Heart failure with preserved ejection fraction: strategies for disease management and emerging therapeutic approaches

ABSTRACT

Approximately 50% of patients with heart failure (HF) have a preserved ejection fraction (HFpEF), and the incidence of HFpEF is increasing relative to HF with reduced ejection fraction (HFrEF). Both types of HF are associated with reduced survival and increased risk for hospitalization. However, in contrast to HFrEF, there are no approved treatments specifically indicated for HFpEF, and current therapy is largely focused on management of symptoms and comorbidities. Diagnosis of HFpEF in the outpatient setting also presents unique challenges compared with HFrEF because of factors including a high burden of comorbidities in HFpEF and difficulties in distinguishing HFpEF from normal aging. Primary care providers (PCPs) play a pivotal role in the delivery of holistic, patient-centric care from diagnosis to management and palliative care. As the prevalence of HF continues to rise in an aging population, PCPs will need to play a greater role in HFpEF care. This article will review HFpEF etiology and pathophysiology, diagnostic workup, and management of symptoms and comorbidities, with a focus on the critical role of PCPs throughout the clinical course of HFpEF.

KEYWORDS:

• Heart failure with preserved ejection fraction
• etiology
• pathophysiology
• diagnosis
• treatment and management

1. Introduction

Heart failure (HF) is a global pandemic affecting approximately 26 million people worldwide, with millions more cases going undiagnosed or misdiagnosed [1]. In the US, the number of people diagnosed with HF is projected to rise to 8 million by 2030, representing a 40% increase from 2015 [2]. HF can be divided into two main types based on left ventricular ejection fraction (LVEF): HF with preserved ejection fraction (HFpEF; LVEF ≥50%) and HF with reduced ejection fraction (HFrEF; LVEF ≤40% [or sometimes <40%]) [3,4]. Hospitalization for HF is often necessary, particularly as many patients with HF are elderly and have multiple comorbidities [5]. Not only is hospitalization associated with advanced disease and poor prognosis in patients with HF [6–8], it is also associated with worse patient quality of life [9] and has significant financial implications for health-care systems [10]. Regardless of hospitalization, survival in patients with HFpEF is compromised, with 5-year mortality rates reported to be >50% [7]. To date, there are no treatments available that prolong survival in patients with HFpEF [3]. Therefore, management of HFpEF currently concentrates on the treatment of symptoms and comorbidities and reducing hospitalization [3,11]. The accompanying review ‘Heart Failure with preserved ejection fraction: disease burden for patients, caregivers, and the health-care system’ outlines the burden of HFpEF in more depth.

The purpose of this article is to review the pathophysiology and treatment of HFpEF, current strategies for disease management, and new and emerging therapeutic approaches, focusing on the role of primary care providers (PCPs).

2. Etiology and pathophysiology of HFpEF

HFpEF etiology and pathophysiology are complex and not fully understood. The diverse etiology includes coronary artery disease, hypertension, valvular heart disease, cardiomyopathy, infiltrative disorders (e.g. amyloidosis), diabetes mellitus (DM), congenital heart disease, pericardial disease, atrial fibrillation, and cardiac toxins (e.g. chemotherapy) [3,4,12–15].

While HFpEF is primarily a disease affecting the heart, it also involves multiple extracardiac systems that can negatively impact cardiac function (Figure 1) [16,17]. HFpEF pathophysiology likely involves a convergence of structural and functional changes to the heart, systemic and pulmonary vascular abnormalities, and extracardiac causes of volume overload such as kidney disease [18].

Figure 1. Schematic of the integrative pathophysiology of HFpEF and the involvement of different extracardiac mechanisms [16,17]. From top left, counterclockwise: lung involvement including primary lung disease leading to PAH, secondary PVH, impaired lung muscle mechanics, and eventual increased pulsatile RV load; abdominal compartment mechanisms including splanchnic circulation (preload), bowel congestion leading to endotoxin translocation and systemic inflammation; skeletal muscle mechanisms including impaired metabolism and peripheral vasodilation; renal mechanisms including passive congestion leading to renal impairment, changes in neurohormonal axis activation, hypertension, abnormal fluid homeostasis, eventual oliguria/renal insufficiency; ventricular-vascular mechanisms including ventricular stiffening leading to systolic and diastolic impairment, diminished systolic reserve, increased cardiac energetic demands and fluid-pressure shift sensitivity. HTN: hypertension; PAH: pulmonary arterial hypertension; PVH: pulmonary venous hypertension; RV: right ventricular. Figure reproduced from Sharma K, Kass DA. Heart failure with preserved ejection fraction: mechanisms, clinical features, and therapies. Circulation Research. 2014;115(1):79–96 [16]. https://www.ahajournals.org/journal/res, with permission from Wolters Kluwer Health

Left ventricular (LV) diastolic dysfunction is considered fundamental to the pathophysiology of HFpEF and is characterized by prolonged isovolumic LV relaxation, slow LV filling, and increased diastolic LV stiffness [19], leading to higher diastolic pressures [13]. The transmission of the higher diastolic pressures to the atrial and pulmonary venous circulations subsequently results in reduced lung compliance and breathlessness [13,15]. Elevated LV diastolic pressures may also increase left atrial (LA) pressure and lead to LA enlargement and structural remodeling, both of which lead to compensatory increase in atrial pump function and electrical remodeling [20,21]. These pathologic changes predispose patients to atrial failure, atrial fibrillation, and subsequently shortness of breath, a hallmark symptom of HFpEF [15,20]. Atrial fibrillation is present in approximately two-thirds of patients with HFpEF and is associated with poor prognosis [22–24]. Additionally, aging and the presence of hypertension, DM, and kidney disease contribute to elevated arterial stiffness in patients with HFpEF [15,16,19]. Arterial stiffening further impairs diastolic relaxation by causing the systemic blood pressure to be more volatile to preload or afterload changes [15,19]. Together, ventricular and arterial stiffening can compound the already prolonged ventricular relaxation and diastolic stiffness and promote pulmonary edema and breathlessness [15,21].

