CHAPTER 70: THE DIAGNOSIS AND MANAGEMENT OF CHRONIC HEART FAILURE

The syndrome of heart failure has existed at least as far back as when humans first began to document disease. Clinical texts attributable to Hippocrates describe patients with shortness of breath, edema, and anasarca, in a manner not too varied from contemporary accounts.¹ It has also long been realized that heart failure is not caused by a single disease; rather, it is an amalgamate of several diseases that have unique etiologies, natural histories, and treatments.² The shared feature of this cluster of illnesses is damage to the cardiac tissue. Initially, the heart compensates in various manners to a loss in reserve; however, once there is a critical degree of impairment in its structure and function, a final common pathway emerges that shares similarities in symptoms and findings.

Over the past several decades, dramatic improvements in management of valvular and ischemic heart disease have decreased mortality from these illnesses and consequently led to an increase in the incidence and prevalence of heart failure. This, coupled with the aging of the population, has led to the growth of heart failure prevalence to epidemic proportions. Currently, heart failure affects more than 23 million people globally, of which over five million are Americans.³ Heart failure is a difficult disease to manage and is associated with high rates of morbidity and mortality and high costs, both in terms of direct financial costs as well as indirect cost of patient and caregiver suffering.⁴ Therefore, it is imperative that we recognize and manage the syndrome using the best currently available information and make concerted efforts to find novel ways to further improve its management.

According to the 2013 American College of Cardiology (ACC)/American Heart Association (AHA) heart failure guidelines:

The definition provided by the 2012 European Society of Cardiology (ESC) heart failure guidelines is as follows: “Heart failure can be defined as an abnormality of cardiac structure or function leading to the failure of the heart to deliver oxygen at a rate commensurate with the requirement of the metabolizing tissues, despite normal filling pressures (or only at the expense of increased filling pressures).”⁶

While previously thought of as simply a syndrome of disordered hemodynamics and fluid balance caused by alterations in the structure of the heart, our understanding of heart failure has progressed to a systemic illness that involves complex interplay between myocardial factors, systemic inflammation, renal abnormalities, and neurohormonal dysfunction (Fig. 70–1).^2,7,8 As a result, there is an evolution in our assessment and treatment from a focus solely on examination findings and improving hemodynamics to measuring and modifying the maladaptive molecular processes that contribute to disease progression.⁹ In the absence of universally agreed upon objective biological descriptors of the syndrome, however, we must rely on an assortment of clinical descriptors to diagnose the syndrome that vary in their ability to adequately phenotype heart failure.¹⁰

Several constructs have been created for the purpose of describing heart failure such that there is greater uniformity in its diagnosis and treatment. The most common of these is the New York Heart Association (NYHA) functional classification that was first introduced in 1928 and still persists due to its ease of use and clinical relevance (Table 70–1).¹¹ Furthermore, currently approved therapies are anchored in this classification, as it was used to select patients for almost all randomized clinical trials in heart failure. Patients with NYHA class I have no symptoms attributable to heart disease; those in NYHA classes II, III, or IV have mild, moderate, or severe symptoms, respectively. It is important to note that this classification has numerous limitations: it correlates poorly with objective measures of heart failure, it is not a static measure, and there is significant intraobserver variability in assignment of class.¹²

Another classification introduced by the ACC/AHA is the heart failure staging approach that emphasizes the importance of development and progression of disease (see Table 70–1).⁵ These stages are progressive and associated with prognosis, and interventions can vary based on stage, including modifying risk factors (stage A), treating structural heart disease (stage B), or reducing morbidity and mortality (stages C and D).

Another key classifier of heart failure is left ventricular (LV) ejection fraction (LVEF), principally because this was a fundamental variable used in clinical trials and there is a belief that it captures a distinct phenotypes of heart failure.^13,14,15 Mathematically, it is LV stroke volume divided by the end-diastolic volume. In patients with reduced contraction and emptying of the left ventricle due to systolic dysfunction, output is maintained by the heart ejecting a smaller percentage of a larger end-diastolic volume.⁶ As systolic function decreases, this is generally paralleled by reduction in ejection fraction and greater end-diastolic and end-systolic volumes. In patients with heart failure symptoms in whom reductions in ejection fraction do not occur, there is generally an increase in thickness of the left ventricle suggestive of chronically high filling pressures, as well as increased left atrial size.

A commonly used clinical classification of heart failure relies on bedside physician assessment of perfusion and volume status (see below).¹⁶ Indicators of congestion include history of orthopnea and/or physical exam evidence of jugular venous distention, rales, hepatojugular reflux, ascites, peripheral edema, leftward radiation of the pulmonic heart sound, or a square wave blood pressure response to the Valsalva maneuver. Compromised perfusion is assessed by the presence of a narrow proportional pulse pressure ([systolic – diastolic blood pressure]/systolic blood pressure < 25%), pulsus alternans, symptomatic hypotension (without orthostasis), cool extremities, and/or impaired mentation.¹⁶

Heart failure itself is not synonymous with a certain degree of LV dysfunction, and similar abnormalities can exist across the spectrum of LVEF.^17,18 However, important clinical decisions hinge on a dichotomous classification of ejection fraction because, to date, only randomized clinical trials that enrolled patients with LVEF ≤ 35% to 40% have shown efficacious results (Table 70–2).^19,20 Therefore, one of the most common ways to describe heart failure is as heart failure with reduced ejection fraction (HFrEF) and heart failure with preserved ejection fraction (HFpEF). The prevalence of HFrEF and HFpEF is approximately 50% each.²¹ Of note, the descriptors “systolic” and “diastolic” are often used in place of HFrEF and HFpEF, but from a purist’s perspective, these are incorrect, because systolic dysfunction and diastolic dysfunction coexist in the majority of patients with heart failure, irrespective of ejection fraction.¹⁵

Finally, there are a few other terminologies used. Patients who have had heart failure for some time are often said to have chronic heart failure. If symptoms in such patients are unchanged for some length of time, they are said to have chronic stable heart failure. If a chronic heart failure patient declines, they are said to have worsening heart failure, and in general, if this leads to hospitalization, it is referred to as acute heart failure or acutely decompensated heart failure. Congestive heart failure is a term still used and implies the symptoms of excessive volume retention that are frequent, but by no means the sine qua non, for heart failure.⁶ Terms such as right-sided heart failure and left-sided heart failure attempt to isolate symptoms to a specific ventricle; for example, lower extremity edema and ascites are ascribed to a failing right ventricle, whereas dyspnea and low blood pressure are ascribed to the left.

Signs and Symptoms

It is often challenging to diagnose heart failure, especially in its early stages. Its signs and symptoms can be fairly nonspecific and may not prompt further testing that would lead to a more certain diagnosis.^22,23,24 In the acute setting, heart failure can be mistaken for disorders such as myocardial ischemia, chronic obstructive pulmonary disease, pulmonary embolism, or infections. In the more chronic setting, heart failure can be mistaken for diseases such as depression, asthma, cirrhosis, or hypothyroidism. Therefore, a high degree of mindfulness is required when a clinician encounters a patient with suspected heart failure such that the appropriate set of diagnostic and therapeutic interventions are performed. Once it is confirmed that abnormalities of the heart are driving the patient’s symptoms, the diversity of causes for cardiomyopathy should be considered.²⁵ Another challenge is that criteria for early detection of heart failure in the community have not been well established at this time.

Table 70–3 outlines some of the findings noted in heart failure. As with many medical conditions, key information can be gleaned from a thoughtful history and physical exam.^26,27 Identification of risk factors, such as prior myocardial infarction and a family history, can provide important clues about etiology.²⁸ The duration and triggers of symptoms can identify potential drivers of cardiac dysfunction. Other clues might come from comorbid conditions and treatments. Social history might reveal use of cocaine, methamphetamine, or alcohol, all of which can cause cardiomyopathy.

Physical examination findings are generally ascribed to findings of increased total-body volume (edema, anasarca) or decreased cardiac output (fatigue, exercise intolerance, decreased urine output) and are useful for ruling in heart failure, rather than ruling it out.^16,29 Findings such as elevations in jugular venous pressure and third heart sound are strongly specific for heart failure, but difficult to detect and suffer from lack of reproducibility.^30,31 Furthermore, they are dependent on the patient’s body habitus and can be misleading in the presence of right ventricular dysfunction and/or tricuspid regurgitation. That said, the astute clinician can garner important clues by examination, especially when it is well integrated.³² A proportional pulse pressure [(systolic – diastolic)/systolic)] of < 25% suggests a cardiac index < 2.2 L·min^–1·m^–2.²³ An elevated jugular venous pressure is the most important exam finding to identify congestion.³⁰

The cardiovascular response to the Valsalva maneuver is a simple and highly sensitive bedside test for estimation of volume status and detection of LV systolic dysfunction. With the blood pressure cuff inflated 15 mm Hg over the systolic pressure, the patient is asked to perform a Valsalva maneuver. A normal response is when Korotkoff sounds are audible only at the onset of straining and at release. In patients with heart failure, Korotkoff sounds can be heard throughout the Valsalva maneuver (the square wave response), or there is a lack of reappearance of the sounds after release (the absent overshoot response).²⁹ The third heart sound (or gallop rhythm) is commonly present with tachycardia (but may be challenging to hear) and volume overload and signifies hemodynamic compromise and elevated left-sided filling pressures. Even in the absence of complaints, examination should be performed to determine the presence of ascites, hepatosplenomegaly, or a pulsatile liver. It is not uncommon for patients with advanced heart failure to present with severe right upper quadrant pain and to undergo cholecystectomy when the actual culprit is acute hepatic congestion.³³ Peripheral edema is common in heart failure, but is nonspecific because it can result from a number of other disease states. The finding of cachexia in patients with end-stage heart failure is associated with a particularly poor prognosis.³⁴ Finally, it is important to remember that signs of pulmonary congestion, such as rales and pulmonary edema, are commonly lacking in patients with heart failure, even elevated pulmonary pressures.^26,27,35

