INTRODUCTION — Heart failure with preserved ejection fraction (HFpEF) is a clinical syndrome in patients with current or prior symptoms of HF with a left ventricular ejection fraction (LVEF) ≥50 percent and evidence of cardiac dysfunction as a cause of symptoms (eg, abnormal LV filling and elevated filling pressures) [1-5]. Most patients with HFpEF display normal LV volumes and evidence of diastolic dysfunction (eg, elevated filling pressures at rest or with exertion) [2,6-8]. By contrast, HF with reduced EF (HFrEF) is characterized by increased LV volumes and reduced LVEF. Previously, HFpEF was termed "diastolic HF" and HFrEF was described as "systolic HF."
The diagnosis of HFpEF is limited to patients with current or prior symptoms of HF (American College of Cardiology/American Heart Association [ACC/AHA] stages C and D HF) but excludes patients with stage A (at high risk for HF but without structural heart disease or symptoms of HF) or stage B HF (structural heart disease but no symptoms or signs of HF) (table 1). Asymptomatic diastolic dysfunction is a common form of ACC/AHA stage B HF which is not encompassed by the term HFpEF but is associated with risk of developing HFpEF. (See "Determining the etiology and severity of heart failure or cardiomyopathy", section on 'Stages in the development of HF' and 'Prevalence and demographics' below and 'Non-HF conditions' below.)
HFpEF should also be distinguished from other causes of HF with an LVEF ≥50 percent that are treated differently, such as valvular heart disease, pericardial disease, cardiac amyloidosis, and high-output HF (table 2). (See 'Differential diagnosis' below.)
The etiology, clinical manifestations, and diagnosis of patients with HFpEF will be reviewed here. Issues related to treatment, prognosis, and pathophysiology are discussed separately. (See "Treatment and prognosis of heart failure with preserved ejection fraction" and "Pathophysiology of heart failure with preserved ejection fraction".) (Related Pathway(s): Heart failure: Diagnosis and classification.)
PREVALENCE AND DEMOGRAPHICS — Among all patients with HF worldwide, nearly half have an LVEF ≥50 percent (including those with HFpEF), nearly half have an LVEF ≤40 percent (HF with reduced ejection fraction [HFrEF]), and the remaining 10 to 24 percent have HF with mid-range ejection fraction (HFmrEF; LVEF 41 to 49 percent). [9-14].
The proportion of patients with HF who have HFpEF appears to be increasing and is higher in older adults [3,4,15-18]. A Mayo Clinic study examined all consecutive patients hospitalized with decompensated HF from 1987 through 2001 [19]. The proportion of patients with the diagnosis of HFpEF increased over time and was significantly higher among community patients than among referral patients (55 versus 45 percent). Over the next decade (2000 through 2010), the proportion of HF patients with HFpEF continued to increase [20]. Data from the Framingham Heart Study and the Cardiovascular Health Study have shown similar findings, with a decreased incidence of HFrEF over the past two decades while the incidence of HFpEF has increased by 45 percent [21].
Data from the Atherosclerosis Risk in Communities (ARIC) study show that HFpEF is by far the dominant form of HF among older adults in the United States, representing 65 to 77 percent of prevalent cases [18]. This study also showed that the majority of older adults in the community have either risk factors for HF (Stage A HF, 52 percent) or structural remodeling (Stage B HF, 30 percent) and are thus at increased risk for developing symptomatic HF (Stage C HF, 13 percent). HFpEF was previously considered to be a disease almost exclusively of aging, but a substantial number of patients with HFpEF are under 65 years of age, and in this cohort, obesity appears to play a larger role [22,23]. (See 'Associated comorbidities' below.)
Community-based studies have shown that the prevalence of HFpEF is similar in men and women after adjusting for age and other risk factors [17,24-28]. Despite the similar age-adjusted prevalence, women outnumber men with HFpEF in the community for two reasons: HFpEF is a disease of aging, and women live longer than men on average, providing more exposure time to develop HFpEF; and female sex is associated with lower risk of death following diagnosis of HFpEF as compared with male sex [29]. By contrast, among patients with HFrEF, men greatly outnumber women, largely owing to the greater burden of coronary disease among men. In a report of more than 100,000 hospitalizations for acute decompensated HF, the 50 percent of patients with a preserved EF (defined in this study as LVEF ≥40 percent and thus including HFpEF and HFmrEF) had the following clinical characteristics compared with those with reduced EF: more likely to be older, female, and hypertensive; less likely to have had a prior myocardial infarction; lower in-hospital mortality (3 versus 4 percent) but similar intensive care unit and hospital length of stay [30].