Arterial hypertension is traditionally thought to promote HFpEF via increased myocardial afterload with septal and left ventricular hypertrophy [14,25,26]. An emerging hypothesis suggests that extracardiac comorbidities such as hypertension, obesity, DM, and chronic obstructive pulmonary disease (COPD) may also contribute to HFpEF pathogenesis by inducing a systemic proinflammatory state that in turn causes coronary microvascular endothelial dysfunction and progressive injury [14]. Coronary microvascular inflammation and oxidative stress produce reactive oxygen species such as superoxide anion and hydroxyl radicals that limit nitric oxide bioavailability (secondary to peroxynitrite anion formation) for adjacent cardiomyocytes and subsequently decrease the activity of protein kinase G [14,27]. Reduced protein kinase G promotes cardiomyocyte hypertrophy [14]. Transforming growth factor β promotes the conversion of fibroblasts into myofibroblasts, thereby increasing myocardial collagen deposition [14,28]. Collectively, stiff cardiomyocytes and collagen deposition by fibroblasts contribute to diastolic dysfunction (Figure 2) [14]. This hypothesis represents a new paradigm for hypertension-induced HFpEF, in which microvascular inflammation removes the ‘brake’ on hypertrophic stimuli triggered by myocardial afterload excess [14].

Figure 2. Schematic summary of the role of inflammation on the pathophysiology of HFpEF [14]. Comorbidities, such as obesity, diabetes mellitus, COPD, and hypertension, contribute to a proinflammatory state that results in cardiac hypertrophy and diastolic dysfunction. Similarly, the upregulation of transforming growth factor β ultimately leads to collagen deposition in the interstitial space and diastolic dysfunction. The inflammatory pathway is getting more recognition as one of the pathways contributing to the pathophysiology of HFpEF. cGMP: cyclic guanosine monophosphate; F_passive: resting tension; COPD: chronic obstructive pulmonary disease; IL-6: interleukin 6; NO: nitric oxide; ONOO: peroxynitrite; PKG: protein kinase G; ROS: reactive oxygen species; sGC: soluble guanylate cyclase; TGF-β: transforming growth factor β; TNF-α: tumor necrosis factor α; VCAM: vascular cell adhesion molecule. Reprinted from the Journal of the American College of Cardiology 62(4), Paulus WJ, Tschöpe C, A novel paradigm for heart failure with preserved ejection fraction: comorbidities drive myocardial dysfunction and remodeling through coronary microvascular endothelial inflammation, pg. 263–271, Copyright 2013, with permission from Elsevier

Neurohormonal imbalance may also play a role in the pathophysiology of HFpEF. Patients with preclinical diastolic dysfunction show impaired renal cyclic guanosine monophosphate activation and natriuretic response to volume expansion that can be corrected with B-type natriuretic peptide (BNP) before treatment [29]. The fact that most patients with HFpEF have a history of hypertension is also suggestive of a role for renin-angiotensin-aldosterone system (RAAS) activation in HFpEF pathophysiology [21,30]. While RAAS is a well-known contributor to HFrEF, its role in HFpEF is less clear and it has been suggested that RAAS activation may only be present in a subset of patients [31,32]. In the context of clinical trial data, no improvements in either mortality or HF-related hospitalizations following treatment with traditional RAAS inhibitors (i.e. angiotensin-converting enzyme inhibitors [ACEI] or angiotensin-receptor blockers [ARB]) have been shown compared with placebo in patients with HFpEF [33].

3. Recognition of HFpEF

The ability to recognize symptoms of HFpEF is important to ensure timely management and patient support, ideally before symptoms escalate and hospitalization is necessary, as hospitalization signifies worsening HF and is associated with progressive decline in cardiac function and quality of life [5,6,9,34].

3.1. Signs and symptoms of HFpEF

Signs and symptoms of HFpEF are similar to those of HFrEF; a comprehensive list of signs and symptoms typical of HF can be found in Table 1 [4]. Patients with HFpEF typically present with dyspnea and exercise intolerance [35]. Therefore, the following symptoms and signs should elevate suspicion of HF: breathlessness, orthopnea, paroxysmal nocturnal dyspnea, limited exercise tolerance, or requirement for more time to recover following exercise, fatigue/tiredness/weakness, and/or ankle swelling [4,13,36]. In patients receiving diuretics for underlying conditions, symptoms of HFpEF may occur in the absence of signs of volume overload, such as ankle edema, elevated jugular venous pressure, and pulmonary rales [36].

Table 1. Signs and symptoms of HF [4]

Symptoms	Signs
Typical	More specific
Breathlessness Orthopnea Paroxysmal nocturnal dyspnea Reduced exercise tolerance Fatigue, tiredness, increased time to recover after exercise Ankle swelling	Elevated jugular venous pressure Hepatojugular reflux Third heart sound (gallop rhythm) Laterally displaced apical impulse
Less typical	Less specific
Nocturnal cough Wheezing Bloated feeling Loss of appetite Confusion (especially in the elderly) Depression Palpitations Dizziness Syncope Bendopnea	Weight gain (>2 kg/week) Weight loss (in advanced HF) Tissue wasting (cachexia) Cardiac murmur Peripheral edema (ankle, sacral, scrotal) Pulmonary crepitations Reduced air entry and dullness to percussion at lung bases (pleural effusion) Tachycardia Irregular pulse Tachypnea Cheyne–Stokes respiration Hepatomegaly Ascites Cold extremities Oliguria Narrow pulse pressure

HF: heart failure

Ponikowski P, Voors AA, Anker SD, et al. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure: The Task Force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC). Developed with the special contribution of the Heart Failure Association (HFA) of the ESC. European Journal of Heart Failure, 2016;18(8):891–975, by permission of Oxford University Press and the Heart Failure Association of the European Society of Cardiology.