Routine Diagnostic Testing

Routine diagnostic testing generally starts with laboratory testing and electrocardiogram (ECG) analysis (Table 70–4).³⁶ Assessments of renal function and potassium can have therapeutic implications. Reversible causes of heart failure such as hypothyroidism, anemia, and hypocalcemia can be ruled out. The ECG can provide information about rhythm and conduction,³⁷ and can help select therapies. Chest x-ray is commonly performed in patients with suspected or new-onset heart failure, but is limited by low sensitivity and specificity,³⁸ and significant LV dysfunction and volume overload can exist in the presence of a normal chest x-ray.³⁹

Biomarkers

Natriuretic peptides, specifically the B-type natriuretic peptide (BNP) and N-terminal pro-BNP (NT-proBNP), have revolutionized the diagnosis and prognostication of heart failure.⁴⁰ Before its activation, BNP is stored as a 108-amino acid polypeptide precursor, proBNP, in secretory granules in both ventricles and, to a lesser extent, in the atria. After proBNP is secreted in response to volume overload and resulting myocardial stretch, it is cleaved to the 76-peptide, biologically inert N-terminal fragment NT-proBNP and the 32-peptide, biologically active hormone BNP. The two fragments are secreted into the plasma in equimolar amounts, and both have been extensively evaluated for use in the management of heart failure. BNP has diuretic, natriuretic, and antihypertensive effects; furthermore, it may provide a protective effect against the detrimental fibrosis and remodeling that occurs in progressive heart failure. Normal natriuretic peptide level in an untreated patient virtually excludes significant cardiac disease, potentially making further testing unnessary.^36,41 Assays for BNP and NT-proBNP are reasonably correlated and either can be used clinically, except when the patient is taking LCZ696 (sacubitril/valsartan; Entresto) where neprilysin inhibition can elevate BNP levels.⁴² Natriuretic peptides have an AHA/ACC Class I recommendation for the diagnosis of acute heart failure and establishment of prognosis in chronic heart failure.⁵ Both BNP and NT-proBNP levels can be elevated from a variety of other causes as well (Table 70–5).⁴³

An abundance of biomarkers can provide unique information about various aspects of heart failure pathophysiology, but only a few are approved for use in the clinical setting (see Fig. 70–1).^9,44 Cardiac troponin⁴⁵ elevations in heart failure are associated with more severe disease and prognosis, but how treatments might be adjusted based on serum elevations remains unclear.⁴⁶ It is reasonable currently to measure levels in the hospital setting to rule out an ischemic trigger and establish prognosis. Soluble ST2 and galectin-3 are predictive of adverse outcomes in chronic heart failure and are additive in their value to natriuretic peptides.⁴⁷ Along with other prognostic factors, they can provide unique and objective information about the patient and are likely be included in future multimarker approaches to heart failure care.⁴⁸ In chronic heart failure, measurement of cardiac troponins, ST2, and galectin-3 has an AHA/ACC Class IIb recommendation for risk stratification.⁴⁹

Transthoracic Echocardiogram

Transthoracic echocardiography plays an important part in the assessment of patients with heart failure (Table 70–6).^13,14,50,51 It can be performed at the bedside, is accurate and safe, and provides key information on chamber size and volumes, systolic and diastolic function, wall thickness, and valve function. A two-dimensional echocardiogram with Doppler during an initial assessment of a patient with suspected heart failure, as well as patients with changes in symptoms, is a Class I recommendation.⁵ The LVEF is measured using several different methods, with the recommended technique being the apical biplane method of disks (the modified Simpson rule). LVEF must be interpreted in its clinical context as it depends on several parameters such as heart rate, preload, afterload, and valvular function, and is not the same as stroke volume. Assessment of diastolic parameters is an important part of the assessment of patients with heart failure. No single echocardiographic parameter can make the diagnosis of diastolic dysfunction in heart failure; a comprehensive inclusion of two-dimensional and Doppler data is required (see Table 70–6). Other key echocardiographic measures in heart failure patients are assessments of right ventricular function, presence of pulmonary hypertension, inferior vena cava diameters, and collapsibility, which are useful in the estimation of right atrial pressure and severity of valve disease.⁵²

Other Noninvasive Imaging

Severe other imaging modalities are available that provide unique information about various aspects of myocardial dysfunction (Table 70–7 and Fig. 70–2).⁵³ Transesophageal echocardiography is useful when chest wall windows are inadequate, in patients with complex valvular or congenital heart disease, and to check for thrombus in the left atrial appendage.⁶ Stress echocardiography is useful to identify ischemia and to gauge if the dysfunctional myocardium is viable.⁵² It is also valuable for the evaluation of aortic stenosis severity or contractile reserve in the setting of reduced LVEF and low transvalvular gradient. Single-photon emission computer tomography (SPECT) or positron emission tomography (PET) may be used to assess ischemia and viability if coronary artery disease is present.

Cardiac magnetic resonance (CMR) can provide almost all the information gleaned from a comprehensive echocardiogram, as well as additional assessments.⁵⁴ It is considered the gold standard for assessment of myocardial volumes, mass, and wall motion. CMR can provide key input in regard to whether an inflammatory or infiltrative condition might be present. Furthermore, it is an extremely helpful tool for the understanding of potential sources of arrhythmias, pericardial disease, and anatomy in patients with congenital heart disease.

Cardiopulmonary Exercise Testing

Cardiopulmonary exercise testing measures a broader range of variables related to cardiorespiratory function, including expiratory ventilation (V̇ E) and pulmonary gas exchange (oxygen uptake [V̇O₂] and carbon dioxide output [V̇CO₂]), and helps quantitatively link metabolic, cardiovascular, and pulmonary responses to exercise.⁵⁵ It objectively correlates the subjective findings in heart failure and helps understand the relative contributions of the heart, lungs, and peripheral muscles to symptoms of dyspnea and fatigue.⁵⁶ The test uses treadmill and bicycle protocols along with gas exchange analysis, and provides key information that can be used to gauge candidacy for advanced therapies such as LV assist devices and transplantation. Cardiopulmonary exercise testing is the gold standard for risk stratification in patients being considered for cardiac transplantation.⁵⁷ Originally, a threshold of peak VO₂ of ≤ 14 mL·kg^–1·min^–1 had been suggested, but in the current era of improved therapies, a lower threshold of 10 to 12 mL·kg^–1·min^–1 has been suggested for consideration of advanced therapies. Some studies have suggested that the VE/VCO₂ slope is a better predictor of outcome than peak VO₂, LVEF, and NYHA class.⁵⁸

Genetic Testing

Familial syndromes are thought to account for up to 20% to 35% of patients with apparent idiopathic dilated cardiomyopathy. Additionally, genetic testing in now readily available to assess for mutations in several other disease-causing genes that can lead to unexplained cardiomyopathies or syndromes with cardiac involvement (Table 70–8).

Invasive Testing

Right heart catheterizations are commonly used to manage heart failure, but there are few data to help guide appropriate settings for when hemodynamic measurements are indicated. Based on expert consensus, invasive monitoring is recommended for patients with respiratory distress or evidence of impaired perfusion in whom clinical assessment falls short. It is also recommended for (1) presumed cardiogenic shock requiring escalating pressor therapy and consideration of mechanical circulatory support; (2) severe clinical decompensation in which therapy is limited by uncertain contributions of elevated filling pressures, hypoperfusion, and vascular tone; (3) apparent dependence on intravenous inotropic infusions; (4) persistent severe symptoms despite adjustment of recommended therapies; or (5) consideration for mechanical circulatory support or transplantation in a patient with advance disease. Routine use of right heart catheterization is not recommended in normotensive patients who are responding appropriately to diuretics and vasodilators.

Based on the premise that providing clinicians with continuous invasive data in heart failure patients could improve outcomes, an implantable device (CardioMEMS; St. Jude Medical, Saint Paul, MN) that allows therapeutic adjustment based on continuous measures of pulmonary artery pressures was tested in a randomized trial of 550 high-risk heart failure patients. The CHAMPION (CardioMEMS Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in NYHA Class III Heart Failure Patients) trial showed that the pulmonary artery pressure–guided treatment arm had a 37% reduction in heart failure hospitalizations compared with the control group.⁵⁹ Based on these results, the device is approved by the US Food and Drug Administration (FDA) for use in patients with heart failure and NYHA class III symptoms. The cost-effectiveness and appropriate subgroup of patients for this device remain unclear.⁶⁰ Left heart catheterizations or coronary angiography should be performed in patients who are candidates for revascularization in whom it is felt that ischemia is contributing to the development or worsening of heart failure.⁶¹ Endomyocardial biopsy should be considered when seeking a specific diagnosis that would influence therapy, such as suspected giant cell myocarditis or cardiac amyloidosis.⁶²

Sodium Intake

Despite the absence of credible data, dietary sodium restriction for the management of heart failure is widely recommended and is borne out of historical beliefs, putative mechanistic data, and observational studies.^63,64 No study to date has evaluated the effects of sodium restriction on outcomes in optimally treated patients,⁶⁵ and there are data showing that reduction in sodium intake might lead to adverse outcomes in heart failure,⁶⁶ underscoring the need for a randomized clinical trial in this area.⁶⁷

Exercise

Both the ESC and the AHA/ACC guidelines give exercise training a Class I recommendation for treatment of HFrEF.^5,6 Several studies have shown that exercise has numerous benefits in heart failure.^68,69 The HF-ACTION (Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training) trial randomized 2331 patients with LVEF ≤ 35% to exercise training versus usual care,⁷⁰ and showed improvements in quality of life, depression, physical fitness, and reduction in risk of heart failure hospitalizations. The Centers for Medicare and Medicaid Services has approved payment for cardiac rehabilitation for heart failure.⁷¹

Pharmacologic Treatments

The goal for pharmacologic treatment of heart failure is to help patients feel better, live longer, and stay out of the hospital. Until the evaluation of vasodilators in 1980s, no options were available to treat heart failure other than diuretics and digoxin.⁷² Subsequently, hydralazine plus isosorbide dinitrate, angiotensin-converting enzyme (ACE) inhibitors,^73,74 β-blockers, and mineralocorticoid receptor antagonists have all been shown to improve outcomes in patients with HFrEF. Two novel therapies, sacubitril/valsartan and ivabradine, were recently approved for use in HFrEF (Fig. 70–3).