Asymptomatic diastolic dysfunction is associated with increased risk of developing HFpEF and is much more common than HFpEF. The difference in the prevalence of these two conditions was illustrated by a community-based survey that evaluated 2042 subjects ≥45 years of age [31]. The overall prevalence of clinical HF in this study was 2.2 percent; of these, almost half had HFpEF. Among subjects without HF, 28.1 percent had some degree of diastolic dysfunction using Doppler echocardiographic criteria. The prevalence and severity of diastolic dysfunction increases with age, and the development or worsening of diastolic dysfunction is associated with an increased risk of developing HF [32,33]. (See 'Differential diagnosis' below and "Echocardiographic evaluation of left ventricular diastolic function in adults".)
CLINICAL MANIFESTATIONS
Symptoms and signs — Clinical manifestations of HFpEF are the same as those for HF generally, including HF with reduced EF (HFrEF) [34-36]. Patients with clinical HF (American College of Cardiology/American Heart Association [ACC/AHA] stage C or D (table 1)) have one or more current or prior symptoms of HF. Dyspnea (including dyspnea on exertion, paroxysmal nocturnal dyspnea, and orthopnea) and fatigue are by far the most common symptoms of HFpEF (table 3) [37]. Patients with HF may or may not have physical signs of HF (such as elevated jugular venous pressure, pulmonary rales, and lower extremity edema). Many patients with HFpEF present with exertional chest pain, so the diagnosis of HFpEF should be entertained in patients with exertional chest pain, particularly when significant epicardial coronary disease is not identified [38-40]. (See "Heart failure: Clinical manifestations and diagnosis in adults", section on 'Clinical presentation'.)
The similarity in clinical manifestations for HFpEF and HFrEF was illustrated in a report in which clinical data from 59 patients aged at least 60 years with symptoms of HF and an LVEF ≥50 percent were compared with data from 60 patients of the same age with HF and an LVEF ≤35 percent and from 28 age-matched healthy controls [41]. The patients with HFpEF had similar clinical manifestations (including peak VO2 and neurohumoral activation) to those with HFrEF, although some parameters were less severe (natriuretic peptide levels, some quality-of-life measures). In other series, cardiopulmonary exercise parameters [42], central cardiac filling pressures, and pulmonary hypertension severity [43] were indistinguishable.
Associated comorbidities — Evaluation of patients with suspected HFpEF includes identifying and evaluating comorbidities, as these affect management and prognosis. (See "Treatment and prognosis of heart failure with preserved ejection fraction".)
As noted above, HFpEF is commonly associated with aging [3,4,15,16] and the following conditions, which are discussed further separately:
●Cardiovascular disorders
•Hypertension [15,44,45] (see "Overview of hypertension in adults")
•Coronary heart disease (CHD) [46] (see "Angina pectoris: Chest pain caused by fixed epicardial coronary artery obstruction" and 'Tests for coronary artery disease' below)
•Atrial fibrillation (AF) [47,48] (see "Atrial fibrillation: Overview and management of new-onset atrial fibrillation")
●Metabolic disorders
•Diabetes mellitus and metabolic syndrome [49,50] (see "Heart failure in patients with diabetes mellitus: Epidemiology, pathophysiology, and management")
•Obesity [22] (see "Obesity in adults: Prevalence, screening, and evaluation")
●Respiratory disorders
•Sleep-disordered breathing [51] (see "Sleep-disordered breathing in heart failure")
•Chronic obstructive lung disease [51] (see "Chronic obstructive pulmonary disease: Definition, clinical manifestations, diagnosis, and staging")
●Kidney disease [52] (see "Early detection of chronic kidney disease" and "Cardiorenal syndrome: Definition, prevalence, diagnosis, and pathophysiology")
●Anemia [51] (see "Diagnostic approach to anemia in adults")
These comorbidities are common in patients with HFpEF as well as in patients with HFrEF [51,53].