The prevalence of noncardiac comorbidities tends to be high in patients with HFpEF. In a retrospective analysis of a large HF outpatient cohort (>9000 patients), patients with HFpEF had a higher mean number of noncardiac comorbidities per patient (4.0 vs 3.5, respectively; P < 0.001) and were more likely to have DM, hypertension, anemia, COPD, and obesity than patients with HFrEF [37]. In one series of patients with HFpEF, 60% had sleep disordered breathing and 62% had obstructive sleep apnea [38]. Obstructive sleep apnea is associated with a 2.2-fold higher risk of hospitalization for HFpEF [39].

Wheezing is another less common, but equally important, symptom: up to 35% of elderly patients with acute HF present with wheezing [36,40]. Wheezing may also be evident in ambulatory HF and is likely caused by compression outside the bronchioles stemming from fluid in the lungs [36]. Although the underlying pathophysiology of wheezing is different in HF and pulmonary diseases (i.e. asthma, COPD), in which compression originates around or inside the bronchioles, the pattern of wheezing will be similar and should not be automatically attributed to pulmonary disease [36].

3.2. Difficulties in diagnosing HFpEF

Diagnosing HFpEF is considered more challenging than diagnosing HFrEF. Numerous factors contribute to diagnostic difficulty in HFpEF, including gaps in our understanding of its pathophysiology, lack of consensus on diagnostic criteria, patient heterogeneity, the presence of noncardiac comorbidities, and a disproportionate focus on HFrEF because of availability of evidence-based therapies [37,41].

The majority of missed HF cases are in patients with HFpEF [42]. A diagnosis of HFpEF may be overlooked, as a normal ejection fraction (EF) and lack of signs of volume overload may redirect suspicion to other causes of dyspnea, such as pulmonary disease or obesity [36]. Active case-finding strategies may be helpful to identify patients with HFpEF in some high-risk groups (i.e. those with type 2 DM or COPD) [36]. In a case-finding study in older patients (aged >60 years) not suspected of having HF but at risk due to type 2 DM, 28% were subsequently diagnosed with HF, 83% of whom had HFpEF, and only 17% of whom had HFrEF [43].

As HFpEF preferentially affects the elderly, symptoms may be confused with changes characteristic of normal aging [44], including changes associated with normal aging such as skeletal muscle mass loss [45], chronic inflammation [46], reduced endothelium-dependent vasodilation [47], as well as increased LV stiffness during systole and diastole [48]. Atypical presentation or presentation with comorbidities can also lead to difficulties in obtaining a differential diagnosis. For instance, the diagnosis of HF in patients with COPD is particularly challenging, as COPD has numerous overlapping signs and symptoms with HF (e.g. dyspnea on exertion, cough, paroxysmal nocturnal dyspnea) and can complicate interpretation of diagnostic tests (chest X-ray, echocardiography, serum levels of natriuretic peptides [NPs]) [49,50].

Misdiagnosis of HFpEF undermines effective management and clinical trial recruitment [41]. In addition, research in other disease states has indicated that the lack of an accurate diagnosis can be difficult and frustrating for patients [51], and patients are relieved once diagnosed [52]. While there are some simplified tools to help guide diagnosis of HF and HFpEF (i.e. the H₂FPEF Score [53]), no single test can establish a diagnosis of HF [3]. HF remains a primarily clinical diagnosis informed by a thorough patient history and physical examination [3].

3.3. Considerations for identification of patients with HFpEF

Early detection of the disease requires recognition of tell-tale signs and symptoms deduced from appropriate history taking and physical examination, which are then validated upon further investigation. As signs and symptoms overlap between HFpEF and HFrEF, additional diagnostic workup including/excluding noncardiac causes of symptoms suggestive of HF is crucial to differentiate these two phenotypes [3,4]. The 2013 American College of Cardiology Foundation/American Heart Association (ACCF/AHA) guideline for the management of HF recommends the following components as part of a thorough patient history: family history, duration of illness, severity, and triggers of dyspnea and fatigue, presence of chest pain, exercise capacity, weight changes, heart rhythm abnormalities (palpitations, ICD shocks, etc.), symptoms of transient ischemic attack or thromboembolism, presence of peripheral edema or ascites, sleep problems, history of hospitalization for HF, diet, and medications [3]. Another important aspect of the patient history is the presence of comorbidities, including ischemic heart disease, type 2 DM, and hypertension [36].

Physical examination enables assessment of volume status and perfusion and assists in determining disease severity [3]. Key components of the HF physical examination include physical appearance (body mass index, peripheral edema, temperature of lower extremities), vital signs (blood pressure [including orthostatic changes], pulse, heart rate, respiratory rate), jugular venous pressure, and cardiopulmonary examination (heart sounds and murmurs, point of maximal impulse, right ventricular heave, rales, pleural effusion) [3].

If signs and symptoms are consistent with HF, additional diagnostic investigations such as electrocardiogram, chest X-ray, blood tests (including assessment of BNP), and echocardiogram are required to confirm or exclude HF [3,4]. BNP levels may be lower in obese patients, which may diminish the sensitivity of the test [11]. Echocardiography is generally considered the most useful test for diagnosis of HF and is a key component of the HFpEF diagnostic workup [3,4]. Echocardiography enables assessment of LVEF, key structural alterations (LV and septal hypertrophy and increased left atrial volume index or LV mass index), longitudinal strain, and tricuspid regurgitation [3,4].