Diuretics

Many of the clinical manifestations of heart failure are related to excessive salt and volume retention and several downstream deleterious consequences.⁷⁵ Diuretics can restore salt and water homeostasis, relieve congestion and symptoms, and improve exercise capacity; however, no long-term studies examining the effects of diuretics on morbidity or mortality have been performed to date.⁷⁶ Figure 70–4 shows the tubular sites of action of commonly used diuretics, and Table 70–9 reviews usual dosage recommendations.

Proximal tubular diuretics act primarily by decreasing proximal tubular sodium reabsorption; include osmotic diuretics, such as mannitol, and carbonic anhydrase inhibitors, such as acetazolamide; and are not very effective when used alone. Although 50% to 70% of the glomerular filtrate is reabsorbed iso-osmotically in the proximal tubule, the distal nephron, particularly the ascending limb of Henle loop, has the capacity to increase its rate of sodium reabsorption substantially. Therefore, an increase in glomerular filtration or decrease in proximal tubular reabsorption alone may not be associated with a significant diuresis. Thus, these agents are rarely used.

Loop of Henle diuretics include ethacrynic acid, furosemide, bumetanide, and torsemide. Loop diuretics inhibit the Na⁺/2Cl^–/K⁺ cotransporter of the thick ascending loop of Henle, resulting in decreased sodium and chloride reabsorption.⁷⁷ The potency of these agents depends on renal blood flow and proximal tubular secretion to deliver these agents to their site of action. Bioavailability of furosemide ranges from 40% to 70% of the oral dose; in contrast, the bioavailability of bumetanide and torsemide exceeds 80%.⁷⁸ Despite belief that continuous infusions of loop diuretics were superior to bolus dosing and that higher doses might be harmful, the Diuretic Strategies in Patients with Acute Decompensated Heart Failure (DOSE) trial showed that these therapeutic strategies largely made no difference, and it appeared to be beneficial to use higher doses of diuretics.⁷⁹ Loop diuretics also induce prostaglandin synthesis, resulting in renal and peripheral vascular smooth muscle relaxation and vasodilatation.

Thiazide diuretics include drugs like chlorthalidone and metolazone.⁷⁷ These agents are chemically different. Metolazone is an oral quinazoline that inhibits the Na⁺/Cl^– transporter in the cortical portion of the ascending loop of Henle, prevents maximal dilution of the urine, and decreases the kidneys’ ability to secrete free water. Thiazide diuretics may contribute to hyponatremia. They also increase Ca²⁺ and decrease Mg²⁺ resorption and may be associated with hypercalcemia and hypomagnesemia. Increased delivery of NaCl and H₂O to the collecting duct can trigger potassium secretion, causing hypokalemia. Metolazone and chlorothiazide can be used in conjunction with loop diuretics to overcome diuretic resistance.

The mineralocorticoid receptor antagonists spironolactone and eplerenone are the most relevant potassium-sparing diuretics in heart failure.⁷⁷ Spironolactone is a synthetic steroid that competes for the cytoplasmic aldosterone receptor. It increases the secretion of water and sodium, while decreasing the excretion of potassium, by competing for the aldosterone-sensitive Na⁺/K⁺ channel in the distal tubule of the nephron. Because sodium exchange in the distal tubule is low, potassium-sparing diuretics are generally relatively weak diuretics. As a nonselective aldosterone receptor antagonist, endocrine-related adverse effects (such as gynecomastia) are relatively common with spironolactone. The selective antagonist eplerenone has a similar mechanism of action, but is associated with fewer endocrine-related side effects. These agents are weak diuretics and have a more essential role as neurohormonal modulators, discussed later.

Loop diuretics are the backbone for the treatment of volume overload, and most patients are on maintenance doses to preserve a euvolemic state. Their dosing is usually adjusted in response to disease progression, dietary changes, and concurrent therapies.⁷⁷ Although once-daily use of furosemide is common, more frequent dosing can limit the “rebound” period during which periods of subtherapeutic diuretic concentration in the tubule may lead to a “rebound” in sodium avidity by the kidney. Although furosemide is by far the most common oral loop diuretic, patients with resistance to oral furosemide therapy may benefit from bumetanide or torsemide, which may offer greater efficacy due to increased bioavailability and potency. All the available loop diuretics share similar toxicity, including the risk of ototoxicity in high doses. Ethacrynic acid may be used in patients with sulfa allergy, as it is the only loop diuretic that does not contain sulfa. Careful monitoring and supplementation of electrolytes, particularly potassium and magnesium, are crucial aspects of loop diuretic therapy.

Over time, some patients develop diuretic resistance. Once issues such as noncompliance and concomitant use of nonsteroidal anti-inflammatory drugs have been ruled out, it is important to achieve an adequate dose of loop diuretics by requiring a higher dose in order to achieve the same level of sodium excretion (Fig. 70–5). This “breaking phenomenon” is felt to be a result of hemodynamic changes at the glomerulus, adaptive changes in the distal nephron, and stimulation of the renin-angiotensin and sympathetic nervous systems. Chronically high amounts of sodium delivered to the distal nephron cause hypertrophy of distal convoluted tubule cells, leading to increased sodium reabsorption and negating the overall natriuretic effect of the loop diuretic. This occurs in a more pronounced manner in patients with renal insufficiency. The addition of either metolazone or chlorothiazide can also be helpful.

Finally, there is lingering concern about whether diuretics might cause some harm in heart failure (Figure 70–6).^80,81 For example, diuretics can upregulate neurohormonal activation, potentially worsening renal function, and cause electrolyte abnormalities. However, in the absence of randomized controlled data, it is difficult to separate causal association from confounding by indication.⁸²

Angiotensin-Converting Enzyme Inhibitors/Angiotensin Receptor Blockers

ACE inhibitors reduce the conversion of angiotensin I to angiotensin II and also upregulate bradykinin. The combination of these actions modifies adverse cardiac remodeling.⁸³ Several trials have shown that ACE inhibitors reduce the risk of death and hospitalization in patients with HFrEF, and ACE inhibitors are an ACC/AHA Class I recommendation for patients with current or prior symptoms of heart failure.⁵ Two key randomized trials were the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS) and the Studies of Left Ventricular Dysfunction (SOLVD)-Treatment trial.^73,74 Both trials showed reductions in mortality, with a relative risk reduction of 27% in CONSENSUS and 16% in SOLVD-Treatment. The Assessment of Treatment with Lisinopril and Survival (ATLAS) trial randomized 3164 patients with moderate to severe heart failure to low- versus high-dose lisinopril and showed a 15% relative reduction in risk of death or hospitalization with the latter approach.⁸⁴ Furthermore, in 4228 patients with low LVEF but no symptoms of heart failure, enalapril in the SOLVD-Prevention trial showed a 20% reduction in risk of death or heart failure hospitalization.⁸⁵

ACE inhibitors should be used with caution in patients with low systemic blood pressures, markedly elevated creatinine, bilateral renal artery stenosis, or elevated levels of serum potassium.⁵ There are no differences among the various ACE inhibitors, and there is a Class I recommendation for their use (Table 70–10). Renal function and potassium should be monitored 1 to 2 weeks after initiation and at periodic time points afterward. Up to 20% of patients develop an ACE inhibitor–induced cough. In these patients and others who are intolerant of ACE inhibitors, angiotensin receptor blockers (ARBs) may provide an equally efficacious alternative.⁵ These medications were developed with the rationale that much of the efficacy of ACE inhibition could be gained via blockade of the angiotensin type I receptor (AT1).⁸⁶ In several placebo-controlled studies, ARBs have been shown to yield similar long-term hemodynamic, neurohormonal, and clinical effects as those of ACE inhibition. Two key trials were the Valsartan Heart Failure Trial (Val-HeFT) and the Candesartan in Heart Failure: Assessment of Mortality and Morbidity (CHARM)-Added trial, which randomized patients with mild-to-severe heart failure to placebo versus valsartan and candesartan, respectively.^87,88 Of note, the majority of patients in both trials were already on an ACE inhibitor. Treatment with an ARB improved symptoms and quality of life and reduced the risk of heart failure hospitalization. Further evidence of the sole benefit of ARB versus ACE inhibition came from the CHARM-Alternative trial that randomized ACE inhibitor–intolerant patients to placebo versus candesartan and showed a relative risk reduction of 23% for cardiovascular death and heart failure hospitalizations.⁸⁹ Other trials in acute myocardial infarction have continued to solidify the proposition that ARBs are worthy alternatives to ACE inhibitors in HFrEF.⁶ Table 70–10 displays the recommended and target doses for ARBs in heart failure. Considerations for initiation of ARBs are similar to those of ACE inhibitors.

