient has preserved left ventricular ejection fraction and a normal natriuretic peptide level." A diastolic stress test using exercise and an echo exam may identify stable patients with early-stage diastolic dysfunction but this requires confirmation, he said.

HFPEF dominates heart failure, lacks good treatment

Experts say the onus on physicians to diagnose and manage HFPEF will grow substantially, since HFPEF is not only highly prevalent but also increasing faster than heart failure with reduced ejection fraction (HFREF). Results from 13 community-based studies published during 1997-2006 showed that HFPEF represented an average of 55% of all heart failure cases, Dr. Carolyn S.P. Lam said last November during a talk at the American Heart Association’s scientific sessions.

Two recent reports of U.S. data documented that the increasing prevalence of HFPEF is outpacing that of HFREF by 1% per year, driven by the aging of the American population (older age is a risk factor for HFPEF) and by the increasing prevalence of comorbidities that contribute to HFPEF. One of the reports, based on U.S. national data collected by the Get With the Guidelines Program, extrapolated that by the end of this decade about 65% of all patients hospitalized for heart failure will have a left ventricular ejection fraction of 40% or greater, the vast majority with HFPEF (Curr. Heart Fail. Rep. 2013;10:401-10). What’s coming is a HFPEF "epidemic," said Dr. Lam, a cardiologist at the National University Heart Centre in Singapore.

The fact that HFPEF is now at least as prevalent as HFREF, that it is about as lethal independent of what comorbidities might contribute, and that it is growing more prevalent than HFREF sharply contrasts with the absence of treatments with proven efficacy. How did that happen?

"Research into HFPEF is decades behind research into HFREF," said Dr. Fonarow. "HFPEF was largely ignored because the prevailing wisdom [20 or so years ago] was that it represented only a small proportion of cases and that outcomes were much better compared with HFREF." Only with registry data and community-based studies run more recently did cardiologists realize that outcomes from HFPEF were as bad as from HFREF. Progress was further hampered by enrollment of patients who did not have HFPEF or a broad spectrum of patients unable to respond equally well to the treatment under study.

"There are large gaps in knowledge of the pathophysiology of HFPEF and little investment has been made to identify potential therapeutic targets. This creates cascading levels of challenges for developing effective treatments despite the massive unmet need," said Dr. Fonarow.

"We don’t know what to treat yet," was Dr. Teerlink’s summation of the HFPEF dilemma.

Dr. Borlaug said that he has been a consultant to GlaxoSmithKline, Merck, Amgen, CardioKinetix, and DC Devices and has received research support from Atcor. Dr. Solomon said he has been a consultant to and received research support from Novartis and more than 10 other drug and device companies. Dr. Pieske said that he has received honoraria from Bayer, Boehringer Ingelheim, Servier, Medtronic, Bristol-Myers Squibb, Menarini, and Novartis and received research support from Bayer and Medtronic. Dr. Shah said that he has been a consultant to Novartis, Bayer Schering, and the Pulmonary Hypertension Association. Dr. Fonarow said that he has been a consultant to Medtronic, Novartis, and Pfizer. Dr. Teerlink said that he has been an adviser to and received research support from Novartis. Dr. Cleland said that he has received honoraria from Novartis and research funding from Pfizer. Dr. Lam said that she has been a consultant to Bayer, Novartis, and DC Devices and has received research support from Boston Scientific, Medtronic, and Vifor Pharma.

[email protected]

On Twitter @mitchelzoler

When the highly-anticipated results from the TOPCAT trial reported last November failed to show a statistically significant efficacy benefit from spironolactone treatment in patients with heart failure and preserved left ventricular ejection fraction for the study’s primary endpoint of death or heart failure hospitalization, the neutral result represented more than another failed treatment trial in this rapidly expanding patient population that remains bereft of treatments with proven efficacy.

TOPCAT’s results also underscored the heterogeneity of patients diagnosed with heart failure with preserved ejection fraction (HFPEF), a signature feature of the syndrome. Heterogeneity is believed to have played a large role in the lack of success in TOPCAT as well as in several other major HFPEF trials.

Although HFPEF is defined as patients who exhibit clear clinical symptoms of heart failure but with a LVEF of at least 45%, and more recently at least 50%, what’s become increasingly clear as neutral trial results accumulate is that HFPEF is a syndrome. Far from a single pathologic entity, it can have different etiologies and present as different subtypes in a spectrum of abnormalities and severities.