NP levels are useful in identifying patients requiring further investigation, such as echocardiography, and can be used to exclude the diagnosis of HF [4]. In the ambulatory setting, a BNP level <35 pg/mL or an N-terminal prohormone of BNP (NT-proBNP) level <125 pg/mL excludes HF [4], whereas in the acute setting, higher levels (BNP <100 pg/mL and NT-proBNP <300 pg/mL) are used to exclude diagnosis (Table 2) [4,54–60]. While these cutoffs are similar between patients with HFrEF and HFpEF, NP levels tend to be lower in patients with HFpEF than in patients with HFrEF [4,61–64]. Even when patients with HFpEF are symptomatic, the rise in levels may not be as pronounced as that in patients with HFrEF [65]. The 2017 update to the 2013 ACCF/AHA guidelines recommends measurement of BNP to help support or exclude a diagnosis of HF and to establish prognosis and disease severity [11].

Table 2. Physiology and clinical utility of natriuretic peptides [4,54–60]

Natriuretic peptides	Physiologic function	Upper limits of cutoff to exclude HF^a (pg/mL) [4]		Clinical utility
		Ambulatory setting (without acute symptoms)	Acute setting
BNP	The release of BNP and NT-proBNP, cleaved fragments of proBNP, is enhanced when myocardial wall stress is detected. To compensate for the ventricular volume/pressure overload, these biomarkers are released to promote natriuresis, diuresis, inhibition of RAAS and SNS, ultimately leading to vasodilatory effects [54,55]	<35 HF unlikely ≥35 warrants additional investigation	<100 HF unlikely ≥100 warrants additional investigation	Useful as one of the initial diagnostic tests to stratify patients who may need further workup, especially in the ambulatory setting in which echocardiography is not readily available. Elevated NPs may identify patients who need additional cardiac investigation [4] Useful for exclusion of HF diagnosis: ○ When the level of NP is below the cutoff threshold, it is very likely that the diagnosis of HF can be ruled out [4,54] ● Predicts incident HFpEF in a community-based cohort ○ Over a mean follow-up of 12 years, among 22,756 participants who were prospectively followed, BNP/NT-proBNP was found to be associated with 27% increase of incident HFpEF (HR, 1.27; 95% CI, 1.16, 1.40; P < 0.001) [60] ○ Coupled with clinical, echocardiographic, and biochemical risk factors, NPs may be part of the prediction model for identifying patients at risk for HFpEF [59] ● Prognostic value: ○ For every twofold increase in BNP, the composite risk of all-cause mortality and hospitalization for HF increases by approximately 1.3 times regardless of EF [57] ○ NP at admission or discharge is useful to establish prognosis in patients with ADHF ● Higher BNP levels at admission are associated with increased in-hospital mortality (HFpEF in-hospital mortality rates: 1.6% if BNP <303 pg/mL vs 4.9% if BNP ≥1070 pg/mL) [58] ● Compared with >60% reduction of NT-proBNP at discharge, attaining only 30%–60% reduction or ≤30% reduction is, respectively, associated with ~3 times and ~5 times higher risk of 6-month all-cause mortality in patients with HFpEF [56]
NT-proBNP		<125 HF unlikely ≥125 warrants additional investigation	<300 HF unlikely ≥300 warrants additional investigation

ADHF: acute decompensated heart failure; BNP: B-type natriuretic peptide; EF: ejection fraction; HF: heart failure; HFpEF: heart failure with preserved ejection fraction; HR: hazard ratio; NP: natriuretic peptide; NT-proBNP: N-terminal prohormone of B-type natriuretic peptide; RAAS: renin-angiotensin-aldosterone system; SNS: sympathetic nervous system.

^aLower cutoff values may be used in HFpEF than in HFrEF.

Current guidelines from the European Society of Cardiology (ESC) recommend that the following criteria are met before making a diagnosis of HFpEF: presence of symptoms and/or signs of HF, a preserved EF (LVEF ≥50%), elevated levels of natriuretic peptides (BNP >35 pg/mL and/or NT-proBNP level >125 pg/mL), and objective evidence of at least one other cardiac functional or structural alteration underlying HF [4]. Additionally, an algorithm involving a stepwise 4-stage approach is recommended (Figure 3). The first and second steps involve initial clinical assessment followed by a diagnostic workup (comprehensive echocardiography and BNP/NT-proBNP). In the case of diagnostic uncertainty, these initial steps should be followed by invasive or noninvasive stress testing. Exercise echocardiography and invasive hemodynamic tests at rest and exercise can help with confirming a diagnosis for HFpEF. If appropriate, tests such as ergometry may be undertaken to identify any specific underlying etiology [66].

Figure 3. Flowchart of a diagnostic algorithm for diagnosis of HFpEF. ECG: electrocardiogram. [66], by permission of Oxford University Press and the European Society of Cardiology

Failure to correctly identify and categorize patients with HFpEF may have significant implications for both patient care and the wider health-care system because of the ineffectiveness of guideline-directed HFrEF therapies in HFpEF [65]. A focus on the early recognition and the correct diagnosis of HFpEF in primary care could therefore reduce the time wasted on ineffective treatments for patients with HFpEF and could also be used to identify patients who may benefit from enrollment in clinical trials [65].