β-Blockers

The use of β-blocker therapy was once considered counterintuitive in heart failure.^72,90 That belief was overturned by three landmark key clinical trials—Cardiac Insufficiency Bisoprolol Study II (CIBIS II), Carvedilol Prospective Randomized Cumulative Survival (COPERNICUS), and Metoprolol CR/XL Randomized Intervention Trial in Congestive Heart Failure (MERIT-HF)—that randomized more than 9000 patients with mild-to-severe HFrEF to a β-blocker (bisoprolol, carvedilol, and metoprolol, respectively) or placebo.^91,92,93 In these trials, more than 90% of patients were on an ACE inhibitor or ARB. Each of these trials showed a significant reduction in risk of both mortality (including sudden cardiac death) and hospitalization by approximately 30%. Another randomized controlled trial, the Carvedilol or Metoprolol European Trial (COMET), showed that carvedilol increased survival compared with short-acting metoprolol.⁹⁴ It is now felt that β-blockers act by obstructing deleterious long-term effects of sustained sympathetic system activation in heart failure.

β-Blockers should be initiated in all stable heart failure patients except those who have a clear contraindication to their use. Types of β-blockers and target doses are shown in Table 70–10; the ones that have been shown to be efficacious in heart failure are bisoprolol and metoprolol succinate (β₁-antagonists) and carvedilol (α₁-, β₁-, and β₂-antagonists). They must be used cautiously in patients with recent decompensations. If a patient is admitted with acute heart failure, β-blockers can be continued, but in the setting of worsening clinical status, the dose should be lowered or the drug should be discontinued. Clinicians should attempt to achieve target doses suggested by clinical trials. Abrupt withdrawal can lead to clinical decompensation. Adverse effects such as hypotension and bradycardia might necessitate lowering dose or discontinuation; however, symptoms of fatigue or volume overload can be managed without medication discontinuation.

Mineralocorticoid Receptor Antagonists

Aldosterone is synthesized by the adrenal glands to preserve intravascular sodium, potassium, and water homeostasis. Aldosterone binds to mineralocorticoid receptors in the kidney, colon, and sweat glands and induces sodium reabsorption with concomitant potassium excretion. In heart failure, aldosterone levels increase as a response to perceived reductions in intravascular volume and cause several unfavorable effects including adverse remodeling of the heart.⁹⁵

Two aldosterone receptor antagonists, spironolactone and eplerenone, block these deleterious effects and have been shown to have significant morbidity and mortality benefits in symptomatic heart failure. The Randomized Aldactone Evaluation Study (RALES) enrolled 1663 patients with severe heart failure (NYHA class III-IV) and LV systolic dysfunction to receive either spironolactone or placebo in combination with ACE inhibition and diuretics.⁹⁶ Only 11% of the patients were on β-blockers. Spironolactone led to a relative risk reduction in death of 30% and in heart failure hospitalization of 35%. The Eplerenone in Patients with Systolic Heart Failure and Mild Symptoms (EMPHASIS-HF) trial randomized 2737 patients with NYHA class II symptoms and LVEF ≤ 30% (≤ 35% if QRS ≥ 130 ms).⁹⁷ Treatment with eplerenone led to a relative risk reduction of 37% in cardiovascular death or heart failure hospitalization. The Eplerenone Post–Acute Myocardial Infarction Heart Failure Efficacy and Survival Study (EPHESUS) trial enrolled 6632 patients with LV systolic dysfunction (LVEF ≤ 40%) within 2 weeks following an acute myocardial infarction and showed a 15% reduction in risk of all-cause death with eplerenone treatment.⁹⁸ Importantly, the majority of patients in both EPHESUS trials were on β-blockers and ACE inhibitors/ARBs.

It is recommended that spironolactone or eplerenone be used in all patients with HFrEF who are already on an ACE inhibitor (or ARB) and β-blocker, but have suboptimal response to therapy and still have NYHA class II symptoms.^5,6 Table 70–10 shows the recommended target doses. The biggest concern is hyperkalemia, and therefore, these medications should not be used in patients with an estimated glomerular filtration rate < 30 mL/min/1.73 m² (serum creatinine of approximately 2.5 mg/dL in men and 2.0 mg/dL in women) and/or potassium > 5 mEq/L. Those with marginal renal function (glomerular filtration rate 30-50 mL/min/1.73 m²) need every-other-day dosing. Even in the absence of these findings, patients on these medications require rigorous monitoring of their renal function and serum potassium levels. Patients should be advised to eat foods low in potassium and be taken off potassium supplementation. Men on spironolactone can experience gynecomastia or breast pain (~10%), in which case, they should be switched to eplerenone, because the major difference between the medications is selectivity of aldosterone receptor antagonism.

Hydralazine and Isosorbide Dinitrate

The effect of long-acting nitrates plus hydralazine on survival was evaluated in the First Veterans Heart Failure Trial (V-HeFT-I),⁹⁹ in which 642 men with mild-to-moderate heart failure were randomized to receive placebo, prazosin, or the combination of isosorbide dinitrate plus hydralazine. Compared with placebo, the risk reduction for death in the group treated with isosorbide dinitrate plus hydralazine was 36% by 3 years. In contrast, the mortality in the prazosin group was the same as that seen in the placebo-treated patients. The V-HeFT-II trial published in 1991 directly compared the combination of isosorbide dinitrate plus hydralazine with enalapril in patients with predominantly NYHA class II or III heart failure.¹⁰⁰ Overall, 804 men were randomized to one of these two regimens for an average of 2.5 years. The study demonstrated the superiority of the ACE inhibitor in reducing cumulative mortality. The African American Heart Failure Trial (A-HeFT) tested the hypothesis that a fixed-dose combination of isosorbide dinitrate and hydralazine (BiDil) could improve outcomes in African Americans with systolic heart failure. In A-HeFT, 1050 African American patients with NYHA class III or IV heart failure were randomized to the fixed-dose combination or placebo. BiDil significantly improved the primary end point, a composite response including morbidity, mortality, and quality of life. Most impressively, a 43% reduction in all-cause mortality was seen in the treatment group. The results of this study led the US FDA to approve BiDil for African American patients with NYHA class III or IV heart failure and reduced ejection fraction.

Clinicians can use the combination of hydralazine and isosorbide dinitrate for African American patients with HFrEF who remain symptomatic while on ACE inhibitors, β-blockers, and aldosterone antagonists. These therapies are not a substitute for ACE inhibitors or ARBs, but may be used in those who are intolerant of these medications and in non–African American patients as well (see Table 70–10). These therapies may not be well tolerated due to need for frequent dosing and numerous adverse reactions such as headache, dizziness, and gastrointestinal complaints.¹⁰¹

Digoxin

Digitalis has been used to treat heart failure for more than 200 years. The most commonly used preparation is digoxin.¹⁰² Its mechanism of action is via inhibition of the Na-K-ATPase pump in myocardial cells, augmenting intracellular calcium, thus increasing inotropy. Digitalis also has an antiadrenergic effect and upregulates parasympathetic tone.¹⁰³ In the Digitalis Investigation Group (DIG) trial,¹⁰⁴ 6800 patients with LVEF ≤ 45% and NYHA class II to IV symptoms were randomized to 0.25 mg of digoxin versus placebo. The patient population was on concomitant diuretics (82%) and ACE inhibitors (95%). Treatment with digoxin did not reduce all-cause or heart failure mortality, but led to a 28% reduction in heart failure admissions. Digoxin was associated with a trend toward an increased incidence of sudden (presumed arrhythmic) death. A post hoc analysis showed a relationship between mortality and plasma digoxin concentration, with a serum concentration > 1 ng/mL associated with increased risk. Clinical improvements have been supported by smaller trials that suggest a role for digoxin in the improvement or prevention of clinical deterioration.^105,106

Clinicians can consider adding digoxin if patients have persistent symptoms on guideline-recommended therapy. Generally, therapy is started at 0.125 to 0.25 mg/d, with lower or every-other-day dosing in patients with impaired renal function, elderly patients, or frail patients. Serum levels of 0.5 to 0.9 ng/mL should be targeted. The major side effects are arrhythmias (ectopic and reentry rhythms and heart block), gastrointestinal symptoms (anorexia, nausea, and vomiting), and neurologic symptoms (visual disturbances, disorientation, and confusion). Toxicity generally occurs at serum levels of > 2 ng/mL, but may be potentiated at lower levels in the presence of hypokalemia, hypomagnesemia, and hypothyroidism. Several medications interact with digoxin, and interactions should be considered prior to initiation.

LCZ696 (Sacubitril and Valsartan)

A first-in-class of drug that involves dual inhibition of neprilysin and the angiotensin II receptor (angiotensin receptor-neprilysin inhibitor [ARNI]; LCZ696) was approved for use in HFrEF and NYHA class II to IV heart failure based on a trial that randomized 8442 patients to LCZ696 versus enalapril (Prospective Comparison of ARNI with ARB Global Outcomes in Heart Failure with Preserved Ejection Fraction [PARADIGM-HF] trial).¹⁰⁷ The trial was stopped early due to significant reductions in the primary end point after a median duration of follow-up of 27 months. LCZ696 reduced the primary composite end point of cardiovascular death or heart failure hospitalization by 20%. Similar results were observed for cardiovascular death, hospitalization, and all-cause mortality. Hypotension was more common in patients receiving LCZ696, although discontinuation because of hypotension was similar in both arms of the study. Elevations in serum creatinine and potassium and cough were less frequent in those assigned to LCZ696.