Changes in ventricular, vascular, and peripheral structure and function causing HFpEF are believed to be related to aging and comorbid conditions commonly associated with HFpEF, but the mechanisms for these changes have not been established. Although comorbidities are an important contributor to the outcome in HFpEF, HFpEF is more than a collection of associated conditions. Abnormalities in cardiac structure, function, and clinical outcomes are all more impaired in patients with HFpEF compared with matched patients with similar comorbidities but no HF [54,55]. The pathophysiology of HFpEF, including proposed mechanisms by which comorbidities lead to HF, is discussed separately. (See "Pathophysiology of heart failure with preserved ejection fraction".)
Common precipitants of exacerbations — Patients with HFpEF have particular difficulty tolerating certain kinds of hemodynamic stress which cause increased LV diastolic pressure. The most common stresses encountered by patients with HFpEF are exercise, AF, hypertension, intravenous fluid load, and ischemia:
●Physical activity requires an increase in cardiac output to the body [56]. In patients with HFpEF, there is an inability to tolerate the increase in blood return to the heart along with changes in cardiac afterload and tachycardia; thus, left atrial and pulmonary venous pressure are elevated [57,58]. This is associated with symptoms of dyspnea, poor aerobic capacity, and increased risk of death [59-61].
●Patients with HFpEF tolerate AF poorly, since the loss of atrial contraction can dramatically reduce LV filling and limit the stroke volume [62,63]. AF is also associated with more severe right-sided HF in HFpEF [47,64,65]. (See "Hemodynamic consequences of atrial fibrillation and cardioversion to sinus rhythm".)
●Elevations in systemic blood pressure due to arterial stiffening increase LV wall stress, which can impair or delay myocardial relaxation and promote greater elevation in pulmonary venous pressures [66]. (See "Treatment of hypertension in patients with heart failure", section on 'Treatment of hypertension in patients with heart failure with preserved ejection fraction (HFpEF)' and "Evaluation of secondary hypertension".)
●Myocardial ischemia causes acute induction or worsening of diastolic dysfunction, thus elevating left atrial and pulmonary venous pressures. This explains why many patients with CHD have respiratory symptoms with their anginal pain, including wheezing, a limited ability to take a deep breath, and shortness of breath. These respiratory symptoms can occur in the absence of anginal pain and are often referred to as "anginal equivalents." Notably, ischemia may develop in HFpEF even in the absence of epicardial coronary artery disease (CAD) and is associated with worse exercise capacity and ventricular function [67-69]. (See "Angina pectoris: Chest pain caused by fixed epicardial coronary artery obstruction".)
Episodes of hemodynamic decompensation may result in pulmonary congestion or edema severe enough to be life threatening. Studies using implantable hemodynamic monitors have shown that LV diastolic pressures rise in a progressive fashion prior to the abrupt onset of HF (figure 1) [6,7,70-74]. While the development of symptoms is rapid, increases in diastolic pressure frequently occur gradually over a period of days to weeks. (See "Pathophysiology of cardiogenic pulmonary edema".)
Additional precipitating factors for acute HF (with HFpEF, HFrEF, or HFmrEF) are discussed separately. (See "Approach to diagnosis and evaluation of acute decompensated heart failure in adults", section on 'Causes and precipitating factors'.)
Initial tests — The initial evaluation of patients with HF commonly includes a chest radiograph and electrocardiogram (ECG). An echocardiogram is a key initial test, as discussed below. (See 'Approach to diagnosis' below and 'Echocardiography' below.)
A chest radiograph is commonly obtained in patients with HF to assess for signs of pulmonary edema and identify other causes of dyspnea. The chest radiograph may show cardiomegaly and/or radiographic evidence of pulmonary edema. Most patients with HFpEF will have a normal chest radiograph, so this should not deter further evaluation [45].
An ECG is routinely performed in patients with HF, although findings are often nonspecific. Evidence of myocardial ischemia or prior infarction may be detected. AF is observed in approximately 40 to 50 percent of patients at any given time, with a lifetime risk of 67 percent [62]. The presence of AF on ECG in a patient with normal LVEF and dyspnea increases the odds that HFpEF is present by more than 20-fold [75].
DIAGNOSIS — HFpEF is a clinical syndrome in which patients have symptoms and signs of HF, a normal or near normal LVEF (≥50 percent), and evidence of cardiac dysfunction as a cause of symptoms (eg, abnormal LV filling and elevated filling pressures) [76,77].