HFPEF "is not just a simple disease of diastolic function and a stiff ventricle. It’s a very complicated, highly integrated, multisystem failure of cardiovascular and peripheral reserve," said Dr. Barry A. Borlaug, a cardiologist at the Mayo Clinic in Rochester, Minn. "A lot of us are coming to the conclusion that there are different subphenotypes of HFPEF, each with unique mechanisms and clinical trajectories that probably need unique treatment," he said in an interview.

"This is a very heterogeneous syndrome with no one etiology and a wide range of patients, some with marked ventricular hypertrophy, some with diastolic dysfunction, some who are fairly asymptomatic but become symptomatic with exercise, some who are compensated and stable, and others who are uncompensated. The spectrum of disease is very broad, and it may be many diseases," said Dr. Scott D. Solomon, professor of medicine at Harvard Medical School and director of noninvasive cardiology at Brigham and Women’s Hospital, Boston.

Last year, the heart failure literature featured two editorials spelling out proposed visions of the HFPEF spectrum. One of these proposed three characteristic types of HFPEF patients (J. Am. Coll. Cardiol. 2013;62:1339-42): those with exercise-induced diastolic dysfunction with no symptoms at rest, minimal fluid retention, and never hospitalized for heart failure but with long-standing hypertension and exercise intolerance; patients with volume overload and edema, recently hospitalized heart failure with dyspnea on exertion, and moderately severe heart failure symptoms; and the worst form, patients who have developed pulmonary hypertension and right ventricular failure as a consequence of their HFPEF and now have frequent heart failure hospitalizations. But patients do not necessarily progress from one severity stage to the next, noted Dr. Sanjiv J. Shah, a cardiologist at Northwestern University, Chicago, who wrote the editorial.

"Patients can progress back and forth across the spectrum" or remain in one stage, he said in an interview. "The syndrome is quite heterogeneous, not only in clinical presentation, but in etiology, and pathophysiology. There is still much to learn about the natural history of HFPEF , how it develops, and the trajectory of patients."

The second editorial described several relatively common HFPEF phenotypes that can appear individually or in combination, including filling limitation, ejection limitation, cardioacceleration, and vasoregulation (Eur. Heart J. 2013;34:1393-5). "Patients with heart failure caused by severe mitral insufficiency or aortic stenosis will clearly behave differently and respond to treatments differently from patients with hypertropic cardiomyopathy, constrictive pericarditis, or high-output heart failure. However, despite this heterogeneity these entities continue currently to be lumped together into the category of HFPEF," Dr. Borlaug wrote in this editorial.

The perils of heterogeneity

In addition to providing a better framework for understanding the causes and consequences of HFPEF, the paradigm of HFPEF as a heterogeneous syndrome also helps explain why the major intervention trials that focused on HFPEF patients, starting a decade ago with studies such as CHARM Preserved (Lancet 2003;362:777-81) and I-PRESERVE (N. Engl. J. Med. 2008;359:2456-67), and continuing through to the recent TOPCAT, have all failed to produce a statistically significant benefit for their primary endpoints.

"Heterogeneity confounds our ability to identify effective treatments. It’s the most likely single explanation" for the neutral HFPEF trials, Dr. Borlaug said. "I don’t think it’s as simple as one thing, but my speculation is that it’s the dominant reason."

"Many trials such as CHARM Preserved and TOPCAT seemed to enroll some patients who in retrospect did not have HFPEF," noted Dr. Gregg C. Fonarow, professor of medicine and associate chief of cardiology at UCLA in Los Angeles. "The key to future clinical trials will be better classification of HFPEF" and better matching of HFPEF patients to their treatment, said Dr. Shah.

"In the neutral megatrials some of the patients did not have HFPEF but had other disorders such as deconditioning and obesity – things that cannot be helped by treatments directed at the heart. We need to identify the patients who will respond to the treatment," said Dr. Burkert M. Pieske, professor and director of cardiology at the Medical University of Graz, Austria.

The TOPCAT (Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist) results reported last November exemplified the problem of patient heterogeneity. The study enrolled more than 3,400 patients in six countries: Argentina, Brazil, Canada, Russia, the Republic of Georgia, and the United States. Retrospective analysis showed that among patients in the placebo group, the primary endpoint of death or heart failure hospitalization during follow-up occurred at a rate of 12.6 events/100 patient-years among the 881 patients treated in the four Western Hemisphere countries, and at a rate of 2.3 events/100 patient-years among the 842 patients treated in Russia or Georgia, a greater than fivefold difference between the two subgroups.