LCZ696 is composed of the ARB valsartan and the neprilysin inhibitor sacubitril. Valsartan blocks the AT1 receptor. Sacubitril is converted to the active neprilysin inhibitor LBQ657, which inhibits neprilysin, an enzyme that breaks down atrial natriuretic peptide, BNP, and C-type natriuretic peptide, as well as other vasoactive substances (Fig. 70–7). NT-proBNP is not a substrate for neprilysin and more appropriately reflects heart failure condition among patients on this therapy. The combined beneficial effects of ARBs in heart failure along with the natriuretic, vasodilatory, antisympathetic, and anti–adverse remodeling effects of natriuretic peptides likely account for the benefit with this agent.

Ivabradine

Ivabradine involves inhibition of the I_f node. It slows down the sinus node rate in patients who are in sinus rhythm. It is approved for stable patients with HFrEF who have a resting heart rate of at least 70 bpm on maximally tolerated β-blockers. The Systolic Heart Failure Treatment with the I_f Inhibitor Ivabradine Trial (SHIFT) enrolled 6588 heart failure patients with NYHA functional class II to IV, sinus rhythm with a rate of ≥ 70 bpm, and an LVEF ≤ 35%.¹⁰⁸ Patients were randomized to ivabradine versus placebo, and ivabradine was associated with an 18% reduction in cardiovascular death or heart failure hospitalization. Ivabradine also improved LV function and quality of life. The major side effects were symptomatic bradycardia (5%) and visual disturbances (3%).

Other Therapies

Anticoagulation

Derangements in the clotting cascade are common in heart failure and the triad of prerequisites for thrombus formation (ie, endothelial dysfunction, hypercoagulopathy, and stasis) commonly coexist in heart failure.^109,110 Interruption of the clotting cascade to normalize its function can be a valuable therapeutic target in heart failure. However, large trials testing vitamin K antagonism in HFrEF without atrial fibrillation yielded no evidence of improvement in outcomes. The WASH (Warfarin/Aspirin Study in Heart Failure) and HELAS (Heart Failure Long-Term Anticoagulation Study) trials showed no benefit, but noted an increased rate of major bleeding with warfarin.^111,112 The WATCH (Warfarin and Antiplatelet Therapy in Chronic Heart Failure) trial (n = 1587) showed no difference in patients randomized to aspirin 162 mg versus clopidogrel 75 mg versus warfarin for the primary end point of death, myocardial infarction, or stroke.¹¹³ The WARCEF (Warfarin Versus Aspirin in Reduced Cardiac Ejection Fraction) trial enrolled 2305 patients with LVEF ≤ 35% and sinus rhythm and found no differences in the rates of the primary end point of death, ischemic stroke, or intracerebral hemorrhage for warfarin versus aspirin.¹¹⁴ A reduced risk of ischemic stroke with warfarin was offset by an increased risk of major hemorrhage. Therefore, at this time, anticoagulants other than aspirin are recommended only for heart failure patients with atrial fibrillation.^5,6

Device-Based Therapies

Three types of devices have been shown to reduce morbidity and mortality in select heart failure patients. They include implantable cardioverter-defibrillators (ICDs), cardiac resynchronization therapy (CRT), and left ventricular assist devices (LVADs). Chapter 73 discusses the use of LVADs in advanced heart failure.

Implantable Cardioverter-Defibrillators

Patients with HFrEF are at increased risk for ventricular arrhythmias and sudden cardiac death (SCD). Whereas therapy with neurohormonal blockade can decrease this risk of SCD, an unacceptable degree of residual risk remains. For these patients, an ICD can be life-saving. The benefits of these therapies have been evaluated in several clinical trials, both for primary and secondary prevention, leading to the current recommendations.

For primary prevention, current recommendations are largely based on three key trials: the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT), Multicenter Automatic Defibrillator Implantation Trial II (MADIT II), and the Defibrillators in Non-ischemic Cardiomyopathy Treatment Evaluation (DEFINITE) trial.^115,116,117 SCD-HeFT enrolled 2521 patients with NYHA class II or III heart failure from ischemic or nonischemic etiologies and LVEF ≤ 35% and randomized them to placebo, amiodarone, or standard medical therapy. Amiodarone did not reduce mortality, but ICD was associated with a relative risk reduction of 23% over the follow-up period of approximately 5 years. The MADIT-II trial enrolled 1232 patients with ischemic cardiomyopathy and an LVEF ≤ 30% who were at least 40 days post–myocardial infarction. Follow-up was prematurely discontinued at 20 months due to a significant 31% reduction in all-cause mortality in the ICD group. The DEFINITE trial attempted to evaluate the role of primary prevention ICDs in 458 patients with a nonischemic etiology and LVEF ≤ 35%. A nonsignificant trend toward reduction in mortality was noted.

The following are Class I recommendations for ICD use in primary prevention, according to the ACC/AHA 2013 guidelines: (1) for the primary prevention of SCD to reduce mortality in patient with both ischemic and nonischemic cardiomyopathy and LVEF ≤ 35% with NYHA class II or III symptoms on guideline-recommended medical therapy for at least 3 months, and (2) for the primary prevention of SCD in patients who are at least 40 days post–myocardial infarction and with LVEF ≤ 30% and NYHA class I symptoms or more while receiving guideline-recommended medical therapy. For secondary prevention, ICDs are indicated for survivors of cardiac arrest and in patients with sustained ventricular arrhythmias, irrespective of LVEF. ICD therapy should only be offered to patients who have a reasonable expectation of survival beyond a year and lack of frailty and/or advanced comorbid conditions.

The decision to recommend an ICD is highly complex and should be approached with a high degree of contemplation, because an injudicious insertion can lead to significant suffering. Patients should be advised that the ICD is not a disease-modifying therapy, which is a common misconception. ICD shocks, whether appropriate or not, can lead to anxiety and even posttraumatic stress disorder. Patients with multiple comorbid conditions have far higher rates of device complications, and competing risks of death from noncardiac issues might blunt benefit from an ICD. There are unclear data on the benefit in patients > 75 years old and those with chronic renal disease. An ICD should not be implanted in patients with NYHA class IV symptoms who are not candidates for LVAD or cardiac transplantation. Finally, heart failure is a progressive disease, and patients with functional ICDs will eventually come to a point where the risk/benefit ratio from the device shifts to increased harm, in which case discussions with the patients and families about deactivation are challenging but obligatory.

Cardiac Resynchronization Therapy

Approximately a third of patients with HFrEF have prolongation of the QRS interval on ECG, inferring a degree of mechanical dyssynchrony of the failing heart, which is associated with worse clinical outcomes (Fig. 70–8).¹¹⁸ CRT or biventricular pacing is accomplished through simultaneous pacing of both the left and the right ventricles and can lead to improvements in echocardiographic and hemodynamic parameters, functional status (average of 1-2 mL/kg/min increase in peak VO₂), and clinical outcomes (reduction in hospitalizations and all-cause mortality). CRT is recommended for patients in normal sinus rhythm with LVEF ≤ 35% and a left bundle branch block (LBBB) with QRS of ≥ 120 ms. Less clarity exists for patients with non-LBBB patterns on ECG, those who are not in normal sinus rhythm, or those with wide QRS but ≤ 150 ms.

Figure 70–9 presents an algorithm for the consideration of CRT in heart failure patients. Key among the studies for moderate to severe heart failure were the Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) and Cardiac Resynchronization in Heart Failure Study (CARE-HF) trials, which together randomized 2333 patients with NYHA class III to ambulatory class IV symptoms, LVEF ≤ 35%, and QRS ≥ 120 ms to CRT versus usual medical care.^119,120 Each of these two trials showed that CRT reduced the risk of death from any cause and hospital admission for worsening heart failure (relative risk reduction for death of 24% with a CRT-pacemaker [CRT-P] and of 36% with CRT-defibrillator [CRT-D] in COMPANION and of 36% with CRT-P in CARE-HF). In CARE-HF, the relative risk reduction in HF hospitalization with CRT-P was 52%. These benefits were additional to those gained with conventional treatment, including a diuretic, digoxin, an ACE inhibitor/ARB, a β-blocker, and an aldosterone antagonist. These trials also showed an improvement in symptoms, quality of life, and ventricular function.

Two key randomized trials explored the data for CRT in 3618 patients with mild-to-moderate heart failure: the Multicenter Automatic Defibrillator Implantation Trial with Cardiac Resynchronization Therapy (MADIT-CRT) and Resynchronization/Defibrillation for Ambulatory Heart Failure Trial (RAFT).^121,122 MADIT-CRT randomized NYHA class I (15%) and NYHA class II (85%) patients with LVEF ≤ 30%, a QRS duration ≥ 130 ms, and normal sinus rhythm to optimal medical therapy plus an ICD or optimal medical therapy plus CRT-D. Each of these two trials showed that CRT reduced the risk of death and heart failure hospitalization by 25% to 35%. Only RAFT showed a decrease in all-cause mortality. In both studies, there was significant heterogeneity in treatment effect based on baseline QRS duration, diminishing the strength of recommendation of CRT in patients with mild heart failure to those with a QRS ≥ 150 ms or ≥ 120 ms plus an LBBB pattern. CRT is also indicated in heart failure patients who have an indication for conventional right ventricular pacing that alters cardiac activation in a manner similar to a native LBBB. Finally, because effective CRT requires high rates of biventricular pacing, the benefit of patients with underlying atrial fibrillation is greatest in patients who have undergone concomitant atrioventricular node ablation.^123,124

Clinicians and their patients should have discussions about the pros and cons of CRT-D versus CRT. Similar discussions and recommendations as have been laid out for ICD apply to CRT-D. In addition, patients with CRT with or without ICD will need monitoring after device implantation for adjustment of parameters to maximize resynchronization, as is discussed in Chapter 89. The use of imaging, typically echocardiography, to adjust device parameters in an effort to reduce dyssynchrony is still the subject of research.^125,126 Potential complications include coronary sinus dissection or perforation, lead dislodgement, device infection, pneumothorax, and pocket erosion. High LV lead pacing capture thresholds can cause diaphragmatic pacing from stimulation of the phrenic nerve that runs parallel to the lateral LV free wall.