4. Management of patients with HFpEF

4.1. The role of PCPs in the management of HFpEF

PCPs play a crucial role in the management of HFpEF. Early diagnosis is crucial to ensure optimal management, and PCPs are optimally positioned to provide patients with ongoing holistic patient-centric care [67]. A PCP providing care to 2000 adult patients may have more than 40 patients with HF [68], half of whom may have HFpEF given estimates that ~50% of patients with HF have HFpEF [3,35]. As the prevalence of HF is increasing at a rate faster than the availability of cardiologists, it is inevitable that PCPs will need to play a greater role in managing these patients [69]. PCPs function at the center of patient care throughout the disease trajectory of HF, ensuring that patients receive comprehensive, coordinated care across specialties [67,70]. They will make appropriate referrals, provide specialists with the necessary patient information and medical history, and work together with specialists. They also manage risk factors and risk stratification of patients with existing comorbidities; manage stable HF; and in instances of suspected acute HF, initiate nitrates and loop diuretics; and, when indicated, refer to an HF specialist for advanced management and/or end-of-life care support [36,71].

PCPs may also collaborate with patients and their caregivers to assess and address any frailty issues. Patients who are diagnosed with HFpEF are generally older compared with patients with HFrEF [15], and frailty is a common issue experienced by elderly patients with HF [72,73]. Frailty is defined as having low grip strength, low energy, slow walking speed, and low physical activity [72] and has been associated with a 92% and 65% increased likelihood of emergency department visits and hospitalizations [73], respectively, in frail patients with HF compared with those who were not frail. Frailty may be negatively impacted by comorbidities, which can impair patients’ ability to care for themselves [72]. As such, greater collaboration among PCPs, patients, and their caregivers may facilitate assessment of their quality-of-life (QoL) issues, end-of-life values, preferences in decision-making, management of comorbidities, and provide educational support [71,74].

4.2. Use of pharmacologic treatments in the management of HFpEF

Currently, the main goals of HFpEF treatment are to improve functional capacity and QoL, through the management of symptoms and comorbidities, and to avoid acute exacerbations and related hospital admissions [75]. Much of HFpEF management is centered around relieving symptoms due to volume overload. The only guideline-recommended treatment for HFpEF is loop diuretics in congested patients to alleviate signs and symptoms [3,4,11]. The use of a hemodynamic monitoring system should be considered in symptomatic patients with previous HF hospitalization [4]. The implantable pulmonary artery pressure monitoring system CardioMEMS™ (Abbott) was shown to be effective in reducing hospitalizations in patients with HFrEF and HFpEF in the CHAMPION (CardioMEMS Heart Sensor Allows Monitoring of Pressure to Improve Outcomes in NYHA Functional Class III Heart Failure Patients) trial, with a 46% reduction in HF-related hospitalizations in patients with LVEF ≥40% in the treatment group vs the control group (N = 119; incidence rate ratio 0.54; 95% confidence interval [CI] 0.38, 0.70; P < 0.0001) [76,77]. This reduction was reflective of more changes in diuretic and vasodilator therapies in the treatment group in response to pulmonary artery pressure information.

Pharmacologic treatments that have shown improvements in mortality and rates of hospitalization in patients with HFrEF have failed to show these benefits in clinical trials for HFpEF [33]. This includes treatment with ACEIs and ARBs. Even in instances in which nonsignificant trends toward improvements were observed, such as in the CHARM-Preserved study comparing candesartan versus placebo, results need to be interpreted with caution, as a large proportion of the patients had what would now be classified as borderline HFpEF or HFmrEF (EF 40% to 49%), and baseline clinical characteristics likely differed from those observed in clinical practice [33,78].

Even the efficacy of β-blockers is largely unresolved, with conflicting evidence for their use, although they are commonly prescribed for this patient population [33]. A cumulative meta-analysis of observational studies evaluating the effect of statins on mortality in patients with HFpEF has shown a 40% reduction in the relative risk of all-cause mortality (P < 0.001) [79], but data from prospective randomized controlled trials are still needed to confirm this effect. The mineralocorticoid-receptor antagonist spironolactone has provided some evidence to suggest improvements in the rates of hospitalization for HF (HHF) in patients with HFpEF, but this did not translate into reductions in mortality [80]. The TOPCAT (Treatment of Preserved Cardiac Function Heart Failure with an Aldosterone Antagonist) trial investigated the effects of spironolactone on the primary composite end point of cardiovascular (CV) death, aborted cardiac arrest, or HHF [80]. Significant reduction in one of the components of the primary composite outcome, HHF, was observed (unadjusted hazard ratio [HR], 0.83: 12% vs 14% of patients in spironolactone vs placebo treatment arms, respectively; P = 0.04) but was not accompanied by a significant reduction of the primary composite end point (unadjusted HR, 0.89; P = 0.14) [80]. Calcium channel blockers (CCBs) have shown some benefit in improving exercise capacity and diastolic function in small studies in HFpEF patients [81,82]. However, a retrospective analysis of propensity-matched study cohorts of older hospitalized HFpEF patients found no association with new discharge prescription for CCBs and mortality or HF hospitalization [83].