"As a whole, the patients in Russia and Georgia had a lower event rate and were certainly less severely ill than the other patients, but they supposedly still had heart failure. Defining this disorder is difficult, and when a patient has signs and symptoms of heart failure and preserved ejection fraction you may not be certain that heart failure is causing the symptoms. That’s why many people think that we should use another criterion" to define HFPEF in trials, such as elevated serum level of some form of natriuretic peptide, said Dr. Solomon, a TOPCAT coinvestigator. By design, patients could enter TOPCAT either because of a recent heart failure hospitalization, which is how 72% of patients got in, or by having a threshold level of natriuretic peptide, the way the remaining 28% entered the study. Within the subgroup enrolled by natriuretic peptide level, spironolactone treatment had a statistically significant effect in reducing the primary endpoint, while in the other 72% the drug produced no discernable benefit over placebo.

"TOPCAT had funding problems, recruitment problems, and monitoring problems. There was not enough money for good monitoring," said Dr. Pieske, who did not participate in TOPCAT. The patients enrolled in Russia and Georgia were "unbelievably stable," with their 2.3/100 patient-years event rate. "Even if the drug works, it can’t exert an effect if there are no events."

Dr. Pieske ran a much smaller study of spironolactone involving 422 patients with HFPEF, the Aldosterone Receptor Blockade in Diastolic Heart Failure (Aldo-DHF) trial, which showed efficacy for the endpoint of improved diastolic function but not for the co-endpoint of improved exercise capacity (JAMA 2013;309:781-91).

"If you look at TOPCAT and Aldo-DHF together and ask, does spironolactone work, the answer is yes if you select the right patients, those with clear evidence of cardiac functional abnormalities and increased natriuretic peptide levels or documented hospitalization for heart failure. The drug clearly works in these patients, but there were enough inconsistencies in these trials so that this will not get into the treatment guidelines," Dr. Pieske said in an interview.

"Spironolactone provides most benefit in patients who are volume overloaded, have right-heart failure, or both. These are also the patients most likely to have elevated natriuretic peptide. Spironolactone does not seem to benefit as much early-stage HFPEF," noted Dr. Shah.

Several experts cautioned that because the TOPCAT results failed to produce a statistically significant benefit for the study’s primary endpoint, any other conclusions from the results must be made very cautiously, and several also said that they wanted to see the full, published results which had not appeared as of late February. They also noted that many patients with HFPEF, most likely a majority, do not have elevated blood levels of natriuretic peptide.

Where HFPEF management stands now

Evidence-based guidelines for the specific treatment of HFPEF are simple. There aren’t any.

The American Heart Association/American College of Cardiology heart failure management guidelines released last October made no disease-specific treatment recommendations (J. Amer. Coll. Cardiol. 2013;62:e147-e239). They call for controlling hypertension in patients who need intervention by prevailing standards, treatment with a diuretic in patients with volume overload, coronary revascularization if coronary disease is present, and management of atrial fibrillation if that’s present. The European Society of Cardiology’s guidelines issued in 2012 say pretty much the same (Eur. J. Cardiol. 2012;33:1787-1847), except they add this statement: "No treatment has yet been shown, convincingly, to reduce morbidity and mortality in patients with HFPEF."

If nothing else, TOPCAT’s results further solidified spironolactone’s role as a reasonably safe drug for blood pressure lowering in HFPEF patients who are also hypertensive. Hypertension control is a must for patients with HFPEF as it’s believed to significantly contribute to HFPEF in most patients. "Hypertension is the most common underlying cause of HFPEF," said Dr. John R. Teerlink, professor of medicine and director of the heart failure program at the San Francisco Veterans Affairs Medical Center.

When it comes to treating hypertension in HFPEF patients, "unless there is a contraindication, spironolactone should be one of the first drugs to try, but I’m not sure you can say it’s the first choice" based on current evidence, Dr. Borlaug said.

A diuretic is an obvious antihypertensive for HFPEF patients with fluid overload. And there were suggestions of benefit from the ACE inhibitor perindopril in elderly patients in the PEP-CHF trial (Eur. Heart J. 2006;27:2338-45), another large HFPEF-treatment trial that failed to show significant benefit for the primary endpoint with perindopril treatment but had positive results in some secondary endpoints.