Approximately half of people with heart failure have HFpEF. The prevalence of HFpEF relative to HFrEF is growing by 10% per decade,¹²⁷ and the hospitalizations caused by HFpEF continue to increase at a high rate.¹²⁸ The increasing HFpEF epidemic is related to increasing rates of common comorbidities that are believed to drive the pathophysiology, including obesity, hypertension, and metabolic syndrome, along with aging of the population.^{127,128,129,130,131,132,133} In striking contrast to HFrEF where numerous drugs and devices improve outcomes, there is no proven effective treatment that prolongs life in HFpEF.¹³⁴ The lack of efficacy of these therapies underscores the fundamental differences between these two phenotypically distinct forms of heart failure,^135,136 as well as our incomplete understanding of the pathophysiology.¹³⁷

Similarities and Differences Between Heart Failure With Preserved Ejection Fraction and Heart Failure With Reduced Ejection Fraction

The traditional cutoff for defining HFpEF has been an ejection fraction of > 50% in the presence of clinical heart failure.¹³⁰ Patients with heart failure and an ejection fraction < 40% are defined as having HFrEF.¹³⁴ Patients with an ejection fraction from 40% to 50% have a borderline decrease in ejection fraction, which has been variably classified as either HFpEF or HFrEF in clinical trials, although emerging evidence supports the idea that this group is more accurately classified as HFrEF.^{138,139,140,141} Although it should be obvious that the dichotomization of heart failure patients according to ejection fraction alone is arbitrary and flawed,^142,143 this method is unlikely to be replaced given the widespread use of echocardiography and the well-characterized risk factor profile, pathophysiology, and outcomes in the two heart failure phenotypes as defined according to this nosology.^130,137,141

There are a number of key differences between HFpEF and HFrEF (Table 70–11), the most obvious of which is ventricular dilatation in HFrEF.¹⁴⁴ Eccentric LV remodeling in HFrEF increases wall stress, resulting in greater natriuretic peptide release.¹⁴⁵ As such, BNP or NT-proBNP levels are typically much lower in people with HFpEF, and among stable outpatients, they are often normal.^146,147 There are also important microscopic differences, including increased cardiomyocyte diameter, higher myofibrillar density, and increased cardiomyocyte stiffness in HFpEF as compared to HFrEF.¹⁴⁴ Patients with HFrEF have depressed contractility. Patients with HFpEF have stiffer ventricles and vasculature, which leads to much greater blood pressure lability with changes in loading associated with dietary indiscretion or physical exercise (Fig. 70–10).^135,148,149 Whereas vasodilator therapies are pushed aggressively in HFrEF despite low arterial pressure, it is unclear whether this practice is beneficial or even could be harmful in HFpEF.¹³⁵

Despite the many differences in ventricular structure and function in HFpEF and HFrEF, there are many similarities. Hemodynamic derangements including elevation in filling pressures, pulmonary hypertension, and impaired cardiac output reserve are common to both, as are symptoms of severe exercise intolerance resulting in reduced aerobic capacity and poor quality of life.^135,150,151

Epidemiology

The relative prevalence of HFpEF increased from 38% to 54% of all heart failure cases in Olmsted County, Minnesota, between 1987 and 2001.¹²⁷ This increase is likely related in part to increased recognition of the syndrome, but it also coincides with rising rates of comorbidities that are associated with the pathogenesis, including hypertension, obesity, atrial fibrillation, and diabetes.^133,152 A recent analysis of US data showed that among incident heart failure cases between 2005 and 2008, 52% of patients had HFpEF, 33% had HFrEF, and 16% had borderline systolic dysfunction (ejection fraction of 40%-50%).¹³¹

Comorbidities are common to both HFpEF and HFrEF, but patients with HFpEF are generally older, more hypertensive, obese, diabetic, and more likely to display atrial fibrillation.^{137,153,154,155,156} Although comorbidities influence ventricular-vascular structure and function in HFpEF, fundamental disease-specific changes drive the disorder, indicating that HFpEF is not merely an amalgamation of comorbidities.¹⁵⁷ Furthermore, morbidity and mortality in patients with HFpEF greatly exceed what is observed in patients with the same comorbid conditions, but who do not manifest heart failure.¹⁵⁸

Risk factors for HFpEF are well established and include hypertension, older age, diastolic dysfunction, kidney disease, anemia, and diabetes.^159,160,161 LV diastolic compliance deteriorates as part of normal aging,^162,163 even in community-dwelling volunteers free of cardiovascular disease.¹⁶⁴ Age-related effects appear to accelerate in the presence of obesity.^164,165,166 Sedentary lifestyle is an independent risk factor for HFpEF, and increasing leisure time activity reduces that risk in a dose-dependent fashion.^167,168 Poor fitness is also associated with concentric ventricular remodeling and diastolic dysfunction,¹⁶⁹ which are common precursors of HFpEF. Prior myocardial infarction is more common in HFrEF than HFpEF,¹⁷⁰ but recent data have shown that coronary artery disease is common in HFpEF and associated with adverse outcome, independent of other predictors (Fig. 70–11).¹⁷¹ Decreases in myocardial microvascular density are also present in HFpEF and may contribute to ischemia.¹⁷²

Hospitalization and rehospitalization risk is similar in HFpEF and HFrEF.^153,154 Although some community-based studies have reported similar mortality in HFpEF and HFrEF,^127,173 others have reported better outcomes in HFpEF.¹⁷⁴ While mode of death is largely cardiovascular in HFpEF patients enrolled in trials,¹⁷⁵ noncardiovascular causes of death are common in HFpEF patients seen in the community.^176,177

Pathophysiology

Diastolic dysfunction was formerly considered to be the sole pathophysiologic driver in HFpEF, but recent studies have shown far more complex abnormalities in LV systolic function, right heart function, the vasculature, endothelium, and periphery (including skeletal muscle) in HFpEF (Fig. 70–12).¹⁷⁸

Left Ventricular Dysfunction

Elevated diastolic filling pressures at rest or with exercise are uniformly present in HFpEF (Fig. 70–13).^146,179 It is important not to equate diastolic dysfunction with HFpEF, as the former is quite prevalent in the general population and is essentially part of normal aging.^164,180 While echocardiographic indices have been established to estimate filling pressures as well as intrinsic chamber stiffness and compliance of the ventricle, these measures have greater variability than invasive assessments.^181,182,183 Indeed, echocardiography does not reveal substantial diastolic dysfunction at rest in up to a third of HFpEF patients.^184,185,186 On the other hand, invasive studies have generally shown diastolic abnormalities to be universal in patients with HFpEF, although in patients with earlier stages of disease, provocative maneuvers such as exercise or saline loading are required to elicit the pathologic changes.^{146,179,187,188,189,190,191,192,193}

The normal enhancement of LV relaxation with increasing heart rates is lost in HFpEF, and this along with abnormalities in chamber compliance may lead to exercise-induced pulmonary venous hypertension and exercise intolerance.^179,194,195 Patients with progressively greater elevation in filling pressures have more advanced HFpEF as evidenced by greater multisystem reserve limitation and increased risk of death.^196,197

Although the ejection fraction is normal or near-normal in HFpEF, LV systolic function is impaired, as shown using tissue Doppler imaging,^198,199 myocardial strain imaging,^200,201 and load-independent indices of chamber and myocardial contractility.¹⁴⁹ Patients with more profound impairments in systolic function and deformation display increased risk of death.^149,201 This may be related to abnormalities in calcium handling,²⁰² β-adrenergic signaling,^203,204 myocardial energetics,^205,206 or tissue perfusion reserve.^171,172

While systolic function at rest is only modestly impaired, a number of studies have also identified dramatic limitations in systolic reserve with stress that importantly contribute to decline in peak cardiac output and exercise intolerance in HFpEF (Fig. 70–12).^{188,189,196,199,205,207,208,209} Systolic reserve limitations beget diastolic limitations, because inability to contract to a smaller LV end-systolic volume in HFpEF limits the amount of elastic recoil favoring diastolic suction of blood from left atrium to ventricle during the following diastole.^{189,204,207,209}

Pulmonary Hypertension and Right Ventricular Dysfunction

Pulmonary hypertension (PH) has been reported to be present in over three-fourths of patients with HFpEF in some studies, initially developing as a form of “passive” PH related to elevated pulmonary capillary wedge pressure (PCWP).^150,210 Chronic elevations in PCWP induce pulmonary arteriolar and venous remodeling, resulting in increased pulmonary vascular resistance.^150,211 The presence and severity of PH in HFpEF are independently associated with increased mortality and represent a potential target for treatment.^210,212

Depending on how it is defined, right ventricular dysfunction is present in roughly one-third of patients with HFpEF.^213,214 Patients with right ventricular dysfunction typically display more advanced heart failure, renal dysfunction, atrial fibrillation, tricuspid regurgitation, and LV dysfunction. The presence of right ventricular dilatation and dysfunction in HFpEF predicts increased risk of death, independent of the severity of PH and other covariates.^213,214,215