When to suspect HFpEF — HFpEF should be suspected in individuals with all three of the following features based upon clinical evaluation including history, physical examination, and echocardiography:
●One or more symptoms of HF such as dyspnea or fatigue; physical signs of HF may or may not be present (table 3). (See 'Clinical manifestations' above and "Heart failure: Clinical manifestations and diagnosis in adults".)
●An LVEF ≥50 percent. Evaluation of patients with HF includes Doppler transthoracic echocardiography to evaluate LVEF, estimate pulmonary artery systolic pressure, assess LV filling pressure, and assess cause of HF. If the LVEF cannot be adequately estimated by echocardiography, other noninvasive cardiac imaging methods are suggested. (See 'Other cardiac imaging' below.)
Patients with similar symptoms and lower LVEF should be evaluated for other subtypes of HF. (See "Heart failure: Clinical manifestations and diagnosis in adults" and "Treatment and prognosis of heart failure with mid-range ejection fraction".)
●No apparent cause of HF symptoms other than HFpEF. Causes of the clinical syndrome of HF with an LVEF ≥50 percent other than HFpEF include a cardiomyopathy (eg, hypertrophic or restrictive cardiomyopathy), cardiac amyloidosis, significant valve disease (severe stenosis or regurgitation or at least moderate mixed stenosis and regurgitation), pericardial disease (eg, constrictive pericarditis), and high-output HF (table 2). Clinical evaluation including echocardiography is helpful in identifying these conditions. Non-HF causes of symptoms should also be excluded. (See 'Differential diagnosis' below.)
Approach to diagnosis — We suggest the following approach for diagnosis of HFpEF after initial clinical assessment of the patient’s history, symptoms, and signs (algorithm 1):
●Calculation of H2FPEF (Heavy, Hypertensive, atrial Fibrillation, Pulmonary hypertension, Elder, Filling pressure) and Heart Failure Association Pretest assessment, Echocardiography and natriuretic peptide, Functional testing, Final etiology (HFA-PEFF) scores in patients with suspected HFpEF to estimate the probability of HFpEF versus other causes of symptoms [45,78,79]. (See 'H2FPEF score' below and 'HFA-PEFF score' below.)
●Score interpretation. The probability that HFpEF is the cause of symptoms increases with increasing H2FPEF score (range 0 to 9) and increasing HFA-PEFF score (range 0 to 6):
•Low probability – A low H2FPEF score of 0 or 1 and HFA-PEFF score 0 or 1 is associated with a low probability of HFpEF.
-A low score suggests that symptoms are most likely due to a noncardiac cause, and such causes should be investigated.
-However, if the cause of symptoms remains uncertain after evaluation for noncardiac causes, cardiology consultation and a hemodynamic exercise test is suggested to determine if HFpEF is present. (See 'Hemodynamic exercise test' below.)
•Intermediate probability – H2FPEF and/or HFA-PEFF score ≥2 and neither score is high probability. In this setting, we suggest cardiology consultation and a hemodynamic exercise test. (See 'Hemodynamic exercise test' below.)
•High probability – An H2FPEF score of 6 to 9 or an HFA-PEFF score of 5 or 6 is associated with a high probability of HFpEF and is therefore considered diagnostic for HFpEF.
●On hemodynamic exercise testing (performed in intermediate-risk and only selected low-risk patients), pulmonary capillary wedge pressure (PCWP) ≥15 mmHg at rest or ≥25 mmHg during exercise is diagnostic for HFpEF. (See 'Hemodynamic exercise test' below.)
Diagnostic scores — We use the H2FPEF and HFA-PEFF scores to estimate the probability of HFpEF versus noncardiac causes of symptoms, while recognizing the limitations of these scores, as described in the sections on these scores below [45,78,79]. One limitation common to both scores is that they include parameters measured by echocardiography (including functional measures for both scores and morphologic measures for the HFA-PEFF score), which may be subject to inaccurate results with suboptimal image acquisition and interpretation. (See 'Echocardiography' below and "Echocardiographic assessment of the right heart", section on 'Pulmonary artery pressure' and "Echocardiographic evaluation of left ventricular diastolic function in adults", section on 'Tissue Doppler imaging'.) (Related Pathway(s): Heart failure: Diagnosis and classification.)