Patients "get these drugs to treat their hypertension, but we believe they may also help their heart failure. It’s a belief system. My colleagues and I already use a lot of spironolactone to treat hypertension, and the TOPCAT results won’t change my practice. I’m not comfortable telling people that you should use an aldosterone antagonist because of TOPCAT," Dr. Teerlink said in an interview.

"An angiotensin-converting enzyme inhibitor will probably work; spironolactone probably works if you get the diagnosis right; we don’t know about beta-blockers; and there is some evidence for using digoxin," said Dr. John G.F. Cleland, professor of cardiology at the University of Hull, Kingston-upon-Hull, England. "There must also be good treatment of hypertension, and judicious use of diuretics."

"I’m careful with beta-blockers; I’ve had some patients who felt miserable on them," said Dr. Borlaug. "Some HFPEF patients don’t have a stroke volume, they don’t have diastole, all they have is heart rate, and if you take that away they are left with no cardiac output."

Aside from controlling blood pressure, experts advise good management of other comorbidities such as coronary disease, atrial fibrillation, diabetes, sleep disordered breathing, and renal disease. "Clinicians should make sure that they are not missing severe coronary artery disease, infiltrative cardiomyopathy, constrictive pericarditis, or other causes of HFPEF that have specific treatments," said Dr. Shah. Some drugs have shown promise in early-phase studies – such as ivabradine and neprilysin – but phase III trials are needed. "My advice on how to manage patients with HFPEF is to make every effort to enroll them in a randomized clinical trial," said Dr. Fonarow.

Another key is making the diagnosis, ideally to prevent development of irreversible cardiovascular damage. "HFPEF is difficult to diagnose with certainty unless you do a cardiac catheterization to measure filling pressures, but in patients with early-stage HFPEF, even their invasive hemodynamics look pretty normal," said Dr. Borlaug. Another approach is an exercise echo, which is noninvasive and can identify stress-induced diastolic dysfunction, but the sensitivity and specificity of this approach remains uncertain, he said. And an elevated natriuretic peptide level can help nail the HFPEF diagnosis in some patients but many other patients have levels within the normal range.

To find HFPEF patients, apply a low index of suspicion and look for breathlessness, loss of functional capacity, signs of congestion, lung crackles, echocardiographic signs, pulmonary artery hypertension, and an enlarged left atrium, he suggested.

"Finding patients is a big challenge," said Dr. Pieske. HFPEF patients tend to be elderly, women, and people who are obese who have hypertension and perhaps diabetes. "They complain of dyspnea and fatigue and many physicians think this is just how it is. They will not consider that there is a true diagnosis behind these symptoms and complaints," especially if the patient has preserved left ventricular ejection fraction and a normal natriuretic peptide level." A diastolic stress test using exercise and an echo exam may identify stable patients with early-stage diastolic dysfunction but this requires confirmation, he said.

HFPEF dominates heart failure, lacks good treatment

Experts say the onus on physicians to diagnose and manage HFPEF will grow substantially, since HFPEF is not only highly prevalent but also increasing faster than heart failure with reduced ejection fraction (HFREF). Results from 13 community-based studies published during 1997-2006 showed that HFPEF represented an average of 55% of all heart failure cases, Dr. Carolyn S.P. Lam said last November during a talk at the American Heart Association’s scientific sessions.

Two recent reports of U.S. data documented that the increasing prevalence of HFPEF is outpacing that of HFREF by 1% per year, driven by the aging of the American population (older age is a risk factor for HFPEF) and by the increasing prevalence of comorbidities that contribute to HFPEF. One of the reports, based on U.S. national data collected by the Get With the Guidelines Program, extrapolated that by the end of this decade about 65% of all patients hospitalized for heart failure will have a left ventricular ejection fraction of 40% or greater, the vast majority with HFPEF (Curr. Heart Fail. Rep. 2013;10:401-10). What’s coming is a HFPEF "epidemic," said Dr. Lam, a cardiologist at the National University Heart Centre in Singapore.

The fact that HFPEF is now at least as prevalent as HFREF, that it is about as lethal independent of what comorbidities might contribute, and that it is growing more prevalent than HFREF sharply contrasts with the absence of treatments with proven efficacy. How did that happen?