Thus, patients with HFpEF need a different approach to management than patients with HFrEF. Research into new treatment strategies in patients with HFpEF is currently focused on a number of emerging agents, including combination therapy with an ARB and a neprilysin inhibitor, in the form of an angiotensin receptor–neprilysin inhibitor (ARNI; e.g. sacubitril/valsartan) [84–88] and the sodium glucose transport protein inhibitor (SGLT2i) dapagliflozin [89] (Table 3) [84,85,87–91]. Neprilysin is an endopeptidase that hydrolyzes a variety of peptide hormones, including those that regulate natriuresis and diuresis [92]. The inhibition of neprilysin with sacubitril reduces the proteolytic destruction of atrial NPs and BNPs in blood, thereby potentiating their beneficial effects on natriuresis, diuresis, and vasodilatation [90,93]. The PARAMOUNT (Prospective comparison of ARNI with ARB on management of heart failure with preserved ejection fraction) trial confirmed that sacubitril/valsartan induces favorable changes in a number of HFpEF parameters, including significant reductions in left atrial size, New York Heart Association class, and serum NT-proBNP [84,90]. At 12 weeks, regardless of systolic blood pressure changes, NT-proBNP was significantly reduced for patients taking sacubitril/valsartan compared with valsartan alone (ratio for change, 0.76; P = 0.008) [84], without significantly affecting renal function (baseline to 12 weeks of change in estimated glomerular filtration rate, −0.2 ± 14.9 vs −2.0 ± 11.9 mL/min/1.73 m²; P = 0.29, sacubitril/valsartan vs valsartan, respectively) [85]. Morbidity and mortality outcomes with sacubitril/valsartan have recently been published in the PARAGON-HF (Prospective Comparison of ARNI with ARB Global Outcomes in HF with Preserved Ejection Fraction) trial [86–88]. The primary composite outcome of total (first or recurrent) HHF and CV death did not differ significantly (although there was a trend for benefit) between the sacubitril/valsartan and valsartan groups (primary events, 894 vs 1009, respectively; rate ratio [RR], 0.87; P = 0.06) [88]. A prespecified secondary analysis showed that sacubitril/valsartan may have sex-specific benefits. When compared with valsartan, sacubitril/valsartan led to a greater reduction of the primary composite end point of total HHF and CV death in women than in men (adjusted RR, 0.73 in women [95% CI, 0.60, 0.90] vs 1.02 in men [95% CI, 0.83, 1.24]; P interaction = 0.0225), with the greatest benefit derived from the reduction of HHF (adjusted RR, 0.67 in women [95% CI, 0.54, 0.84] vs 1.06 in men [95% CI, 0.84, 1.34]; P interaction = 0.0048) [94]. This subgroup analysis is important in the setting of HFpEF, since most patients diagnosed with HFpEF are women [3,7,94]. Additionally, in a prespecified pooled analysis of the PARAGON-HF (LVEF defined as ≥45%) and PARADIGM-HF (LVEF defined as ≤40%) studies, greater benefit was observed in those with mildly reduced LVEF [95].

Table 3. Characteristics of studies investigating treatments for HFpEF

Study	Purpose	Study design/ methodology	Treatment arms	Key outcomes measured	Key results
PARAMOUNT Jhund PS, et al. 2014 Voors AA, et al. 2015 Solomon SD, et al. 2012 [84,85,90]	Prospective comparison of the efficacy and safety of LCZ696 (sacubitril/ valsartan) to valsartan in management of HFpEF	Phase 2, randomized, double-blind, parallel-group, active-controlled trial Eligibility: Patients aged ≥40 years with LVEF ≥45% and documented history of HF with associated signs or symptoms Trial duration: 36 weeks – 12 weeks main study period and 24-week extension period	301 HFpEF patients randomly assigned to sacubitril/ valsartan or valsartan Treatment stratified by previous use of ACEI or ARB and region 274 patients (sacubitril/valsartan, n = 137; valsartan, n = 137) had valid SBP reading at baseline and at 12 weeks	Primary: Change from baseline in NT-proBNP at 12 weeks Secondary: Changes in echocardiographic measures, BP, and NYHA class Other: Renal function (assessed by serum creatinine, eGFR, cystatin C, UACR, and WRF)	Primary: At 12 weeks, NT-proBNP was significantly reduced in the sacubitril/valsartan group compared with the valsartan group (ratio of change, 0.77; 95% CI, 0.64, 0.92; P = 0.005) Secondary: Mean reduction in SBP in the sacubitril/valsartan group at 12 weeks, was −9 mm Hg and in the valsartan group was −3 mm Hg; P = 0.002 Greater improvements in left atrial diameter and volume index at 36 weeks with sacubitril/valsartan vs valsartan Greater proportion of patients showed improvement in NYHA class at 36 weeks with sacubitril/valsartan Renal function: Less decline in eGFR with sacubitril/valsartan vs valsartan (–1.5 ± 13.1 vs –5.2 ± 11.4 mL/min per 1.73 m^2; P = 0.008) Fewer patients had WRF in the sacubitril/valsartan (12%) vs valsartan (18%) groups; adjusted P = 0.28 UACR increased with sacubitril/valsartan (2.4–2.9 mg/mmol) and remained stable with valsartan (2.1–2.0 mg/mmol; P for difference = 0.016). Effects of sacubitril/valsartan on left atrial size, NT-proBNP, NYHA class, and eGFR were independent of change in blood pressure
PARAGON-HF Solomon SD, et al. 2017 and 2019 [87,88]	To compare the long-term efficacy and safety of sacubitril/valsartan to valsartan in patients with chronic symptomatic HFpEF	Phase 3, randomized, double-blind, parallel-group, active-controlled, 2-arm, event-driven trial comparing the long-term efficacy and safety of valsartan and sacubitril/valsartan in patients with chronic symptomatic HFpEF. Patients aged ≥50 years, have LVEF ≥45%, evidence of structural heart disease, including either LV hypertrophy or left atrial enlargement	Valsartan 160 mg bid or sacubitril/valsartan 97/103 mg bid (LCZ696 200 mg bid)	Primary: Composite of CV death and total (first and recurrent) HHF Secondary: Change in KCCQ and NYHA class at 8 months, time to first occurrence of either ≥50% eGFR decrease from baseline, ESRD, or death due to renal failure; time to all-cause death	Primary: The primary composite outcome was reduced by sacubitril/valsartan, but failed to reach statistical significance (rate ratio, 0.87; 95% CI, 0.75, 1.01; P = 0.06) Secondary: More patients in the sacubitril/valsartan group had an improvement in NYHA class than those in the valsartan group (odds ratio for improvement, 1.45; 95% CI, 1.13, 1.86) Worsening renal function occurred more commonly in the valsartan group than the sacubitril/valsartan group (2.7% vs 1.4%; HR, 0.50; 95% CI, 0.33, 0.77) At 8 months, the sacubitril/valsartan group experienced less reduction in KCCQ clinical summary score compared with the valsartan group (mean decrease in score was 1.6 points and 2.6 points, respectively [95% CI, 0.0, 2.1]) All-cause death did not differ between groups (HR, 0.97; 95% CI, 0.84, 1.13)
DECLARE-TIMI 58 Wiviott SD, et al. 2019 [89]	To determine the effect of dapagliflozin on renal and cardiovascular outcomes in patients with established type 2 DM and risk factors for or with established ASCVD	Phase 3, randomized, double-blind, placebo-controlled, international study Patients aged ≥40 years, had type 2 DM and established or with multiple risk factors (aged ≥55 years for men; ≥60 years for women) for ASCVD, which is defined as ischemic heart disease, ischemic cerebrovascular disease, or peripheral artery disease	Dapagliflozin 10 mg daily or matching placebo	Primary safety outcome: MACE (defined as CV death, MI, or ischemic stroke) Primary efficacy outcomes: MACE and the composite of CV death or HHF Secondary renal composite outcome: eGFR declined by 40% or more to <60 mL/min/1.73 m^2, new diagnosis of ESRD, or CV or renal death Other secondary outcome: All-cause death	Primary safety outcome: Dapagliflozin was noninferior to placebo in the reduction of MACE (P < 0.001 for noninferiority) Primary efficacy outcomes: Compared with placebo, patients in the dapagliflozin group had a lower rate of CV death or HHF (4.9% dapagliflozin vs 5.8% placebo; HR, 0.83; P = 0.005), and this was mainly because of a lower rate of HHF in the dapagliflozin group (HR, 0.73; 95% CI, 0.61, 0.88) Secondary renal composite outcome: 4.3% of patients in the dapagliflozin group experienced the secondary renal composite outcome, compared with 5.6% in the placebo group (HR, 0.76; 95% CI, 0.67, 0.87) All-cause death did not differ significantly between groups (6.2% dapagliflozin vs 6.6% placebo; HR, 0.93; 95% CI, 0.82, 1.04)
EMPEROR-Preserved (NCT03057951) [91]	To evaluate the effects of empagliflozin versus placebo on morbidity and mortality (on top of guideline-directed medical therapy for HF) in patients with HFpEF	Phase 3, randomized, double-blind, parallel-group, placebo-controlled study Patients aged ≥18 years, with a LVEF >40%, chronic HF (NYHA class II–IV) for ≥3 months, elevated NT-proBNP, and evidence of structural heart disease	Empagliflozin 10 mg daily or placebo	Primary: Composite of time to first event of CV death or HHF Secondary: Occurrence of HHF; change from baseline in eGFR; CV death; time to DM onset; all-cause death	Study initiated 2 March 2017 Results expected second half of 2020
DELIVER (NCT03619213)	To determine whether dapagliflozin is superior to placebo, when added to standard of care, in reducing the composite of CV death and HF events	Phase 3, international, multicenter, parallel-group, event-driven, randomized, placebo-controlled, double-blind study Adult patients aged ≥40 years with HFpEF (LVEF >40% and evidence of structural heart disease) and NYHA class II–IV	Dapagliflozin 10 mg daily versus placebo	Primary: Time to the first occurrence of any of the components of this composite: (1) CV death; (2) HHF; (3) urgent HF visit Secondary: Total number of HHF and CV deaths; change from baseline in total symptom KCCQ; proportion of patients with worsened NYHA class from baseline; time to occurrence of death from any cause	Study start date: 27 August 2018 Results expected second half of 2021