H2FPEF score — The H2FPEF score was developed and clinically validated by our group at the Mayo Clinic. The score is the sum of points assigned to the following clinical variables (score range 0 to 9) (algorithm 1) [45]:
●Heavy – Body mass index >30 kg/m2 (2 points) (calculator 1) (see "Obesity in adults: Prevalence, screening, and evaluation")
●Hypertensive – Hypertensive and treated with two or more antihypertensive medicines (1 point) (see "Overview of hypertension in adults")
●Atrial Fibrillation (AF) – Paroxysmal or persistent (3 points) (see "Atrial fibrillation: Overview and management of new-onset atrial fibrillation")
●Pulmonary hypertension (PH) – Pulmonary artery systolic pressure >35 mmHg using Doppler echocardiography (1 point) (see "Echocardiographic assessment of the right heart", section on 'Estimation of pulmonary artery systolic pressure')
●Elder – Age >60 years (1 point)
●Filling pressure – Doppler echocardiographic E/e' >9 (1 point; E is the peak velocity of early LV filling; e' is the peak early diastolic velocity of LV myocardium adjacent to the mitral annulus by tissue Doppler) (see "Echocardiographic evaluation of left ventricular diastolic function in adults", section on 'Tissue Doppler imaging')
The H2FPEF score was derived based on data in 414 patients with an LVEF ≥50 percent (267 with HFpEF and 147 with noncardiac dyspnea) and validated in a test cohort of 100 patients (61 with HFpEF) [45]. Diagnoses for all patients in this study were validated by invasive hemodynamic exercise test. The odds of HFpEF doubled for each 1-unit H2FPEF score increase (odds ratio 1.98, 95% CI 1.74-2.30), with an area under the curve of 0.841. The H2FPEF score was superior to an algorithm based on expert consensus (increase in area under the curve of 0.169, 95% CI 0.120-0.217). In the independent test cohort, the area under the curve was 0.886.
The generalizability and prognostic value of the H2FPEF score has been confirmed in other patient populations [80,81]. In a study of data from the TOPCAT trial, higher H2FPEF scores were associated with increased risk of the primary outcome (composite of hospitalization for HF, cardiovascular death, or aborted cardiac arrest; hazard ratio 1.12 per point, 95% CI 1.02-1.23) [81]. One study found that the discriminatory capacity of the H2FPEF score was high in patients with normal LVEF and low in patients presenting with dyspnea [82]; however this finding is of uncertain significance as definitive diagnoses for the study subjects were not provided.
HFA-PEFF score — The HFA-PEFF score is based upon assessing criteria in three domains (functional, morphologic, and biomarker) [78]. Each domain can contribute a maximum of 2 points (score range 0 to 6):
●Functional – These parameters are measured by Doppler echocardiography. (See 'Echocardiography' below.)
•Major criteria (2 points if ≥1 of the following is present):
-Septal e’ <7 cm/s
-Lateral e’ <10 cm/s
-Average E/e’ ratio ≥15
-Peak tricuspid regurgitation velocity >2.8 m/s (pulmonary artery systolic pressure [PASP] >35 mmHg)
•Minor criteria (1 point if ≥1 of the following is present):
-Average E/e’ 9 to 14
-Global longitudinal strain (GLS) <16 percent
●Morphologic – These parameters are measured by echocardiography (or cardiovascular magnetic resonance [CMR] if not adequately measured by echocardiography) and indexed to body surface area in m2.
RWT = LVPW X 2 / LVIDD
in which RWT is relative wall thickness, LVPW is the LV posterior wall thickness measured at end-diastole, and LVIDD is the LV internal end-diastolic diameter. (See 'Echocardiography' below.)
•Major criteria (2 points if ≥1 of the following):
-Left atrial volume index (LAVI) >34 mL/m2
-LV mass index (LVMI) ≥149 g/m2 (men) or ≥122 g/m2 (women) and RWT >0.42
•Minor criteria (1 point if ≥1 of the following):
-LAVI 29 to 34 mL/m2
-LVMI >115 g/m2 (men) or >95 g/m2 (women)
-RWT >0.42
-LV wall thickness ≥12 mm
●Biomarker – Different threshold serum natriuretic peptide levels are used based on whether the patient is in sinus rhythm or AF. (See 'Natriuretic peptide level' below.)