"Research into HFPEF is decades behind research into HFREF," said Dr. Fonarow. "HFPEF was largely ignored because the prevailing wisdom [20 or so years ago] was that it represented only a small proportion of cases and that outcomes were much better compared with HFREF." Only with registry data and community-based studies run more recently did cardiologists realize that outcomes from HFPEF were as bad as from HFREF. Progress was further hampered by enrollment of patients who did not have HFPEF or a broad spectrum of patients unable to respond equally well to the treatment under study.

"There are large gaps in knowledge of the pathophysiology of HFPEF and little investment has been made to identify potential therapeutic targets. This creates cascading levels of challenges for developing effective treatments despite the massive unmet need," said Dr. Fonarow.

"We don’t know what to treat yet," was Dr. Teerlink’s summation of the HFPEF dilemma.

Dr. Borlaug said that he has been a consultant to GlaxoSmithKline, Merck, Amgen, CardioKinetix, and DC Devices and has received research support from Atcor. Dr. Solomon said he has been a consultant to and received research support from Novartis and more than 10 other drug and device companies. Dr. Pieske said that he has received honoraria from Bayer, Boehringer Ingelheim, Servier, Medtronic, Bristol-Myers Squibb, Menarini, and Novartis and received research support from Bayer and Medtronic. Dr. Shah said that he has been a consultant to Novartis, Bayer Schering, and the Pulmonary Hypertension Association. Dr. Fonarow said that he has been a consultant to Medtronic, Novartis, and Pfizer. Dr. Teerlink said that he has been an adviser to and received research support from Novartis. Dr. Cleland said that he has received honoraria from Novartis and research funding from Pfizer. Dr. Lam said that she has been a consultant to Bayer, Novartis, and DC Devices and has received research support from Boston Scientific, Medtronic, and Vifor Pharma.

[email protected]

On Twitter @mitchelzoler

When the highly-anticipated results from the TOPCAT trial reported last November failed to show a statistically significant efficacy benefit from spironolactone treatment in patients with heart failure and preserved left ventricular ejection fraction for the study’s primary endpoint of death or heart failure hospitalization, the neutral result represented more than another failed treatment trial in this rapidly expanding patient population that remains bereft of treatments with proven efficacy.

TOPCAT’s results also underscored the heterogeneity of patients diagnosed with heart failure with preserved ejection fraction (HFPEF), a signature feature of the syndrome. Heterogeneity is believed to have played a large role in the lack of success in TOPCAT as well as in several other major HFPEF trials.

Although HFPEF is defined as patients who exhibit clear clinical symptoms of heart failure but with a LVEF of at least 45%, and more recently at least 50%, what’s become increasingly clear as neutral trial results accumulate is that HFPEF is a syndrome. Far from a single pathologic entity, it can have different etiologies and present as different subtypes in a spectrum of abnormalities and severities.

HFPEF "is not just a simple disease of diastolic function and a stiff ventricle. It’s a very complicated, highly integrated, multisystem failure of cardiovascular and peripheral reserve," said Dr. Barry A. Borlaug, a cardiologist at the Mayo Clinic in Rochester, Minn. "A lot of us are coming to the conclusion that there are different subphenotypes of HFPEF, each with unique mechanisms and clinical trajectories that probably need unique treatment," he said in an interview.

"This is a very heterogeneous syndrome with no one etiology and a wide range of patients, some with marked ventricular hypertrophy, some with diastolic dysfunction, some who are fairly asymptomatic but become symptomatic with exercise, some who are compensated and stable, and others who are uncompensated. The spectrum of disease is very broad, and it may be many diseases," said Dr. Scott D. Solomon, professor of medicine at Harvard Medical School and director of noninvasive cardiology at Brigham and Women’s Hospital, Boston.

Last year, the heart failure literature featured two editorials spelling out proposed visions of the HFPEF spectrum. One of these proposed three characteristic types of HFPEF patients (J. Am. Coll. Cardiol. 2013;62:1339-42): those with exercise-induced diastolic dysfunction with no symptoms at rest, minimal fluid retention, and never hospitalized for heart failure but with long-standing hypertension and exercise intolerance; patients with volume overload and edema, recently hospitalized heart failure with dyspnea on exertion, and moderately severe heart failure symptoms; and the worst form, patients who have developed pulmonary hypertension and right ventricular failure as a consequence of their HFPEF and now have frequent heart failure hospitalizations. But patients do not necessarily progress from one severity stage to the next, noted Dr. Sanjiv J. Shah, a cardiologist at Northwestern University, Chicago, who wrote the editorial.