ACEI: angiotensin-converting enzyme inhibitor; ARB: angiotensin receptor blocker; bid: twice a day; ASCVD: atherosclerotic cardiovascular disease; BP: blood pressure; CV: cardiovascular; DM: diabetes mellitus; EF: ejection fraction; eGFR: estimated glomerular filtration rate; ESRD: end-stage renal disease; HF: heart failure; HFpEF: heart failure with preserved ejection fraction; HHF: hospitalization due to heart failure; HR: hazard ratio; KCCQ: Kansas City Cardiomyopathy Questionnaire; LV: left ventricular; LVEF: left ventricular ejection fraction; MI: myocardial infarction; NT-proBNP: N-terminal prohormone of B-type natriuretic peptide; NYHA: New York Heart Association; SBP: systolic blood pressure; UACR: urinary albumin to creatinine ratio; WRF: worsening renal function (determined as serum creatinine increase of >0.3 mg/dL and/or >25% between 2 time points).

The DECLARE-TIMI 58 (Dapagliflozin Effect on Cardiovascular Events-Thrombolysis in Myocardial Infarction [MI] 58) study evaluated the effect of the SGLT2i dapagliflozin on the two primary efficacy outcomes of MACE (defined as CV death, myocardial infarction, or ischemic stroke) and a composite of CV death or HHF [89]. This study included patients with baseline type 2 DM and either a history of atherosclerotic CV disease or multiple risk factors for atherosclerotic CV disease [89]. After a median follow-up of 4.2 years, dapagliflozin did not result in a lower rate of MACE compared with placebo but did lower the rate of CV death or HHF by 17% (HR, 0.83; P = 0.005); the lower rate was largely the result of the reduction of HHF in the dapagliflozin group (HR, 0.73 [95% CI, 0.61, 0.88]) [89]. Subsequent analysis of this trial indicated that 11.6% of the enrolled patients had history of HF: 3.9% of patients had documented HFrEF (EF <45%) and 7.7% of patients did not have reduced EF (defined as either having documented EF ≥45% [i.e. HFpEF] or without a documented EF) [96]. Among patients with baseline HF history, dapagliflozin did not reduce the risk of CV death or HHF in patients with HFpEF but instead showed a greater extent of benefit in patients with HFrEF (HFpEF HR, 0.88 [95% CI, 0.66, 1.17] vs HFrEF HR, 0.62 [95% CI, 0.45, 0.86]) [96].