Chronotropic Incompetence

Heart rate is the major determinant of cardiac output augmentation with exercise as opposed to stroke volume, with an increase of 200% to 400% at peak exercise.^216,217 Most studies in HFpEF have demonstrated chronotropic incompetence in the majority of studied patients.^{188,189,207,208,218,219,220} This appears to be mediated more by reduced adrenergic sensitivity than decreased sympathetic outflow, since catecholamine levels increase similarly in patients with HFpEF and controls despite marked differences in chronotropic reserve.²⁰⁸ There is additional evidence for autonomic dysregulation in HFpEF. Three studies have shown that the normal decrease in heart rate following exercise is blunted, and one study has observed that arterial baroreflex sensitivity is depressed in HFpEF.^207,208,220 Reduction in heart rate was recently shown to decrease exercise capacity in patients with HFpEF in a small trial.²²¹

Peripheral Abnormalities

According to the Fick principle, decreased peak oxygen (O₂) consumption (VO₂) can be related to limitations in cardiac output reserve, limitations in arterial-venous O₂ content difference, or both.¹⁷⁸ The latter is related to how effectively O₂ is distributed, extracted, and used in the skeletal muscles during exercise.²²² Abnormalities in cardiac output reserve are clearly present in HFpEF.^{188,189,191,207,208,223,224} Recent research has also identified abnormalities in peripheral O₂ extraction and utilization in a significant subset of patients with HFpEF.^192,225,226 The mechanism of inadequate peripheral utilization of O₂ may be related to impaired muscle O₂ extraction, impaired autonomic regulation of blood flow, or problems with macro- or microvascular perfusion to the muscular bed. Recent studies examining histopathologic changes in both skeletal and cardiac muscle have revealed that vascular rarefaction may be present in HFpEF patients, potentially contributing to impaired nutritive blood flow in the heart and periphery.^{172,227,228,229} Exercise training, which improves aerobic capacity and quality of life in HFpEF,^230,231 appears to work mainly through beneficial effects in the periphery in people with HFpEF,²³² although one study did observe evidence of direct cardiac effects.²³³

Endothelial Dysfunction

Impaired flow-mediated vasodilation, a marker of abnormal endothelial function, has been observed in HFpEF, and the degree of dysfunction has been correlated with severity of symptoms, exercise impairment, pulmonary vascular disease, and risk of heart failure hospitalization.^{207,234,235,236} However, not all studies in HFpEF have observed abnormal endothelial function, at least when examined in larger conduit vessels.^237,238,239 Nonetheless, there is substantial evidence that endothelial dysfunction plays a central role in the pathophysiology in a number of patients,²⁴⁰ and numerous clinical trials have been or are being performed targeting this pathway, as discussed later.^{223,241,242,243,244,245,246}

Diagnosis

A patient with a low ejection fraction who is short of breath can be confidently diagnosed with an echocardiogram. In contrast, patients with a normal ejection fraction who are short of breath may have HFpEF, a non-HFpEF cardiac cause of symptoms (eg, valvular heart disease), or a noncardiac etiology. These patients often require a much more thoughtful evaluation to demonstrate objective evidence of elevated filling pressures and/or inadequate cardiac output. This is further complicated by the presence of early HFpEF where there may not be clinically detectable signs of congestion at rest, but there is marked elevation of filling pressures with exercise.¹⁴⁶

The clinical diagnosis of HFpEF can be confidently performed at the bedside purely by physical examination and history, provided that the patient has demonstrable evidence of elevated filling pressures at rest and is in decompensated heart failure. Once a clinical diagnosis of heart failure is made in a patient with normal ejection fraction (≥ 50%), it is important to exclude alternative etiologies, such as pericardial or valvular disease, that would be treated differently.²⁴⁷

Elevated natriuretic peptide levels are strongly supportive of the diagnosis of HFpEF with a normal ejection fraction and have prognostic value.¹³⁴ However, it is important to remember that BNP levels are lower in obesity and that elevated BNP levels can occur with atrial fibrillation (even in the absence of heart failure), renal failure, or right ventricular strain from primary pulmonary arterial hypertension or pulmonary emboli. Importantly, a normal BNP (or NT-proBNP) level does not exclude the diagnosis of HFpEF.^146,147

Echocardiography is essential in the evaluation of patients suspected of having HFpEF. Beyond measurement of ejection fraction, echocardiography provides assessment of LV diastolic function,^215,248 LV filling pressures (high E/e′ ratio),²⁴⁹ systolic function and LV strain,^200,201 left atrial remodeling,^250,251 structural remodeling such as concentric LV hypertrophy, inferior vena caval dilatation (an assessment of right atrial pressure),^251,252 elevation in right ventricular systolic pressures (to estimate pulmonary artery pressure), and qualitative or quantitative impairments in right ventricular function.^210,213,214 In the proper clinical setting, identification of one or more of these echocardiographic abnormalities is highly supportive of the diagnosis of HFpEF (Table 70–12).

The gold standard diagnostic test remains direct measurement of intracardiac pressures by right and left heart catheterization.^253,254 Determination of filling pressures at rest may be inadequate if patients are euvolemic or if symptoms occur only with exercise.^146,188 In these situations, right heart catheterization with exercise stress should be considered. The presence of an elevated mean PCWP (≥ 15 mm Hg at rest or ≥ 25 mm Hg with exercise) is diagnostic of HFpEF.^{130,146,188,191,192}

Treatment

There is no proven effective treatment that improves mortality in HFpEF (Fig. 70–14). Current practice guidelines recommend control of volume overload with diuretics and management of comorbidities.¹³⁴ Although treatment of arterial hypertension clearly reduces the risk of incident HF,^255,256 prior trials testing antihypertensive agents in HFpEF have not been associated with improved outcomes,^257,258,259 and at this point, there is no clear answer as to which agents to use to treat hypertension or how aggressively to treat blood pressure in HFpEF. Recent studies have emphasized the importance of lifestyle interventions such as exercise and weight loss, which improve exercise capacity and quality of life.^231,260,261

Diuretics

Diuretics decrease filling pressures to effectively treat acute decompensated heart failure, regardless of etiology, and are recommended to control fluid overload.^6,134 The CHAMPION trial has provided strong evidence supporting the treatment of elevated filling pressures in order to prevent fluid accumulation and decompensation of heart failure, including HFpEF.^262,263 In this trial, patients with recent heart failure hospitalization and NYHA class III symptoms underwent implantation of a pulmonary artery pressure monitor and were randomized to care based on investigator knowledge of invasive pressures versus standard care. The trial showed a significant reduction in heart failure hospitalizations overall and in ancillary analysis restricted to patients with HFpEF. The number needed to treat to prevent one HFpEF hospitalization over 18 months was 2. Although this was not a trial of diuretics per se, the vast majority of medication changes based on invasive pressure data were changes in diuretic dosing, and these data certainly reinforce the importance of control of congestion with diuretics in people with HFpEF. These data also reveal a new role for implantable hemodynamic monitors to help manage patients with advanced (NYHA class III) HFpEF with recent heart failure hospitalization.^262,263

Renin-Angiotensin-Aldosterone System Inhibition

Three large, placebo-controlled, randomized trials testing ACE inhibitors and ARBs in HFpEF have been conducted to date.^257,258,259 There was a nonsignificant reduction in the composite outcome of cardiovascular death or heart failure hospitalizations with candesartan in the CHARM-Preserved study.²⁵⁷ This trial defined HFpEF using an ejection fraction partition value of 40% and included more men and patients with coronary disease than are generally seen in HFpEF, suggesting that many of the participants had a phenotype resembling HFrEF. The larger Irbesartan in Heart Failure With Preserved Systolic Function (I-PRESERVE) trial comparing irbesartan to placebo in a more typical HFpEF patient population was unequivocally neutral, overall and across all of the prespecified secondary analyses presented.²⁵⁹ The Perindopril in Elderly People With Chronic Heart Failure (PEP-CHF) trial randomized older patients (age ≥ 70 years) to perindopril or placebo and found that perindopril was associated with a trend toward decreased all-cause mortality and hospitalization for heart failure at 1 year, but over the entire 3-year study period, there was no reduction in the primary end point.²⁵⁸ Thus, the available data do not support the use of ACE inhibitors/ARBs as a disease-modifying therapy in HFpEF, although they are clearly safe and still may be effective as antihypertensives in this population.

The mineralocorticoid receptor antagonist spironolactone was recently tested in a randomized controlled trial in HFpEF (Treatment of Preserved Cardiac Function Heart Failure With an Aldosterone Antagonist [TOPCAT]), and there was no significant reduction in the composite primary end point of cardiovascular death, heart failure hospitalization, or aborted cardiac death as compared to placebo.²⁶⁴ There was signal of benefit in higher risk populations, including patients enrolled on the basis of elevated natriuretic peptide levels. A post hoc subgroup analysis identified heterogeneity in response comparing subjects enrolled in the Americas as opposed to Russia and the Republic of Georgia, wherein spironolactone improved the composite end point of the trial in subjects enrolled in the Americas but not in Eastern Europe.²⁶⁵ Notably, the event rate in the latter group was much lower as compared to participants in the Americas, raising concern over whether all of these patients truly had heart failure. In another ancillary study from TOPCAT, it was found that ejection fraction modified the treatment response to spironolactone.¹⁴⁰ Patients with lower ejection fraction (but still meeting the entry criteria) were observed to benefit from spironolactone as compared to placebo, whereas patients with higher ejection fraction (especially > 55%) did not.