"Patients can progress back and forth across the spectrum" or remain in one stage, he said in an interview. "The syndrome is quite heterogeneous, not only in clinical presentation, but in etiology, and pathophysiology. There is still much to learn about the natural history of HFPEF , how it develops, and the trajectory of patients."

The second editorial described several relatively common HFPEF phenotypes that can appear individually or in combination, including filling limitation, ejection limitation, cardioacceleration, and vasoregulation (Eur. Heart J. 2013;34:1393-5). "Patients with heart failure caused by severe mitral insufficiency or aortic stenosis will clearly behave differently and respond to treatments differently from patients with hypertropic cardiomyopathy, constrictive pericarditis, or high-output heart failure. However, despite this heterogeneity these entities continue currently to be lumped together into the category of HFPEF," Dr. Borlaug wrote in this editorial.

The perils of heterogeneity

In addition to providing a better framework for understanding the causes and consequences of HFPEF, the paradigm of HFPEF as a heterogeneous syndrome also helps explain why the major intervention trials that focused on HFPEF patients, starting a decade ago with studies such as CHARM Preserved (Lancet 2003;362:777-81) and I-PRESERVE (N. Engl. J. Med. 2008;359:2456-67), and continuing through to the recent TOPCAT, have all failed to produce a statistically significant benefit for their primary endpoints.

"Heterogeneity confounds our ability to identify effective treatments. It’s the most likely single explanation" for the neutral HFPEF trials, Dr. Borlaug said. "I don’t think it’s as simple as one thing, but my speculation is that it’s the dominant reason."

"Many trials such as CHARM Preserved and TOPCAT seemed to enroll some patients who in retrospect did not have HFPEF," noted Dr. Gregg C. Fonarow, professor of medicine and associate chief of cardiology at UCLA in Los Angeles. "The key to future clinical trials will be better classification of HFPEF" and better matching of HFPEF patients to their treatment, said Dr. Shah.

"In the neutral megatrials some of the patients did not have HFPEF but had other disorders such as deconditioning and obesity – things that cannot be helped by treatments directed at the heart. We need to identify the patients who will respond to the treatment," said Dr. Burkert M. Pieske, professor and director of cardiology at the Medical University of Graz, Austria.

The TOPCAT (Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist) results reported last November exemplified the problem of patient heterogeneity. The study enrolled more than 3,400 patients in six countries: Argentina, Brazil, Canada, Russia, the Republic of Georgia, and the United States. Retrospective analysis showed that among patients in the placebo group, the primary endpoint of death or heart failure hospitalization during follow-up occurred at a rate of 12.6 events/100 patient-years among the 881 patients treated in the four Western Hemisphere countries, and at a rate of 2.3 events/100 patient-years among the 842 patients treated in Russia or Georgia, a greater than fivefold difference between the two subgroups.

"As a whole, the patients in Russia and Georgia had a lower event rate and were certainly less severely ill than the other patients, but they supposedly still had heart failure. Defining this disorder is difficult, and when a patient has signs and symptoms of heart failure and preserved ejection fraction you may not be certain that heart failure is causing the symptoms. That’s why many people think that we should use another criterion" to define HFPEF in trials, such as elevated serum level of some form of natriuretic peptide, said Dr. Solomon, a TOPCAT coinvestigator. By design, patients could enter TOPCAT either because of a recent heart failure hospitalization, which is how 72% of patients got in, or by having a threshold level of natriuretic peptide, the way the remaining 28% entered the study. Within the subgroup enrolled by natriuretic peptide level, spironolactone treatment had a statistically significant effect in reducing the primary endpoint, while in the other 72% the drug produced no discernable benefit over placebo.

"TOPCAT had funding problems, recruitment problems, and monitoring problems. There was not enough money for good monitoring," said Dr. Pieske, who did not participate in TOPCAT. The patients enrolled in Russia and Georgia were "unbelievably stable," with their 2.3/100 patient-years event rate. "Even if the drug works, it can’t exert an effect if there are no events."

Dr. Pieske ran a much smaller study of spironolactone involving 422 patients with HFPEF, the Aldosterone Receptor Blockade in Diastolic Heart Failure (Aldo-DHF) trial, which showed efficacy for the endpoint of improved diastolic function but not for the co-endpoint of improved exercise capacity (JAMA 2013;309:781-91).