More data on morbidity and mortality outcomes with SGLT2i are being investigated in the ongoing Empagliflozin outcome trial in patients with chronic heart failure with preserved ejection fraction (EMPEROR-Preserved; NCT03057951) and Dapagliflozin evaluation to improve the lives of patients with preserved ejection fraction heart failure (DELIVER; NCT03619213) studies. These studies will allow for evaluation of the effect of empagliflozin and dapagliflozin added on top of standard of care versus placebo on hospitalizations and/or cardiovascular mortality in patients with HFpEF, with results anticipated in 2020 and 2021, respectively.

4.3. A stage-based approach to HFpEF management

As reflected in the ACCF/AHA guidelines, the management and treatment of patients with HFpEF should follow a similar approach to that used for patients with HFrEF [3]. The guidelines categorize HF into stages A–D depending on disease severity, and a stage-based approach is recommended to guide therapy in HFpEF [3]. Patients with stage A HF are at risk for HF but without structural heart disease or symptoms of HF; management includes treatment of hypertension and hyperlipidemia with antihypertensive drugs and statins as indicated by guidelines [3]. Management of other conditions (e.g. obesity, cardiotoxic agents) that may increase risk of HF should also be addressed. Patients with stage B HF have structural heart disease without signs or symptoms of HF; management focuses on prevention of symptomatic HF and CV events with statins and antihypertensive drugs [3]. Patients with HFpEF may receive treatment with guideline-directed medical therapy similar to patients with HFrEF such as ACEI, ARBs, and/or β-blockers [3]. Stage C HF describes patients with structural heart disease with prior or current symptoms of HF, and stage D describes patients with refractory HF requiring specialized interventions [3]. Management of stage C focuses on treating volume overload with diuretics, controlling blood pressure according to standard clinical guidelines and treating hypertension with β-blockers, ACEIs, and/or ARBs, and treating atrial fibrillation according to standard clinical guidelines [11]. Mineralocorticoid-receptor antagonist, in select patients, and ARBs might be considered to decrease hospitalizations [11]. Last, for patients with stage C HFpEF and coronary artery disease who have symptoms or myocardial ischemia that is having an adverse impact, consider treating with coronary revascularization [11]. Advanced interventions (i.e. left ventricular assist devices, transplantation) are required to manage patients with stage D HF [3]. Strategies to reduce elevated left atrial pressure with an interatrial shunt device are being investigated for HFpEF [97].

4.4. Lifestyle interventions

Exercise intolerance is one of the primary symptoms of HFpEF. Therefore, improving exercise capacity should be a key focus of disease management. Randomized controlled trials that have assessed the impact of exercise training on aerobic fitness and QoL exclusively in patients with HFpEF have demonstrated the benefits of exercise training [15,98–102]. Similarly, dietary salt restriction may be beneficial for patients with hypertensive HFpEF. Patients who followed the Dietary Approaches to Stop Hypertension diet and maintained a daily sodium intake of 50 mmol (1150 mg)/2100 kcal exhibited a significant reduction in blood pressure, arterial stiffness, and oxidative stress after 3 weeks [103]. Other dietary restrictions to improve glucose control in patients with HFpEF should also be considered; evidence suggests a significant association between hemoglobin A1c and the risk of incident HF (HFrEF and HFpEF) among patients with diabetes [104]. Within the primary care setting, QoL assessments can aid in the identification of patients who may benefit from a proactive, holistic approach to care which includes multidisciplinary input (e.g. social support) [105].

Thus, in the absence of any disease-modifying therapies, HFpEF management focuses on addressing underlying comorbidities [35]. Phenotypic categorization of patients based on comorbidities and treatable etiologies can assist in identifying the best management strategy for each individual patient [35].

5. Conclusions

The diagnosis of HFpEF is challenging because there is a need to differentiate it from potential noncardiac causes of symptoms suggestive of HF. Recommended therapies for HFpEF are directed at treating symptoms (especially comorbidities) and reducing risk factors that may worsen CV disease. New and emerging therapies have demonstrated improvements across numerous parameters of HFpEF in this patient population, and the results from recent clinical trials should provide insights on morbidity and mortality outcomes and serve to inform alternative treatment strategies for this disease. PCPs play a crucial role in the management of patients with HFpEF. Early and accurate identification of HFpEF is important to optimize treatment opportunities and reduce the likelihood of disease progression, hospitalization and/or rehospitalization, and premature death.

Declaration of funding

Novartis Pharmaceuticals Corporation provided funding for medical writing and editorial assistance and reviewed the content for medical accuracy but had no role in the preparation of the manuscript or decision to publish it.

Declaration of financial/other relationships

Neither P.P.T. nor D.G. received honorarium or other form of compensation for their participation in this project. The coauthors were involved in every stage of manuscript development and had full control over final content.

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Declaration of interest

Neither P.P.T. nor D.G. received honoraria or other form of compensation for their participation in this project. The co-authors were involved in every stage of manuscript development and had full control over final content. Novartis Pharmaceuticals Corporation provided funding for medical writing and editorial assistance and reviewed the content for medical accuracy but had no role in the preparation of the manuscript or decision to publish it.

Acknowledgments

Medical writing and editorial assistance were provided by Traci Stuve, MA, and Madeline Pfau, PhD, of ApotheCom (Yardley, PA, USA), and Matilda Toivakka, PhD, and Brittany Jarrett, PhD, of Complete HealthVizion (Chicago, IL, USA), and was funded by Novartis Pharmaceuticals Corporation.