A smaller trial observed that spironolactone improved estimated LV filling pressures (E/e′ ratio) in people with early-stage HFpEF, but the co-primary end point of exercise capacity and secondary end points including quality of life were not improved.²⁶⁶ Given the signal of benefit in TOPCAT and the fact that filling pressures may improve, mineralocorticoid antagonists are reasonable choices in patients with HFpEF where azotemia and hyperkalemia are not problematic, particularly if there is additional role for more diuresis.

β-Blockers

β-Blockers are commonly prescribed for patients with diastolic dysfunction based on the notion that slowing the sinus rate will prolong diastole and promote ventricular filling.²⁶⁷ Despite this theoretical benefit, there is little to no evidence that β-blockers improve exercise capacity or clinical outcomes, although data are quite limited in this regard.^{268,269,270,271,272} A recent meta-analysis suggested a possible decrease in all-cause mortality associated with β-blockade in HFpEF,²⁷³ and analysis of data from the Study of Effects of Nebivolol Intervention on Outcomes and Rehospitalization in Seniors With Heart Failure (SENIORS) trial showed no interaction between ejection fraction and the benefits of the β-blocker nebivolol, although the number of patients with “true HFpEF” (ie, ejection fraction ≥ 50%) in this trial was great.²⁶⁹ One concern with β-blockers is the common presence of chronotropic incompetence in people with HFpEF, which will predictably worsen with initiation of a β-blocker.^208,268 A large multicenter trial testing β-blockers in HFpEF is urgently needed to better inform treatment decisions.

Digoxin

The DIG trial included a cohort of 988 patients with heart failure and relatively preserved ejection fraction (> 45%) in sinus rhythm.²⁷⁴ There was a trend to decreased heart failure hospitalizations, but this was negated by an increased number of admissions for unstable angina. There was no effect on mortality. Digoxin is often used for rate control in atrial fibrillation when low blood pressure limits rate-controlling medication, given its minimal hypotensive effects. Retrospective data suggest an increased mortality with digoxin use for atrial fibrillation in heart failure, but prospective trials are lacking.^275,276

Organic Nitrates

Direct nitric oxide (NO) donors such as the organic nitrates increase cellular levels of cyclic guanosine monophosphate (cGMP), a key second messenger that activates kinases that may improve diastolic relaxation and compliance. Nitrates also decrease preload, allowing the ventricle to function at lower volumes, where operating stiffness may be lower as a result of the curvilinear end-diastolic pressure–volume relationship in HFpEF.¹³⁰ However, patients with HFpEF may display excessive reduction in blood pressure and stroke volume with nitrates.¹³⁵

In a placebo-controlled crossover trial, isosorbide mononitrate was recently tested in 110 subjects with HFpEF.²⁴⁵ The primary end point of the trial was chronic activity levels, assessed using hip-worn accelerometry devices, and key secondary end points included exercise capacity (6-minute walk distance), natriuretic peptide levels, and quality of life scores. Compared to placebo, isosorbide mononitrate tended to reduce activity levels, an effect that became more dramatic at higher doses. There was no improvement in exercise capacity and a numerical trend to worsening quality of life and natriuretic peptide levels.²⁴⁵ These data do not support the use of organic nitrates for HFpEF, but they should not be interpreted to indicate that other NO-enhancing therapies will not be effective. Isosorbide mononitrate has been shown to worsen endothelial function in humans,²⁷⁷ which is known to be problematic in HFpEF.²⁰⁷ In addition, the exaggerated blood pressure–lowering effects might have contributed to decreased activity levels.¹³⁵ More targeted interventions that deliver NO at the time of greatest need, such as inorganic nitrite, might be more effective and are being evaluated in this population.²²³

Phosphodiesterase-5 Inhibitors

An alternative method to restore NO signaling is to decrease the degradation of its downstream second messenger (cGMP) using phosphodiesterase-5 inhibitors. In animal models of pressure-overload hypertrophy, sildenafil improves ventricular structure and function dramatically,²⁷⁸ and a small pilot trial reported improvements in PH and right heart function with sildenafil.²⁷⁹ However, in the Phosphdiesterase-5 Inhibition to Improve Clinical Status and Exercise Capacity in Diastolic Heart Failure (RELAX) trial, sildenafil did not improve exercise capacity, neurohormones, ventricular function, or quality of life.²⁴² There may also be untoward effects of sildenafil on ventricular contractility that offset any potential benefits on endothelial and vascular function.²⁸⁰ A more recent trial testing sildenafil in patients with HFpEF and invasively proven PH again showed no improvement in hemodynamics or clinical parameters.²⁸¹

Statins

In a prospective observational study of HFpEF, statin therapy was associated with significantly lower mortality (relative risk of death 0.20) over a median follow-up of 12 months, whereas ACE inhibitor, ARB, β-blocker, and calcium channel blocker therapy had no association with survival.²⁸² However, the Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Miocardico Heart Failure [GISSI-HF] trial, which tested the effects of rosuvastatin on death and death or cardiovascular hospitalization in patients with chronic heart failure, found no benefit in the subgroup of patients included with relatively preserved ejection fraction (> 40%).²⁸³ More recent data from a Swedish registry and meta-analysis of prior trials have suggested that statins may be beneficial in HFpEF.^284,285 As with β-blockers, a large-scale, multicenter, placebo-controlled trial is urgently needed to clarify the role of statins in patients with HFpEF.

Nonpharmacologic Interventions

Relatively small-sized trials have consistently shown that exercise training in HFpEF improves exercise capacity as well as quality of life.^{230,231,232,233,237,260,261} It appears that these benefits are mediated by the periphery and are largely independent of the heart,²³² although one study did observe improvements in diastolic function.²³³ Further study is needed to clarify the best modes and duration of training as well as better methods to optimize adherence and chronically sustain increases in activity levels.²⁶¹

Dietary interventions may also prove effective. In a small, single-center trial, 3 weeks of treatment with a salt-restricted Dietary Approaches to Stop Hypertension (DASH) diet improved diastolic function, arterial stiffness, and ventricular arterial coupling in 13 subjects with HFpEF.²⁸⁶ Kitzman et al²³¹ performed a randomized trial comparing exercise training, caloric restriction, or both to attention control. Both exercise training and caloric restriction improved exercise capacity, and the combination of both interventions was additive, supporting the concept that increases in activity levels and diet may be effective to at least improve morbidity in HFpEF.

Management of Comorbidities

Coronary artery disease is common in patients with HFpEF, seen in roughly 50% to 67% of patients defined angiographically or anatomically from postmortem exams.^171,172 Hwang et al¹⁷¹ have recently shown that the presence of coronary disease is independently associated with increased mortality in HFpEF. Patients undergoing complete revascularization had better maintenance of ventricular function over time and improved survival as compared to patients not receiving complete revascularization (Fig. 70–15).¹⁷¹ Intriguingly, stress testing identified the presence or absence of coronary disease poorly, with very high rates of false-positive and false-negative tests. At present, there are no prospective trial data to guide decisions regarding revascularization for patients with coronary disease and HFpEF, although consensus guidelines recommend revascularization if symptoms of heart failure are deemed to be related to coronary ischemia.¹³⁴

Atrial fibrillation is common in HFpEF, seen in two-thirds of patients at some point during their lifespan.²⁸⁷ Patients with HFpEF and atrial fibrillation are more likely to display right heart dilatation and dysfunction, tricuspid regurgitation, and exercise limitation, and they have a higher risk of death in long-term follow-up, independent of other confounders.^{213,214,250,287,288} There is no clear-cut advantage for rhythm control as opposed to rate control in HFrEF,²⁸⁹ but this has never been tested in patients with HFpEF. Recent studies have shown that aggressive efforts to restore and maintain sinus rhythm in patients with HFpEF and atrial fibrillation can be successful and may be associated with improvements in ventricular function if sinus rhythm can be maintained.²⁹⁰ Further study is required to understand the management of atrial fibrillation in patients with HFpEF.

Given the high prevalence of obesity in HFpEF, diabetes, metabolic syndrome, and sleep apnea are also commonly observed in patients. There are no data evaluating the effects of treating these comorbidities in this population, and treatment is recommended according to their respective guidelines.

Future Directions

Numerous promising therapies are currently being studied in clinical trials for HFpEF. Sacubitril (a combined valsartan/neprilysin inhibitor) showed convincing superiority compared to enalapril in HFrEF,²⁹¹ and in the phase II Prospective Comparison of ARNI with ARB on Management of Heart Failure with Preserved Ejection Fraction (PARAMOUNT) trial, this treatment showed greater improvements in NT-proBNP as compared to ARB alone.²⁴¹ A larger pivotal trial (PARAGON [Efficacy and Safety of LCZ696 Compared with Valsartan, on Morbidity and Mortality in Heart Failure Patients With Preserved Ejection Fraction]) is currently under way (NCT01920711) with intended enrollment of 4300 patients with an ejection fraction > 45%.

A recent double-blind, randomized trial demonstrated that acute administration of inorganic sodium nitrite markedly improved hemodynamic abnormalities developing during exercise, with substantial reduction in exercise LV filling pressures and improved cardiac output reserve.²²³ Another acute crossover study testing inorganic nitrate (precursor to nitrite) delivered as beetroot juice also observed improvements in peak exercise capacity.²⁴⁶ Larger scale studies testing this novel therapy are currently under way. The I_f current blocker ivabradine, which slows the heart rate in patients in sinus rhythm, has variably been shown to improve or worsen exercise capacity in patients with HFpEF.^221,292 A larger scale clinical trial is currently under way to help inform understanding regarding this novel approach. As described earlier, phase III studies testing β-blockers and statins remain a critical unmet need in our understanding regarding the treatment of HFpEF.