"If you look at TOPCAT and Aldo-DHF together and ask, does spironolactone work, the answer is yes if you select the right patients, those with clear evidence of cardiac functional abnormalities and increased natriuretic peptide levels or documented hospitalization for heart failure. The drug clearly works in these patients, but there were enough inconsistencies in these trials so that this will not get into the treatment guidelines," Dr. Pieske said in an interview.

"Spironolactone provides most benefit in patients who are volume overloaded, have right-heart failure, or both. These are also the patients most likely to have elevated natriuretic peptide. Spironolactone does not seem to benefit as much early-stage HFPEF," noted Dr. Shah.

Several experts cautioned that because the TOPCAT results failed to produce a statistically significant benefit for the study’s primary endpoint, any other conclusions from the results must be made very cautiously, and several also said that they wanted to see the full, published results which had not appeared as of late February. They also noted that many patients with HFPEF, most likely a majority, do not have elevated blood levels of natriuretic peptide.

Where HFPEF management stands now

Evidence-based guidelines for the specific treatment of HFPEF are simple. There aren’t any.

The American Heart Association/American College of Cardiology heart failure management guidelines released last October made no disease-specific treatment recommendations (J. Amer. Coll. Cardiol. 2013;62:e147-e239). They call for controlling hypertension in patients who need intervention by prevailing standards, treatment with a diuretic in patients with volume overload, coronary revascularization if coronary disease is present, and management of atrial fibrillation if that’s present. The European Society of Cardiology’s guidelines issued in 2012 say pretty much the same (Eur. J. Cardiol. 2012;33:1787-1847), except they add this statement: "No treatment has yet been shown, convincingly, to reduce morbidity and mortality in patients with HFPEF."

If nothing else, TOPCAT’s results further solidified spironolactone’s role as a reasonably safe drug for blood pressure lowering in HFPEF patients who are also hypertensive. Hypertension control is a must for patients with HFPEF as it’s believed to significantly contribute to HFPEF in most patients. "Hypertension is the most common underlying cause of HFPEF," said Dr. John R. Teerlink, professor of medicine and director of the heart failure program at the San Francisco Veterans Affairs Medical Center.

When it comes to treating hypertension in HFPEF patients, "unless there is a contraindication, spironolactone should be one of the first drugs to try, but I’m not sure you can say it’s the first choice" based on current evidence, Dr. Borlaug said.

A diuretic is an obvious antihypertensive for HFPEF patients with fluid overload. And there were suggestions of benefit from the ACE inhibitor perindopril in elderly patients in the PEP-CHF trial (Eur. Heart J. 2006;27:2338-45), another large HFPEF-treatment trial that failed to show significant benefit for the primary endpoint with perindopril treatment but had positive results in some secondary endpoints.

Patients "get these drugs to treat their hypertension, but we believe they may also help their heart failure. It’s a belief system. My colleagues and I already use a lot of spironolactone to treat hypertension, and the TOPCAT results won’t change my practice. I’m not comfortable telling people that you should use an aldosterone antagonist because of TOPCAT," Dr. Teerlink said in an interview.

"An angiotensin-converting enzyme inhibitor will probably work; spironolactone probably works if you get the diagnosis right; we don’t know about beta-blockers; and there is some evidence for using digoxin," said Dr. John G.F. Cleland, professor of cardiology at the University of Hull, Kingston-upon-Hull, England. "There must also be good treatment of hypertension, and judicious use of diuretics."

"I’m careful with beta-blockers; I’ve had some patients who felt miserable on them," said Dr. Borlaug. "Some HFPEF patients don’t have a stroke volume, they don’t have diastole, all they have is heart rate, and if you take that away they are left with no cardiac output."

Aside from controlling blood pressure, experts advise good management of other comorbidities such as coronary disease, atrial fibrillation, diabetes, sleep disordered breathing, and renal disease. "Clinicians should make sure that they are not missing severe coronary artery disease, infiltrative cardiomyopathy, constrictive pericarditis, or other causes of HFPEF that have specific treatments," said Dr. Shah. Some drugs have shown promise in early-phase studies – such as ivabradine and neprilysin – but phase III trials are needed. "My advice on how to manage patients with HFPEF is to make every effort to enroll them in a randomized clinical trial," said Dr. Fonarow.

Another key is making the diagnosis, ideally to prevent development of irreversible cardiovascular damage. "HFPEF is difficult to diagnose with certainty unless you do a cardiac catheterization to measure filling pressures, but in patients with early-stage HFPEF, even their invasive hemodynamics look pretty normal," said Dr. Borlaug. Another approach is an exercise echo, which is noninvasive and can identify stress-induced diastolic dysfunction, but the sensitivity and specificity of this approach remains uncertain, he said. And an elevated natriuretic peptide level can help nail the HFPEF diagnosis in some patients but many other patients have levels within the normal range.

To find HFPEF patients, apply a low index of suspicion and look for breathlessness, loss of functional capacity, signs of congestion, lung crackles, echocardiographic signs, pulmonary artery hypertension, and an enlarged left atrium, he suggested.

"Finding patients is a big challenge," said Dr. Pieske. HFPEF patients tend to be elderly, women, and people who are obese who have hypertension and perhaps diabetes. "They complain of dyspnea and fatigue and many physicians think this is just how it is. They will not consider that there is a true diagnosis behind these symptoms and complaints," especially if the patient has preserved left ventricular ejection fraction and a normal natriuretic peptide level." A diastolic stress test using exercise and an echo exam may identify stable patients with early-stage diastolic dysfunction but this requires confirmation, he said.

HFPEF dominates heart failure, lacks good treatment

Experts say the onus on physicians to diagnose and manage HFPEF will grow substantially, since HFPEF is not only highly prevalent but also increasing faster than heart failure with reduced ejection fraction (HFREF). Results from 13 community-based studies published during 1997-2006 showed that HFPEF represented an average of 55% of all heart failure cases, Dr. Carolyn S.P. Lam said last November during a talk at the American Heart Association’s scientific sessions.

Two recent reports of U.S. data documented that the increasing prevalence of HFPEF is outpacing that of HFREF by 1% per year, driven by the aging of the American population (older age is a risk factor for HFPEF) and by the increasing prevalence of comorbidities that contribute to HFPEF. One of the reports, based on U.S. national data collected by the Get With the Guidelines Program, extrapolated that by the end of this decade about 65% of all patients hospitalized for heart failure will have a left ventricular ejection fraction of 40% or greater, the vast majority with HFPEF (Curr. Heart Fail. Rep. 2013;10:401-10). What’s coming is a HFPEF "epidemic," said Dr. Lam, a cardiologist at the National University Heart Centre in Singapore.

The fact that HFPEF is now at least as prevalent as HFREF, that it is about as lethal independent of what comorbidities might contribute, and that it is growing more prevalent than HFREF sharply contrasts with the absence of treatments with proven efficacy. How did that happen?

"Research into HFPEF is decades behind research into HFREF," said Dr. Fonarow. "HFPEF was largely ignored because the prevailing wisdom [20 or so years ago] was that it represented only a small proportion of cases and that outcomes were much better compared with HFREF." Only with registry data and community-based studies run more recently did cardiologists realize that outcomes from HFPEF were as bad as from HFREF. Progress was further hampered by enrollment of patients who did not have HFPEF or a broad spectrum of patients unable to respond equally well to the treatment under study.

"There are large gaps in knowledge of the pathophysiology of HFPEF and little investment has been made to identify potential therapeutic targets. This creates cascading levels of challenges for developing effective treatments despite the massive unmet need," said Dr. Fonarow.

"We don’t know what to treat yet," was Dr. Teerlink’s summation of the HFPEF dilemma.

Dr. Borlaug said that he has been a consultant to GlaxoSmithKline, Merck, Amgen, CardioKinetix, and DC Devices and has received research support from Atcor. Dr. Solomon said he has been a consultant to and received research support from Novartis and more than 10 other drug and device companies. Dr. Pieske said that he has received honoraria from Bayer, Boehringer Ingelheim, Servier, Medtronic, Bristol-Myers Squibb, Menarini, and Novartis and received research support from Bayer and Medtronic. Dr. Shah said that he has been a consultant to Novartis, Bayer Schering, and the Pulmonary Hypertension Association. Dr. Fonarow said that he has been a consultant to Medtronic, Novartis, and Pfizer. Dr. Teerlink said that he has been an adviser to and received research support from Novartis. Dr. Cleland said that he has received honoraria from Novartis and research funding from Pfizer. Dr. Lam said that she has been a consultant to Bayer, Novartis, and DC Devices and has received research support from Boston Scientific, Medtronic, and Vifor Pharma.

[email protected]

On Twitter @mitchelzoler