•Major criteria (2 points if either of the following):
-In sinus rhythm: NT-proBNP (N-terminal pro-B-type natriuretic peptide) >220 picograms/mL or BNP (B-type natriuretic peptide) >80 picograms/mL
-In AF: NT-proBNP >660 picograms/mL or BNP >240 picograms/mL
•Minor criteria (1 point if either of the following):
-In sinus rhythm: NT-proBNP 125 to 220 picograms/mL or BNP 35 to 80 picograms/mL
-In AF: NT-proBNP 365 to 660 picograms/mL or BNP 105 to 240 picograms/mL
A study assessed the diagnostic and predictive value of the HFA-PEFF score in two separate cohorts, the Maastricht cohort (228 patients with clinical diagnosis of HFpEF and 42 controls) and the Northwestern Chicago cohort (459 patients with clinical diagnosis of HFpEF) [83]. Compared with clinical diagnosis by a cardiologist with expertise in HFpEF, a high HFA-PEFF score (5 or 6) identified HFpEF with a sensitivity of 59 percent and specificity of 93 percent. A low score (0 or 1) ruled out HFpEF with a sensitivity of 99 percent and specificity of 19 percent. However, 36 percent of patients were categorized as intermediate risk, for which additional diagnostic testing is advised. The risk of hospitalization for HF was elevated among patients with a HFA-PEFF score of ≥5 compared with those with lower scores in both the Maastricht and the Northwestern Chicago cohorts.
Comparison of scores — A study applying the H2FPEF and HFA-PEFF scores to a population of older adults found that the scores had similar prognostic value for adverse clinical outcomes but identified only slightly overlapping high-risk groups [84]. The study included 4892 Atherosclerosis Risk in Communities (ARIC) study participants (mean age 75±5 years; 58 percent women). Participants were categorized as asymptomatic (76.6 percent), symptomatic with clinical diagnosis of HFpEF (10.3 percent), or symptomatic with at least moderate unexplained dyspnea (13.1 percent) based upon clinical evaluations [84].
●At mean follow-up of 5.3 years, rates of HF hospitalization or death per 1000 person-years were higher among patients with known HFpEF compared with asymptomatic patients (71.6 versus 20.7).
●Among the 641 participants with unexplained dyspnea, rates of HF hospitalization or death per 1000 person-years were higher among those with higher H2FPEF or HFA-PEFF scores.
•H2FPEF score – low (score 1 to 2; event rate 26.8 per 1000 person years), intermediate (score 3 to 4; 47.1), high (score ≥6; 84.9)
•HFA-PEFF score – low (score 0 to 2; event rate 31.8 per 1000 person years), intermediate (score 3; 32.4), high (score 5 to 6; 61.8)
●Among participants with unexplained dyspnea, 28 percent had discordant findings (high risk by only 1 score), with only 4 percent high risk by both scores.
Key tests — The key initial diagnostic tests in patients with suspected HFpEF are Doppler echocardiography and serum natriuretic peptide levels. If echocardiography is suboptimal for measurement of morphologic parameters, then other noninvasive cardiac imaging is indicated. (See 'Other cardiac imaging' below.)
Cardiology consultation and hemodynamic exercise test are indicated in selected patients with intermediate H2FPEF scores (and selected patients with low H2FPEF scores and undetermined cause of symptoms despite evaluation for noncardiac causes) (algorithm 1). (See 'H2FPEF score' above.)
Echocardiography — Doppler echocardiography is a key component of the diagnosis and evaluation of patients with suspected HF. Echocardiography is helpful in determining when to suspect HFpEF and contributes to the H2FPEF and HFA-PEFF score estimates of the probability of HFpEF [3,4,45,78,85,86]:
●In identifying patients with suspected HFpEF, echocardiography is helpful in demonstrating that the LVEF is preserved (≥50 percent) and LV volume is normal. (See "Tests to evaluate left ventricular systolic function", section on 'Echocardiography'.)
●Two components of the H2FPEF score (estimated PASP and E/e’) and two domains of the HFA-PEFF score (functional and morphologic) are derived from Doppler echocardiography. (See 'H2FPEF score' above and 'HFA-PEFF score' above.)
•PASP >35 mmHg is a criterion in both the H2FPEF and HFA-PEFF scores. Elevation in PASP estimated by echocardiography is very common in patients with HFpEF, and the identification of an elevated PASP in an older patient with dyspnea should trigger consideration for the diagnosis of HFpEF [87]. Methodology for estimating PASP and limitations of such estimates are discussed separately. (See "Echocardiographic assessment of the right heart", section on 'Pulmonary artery pressure'.)
•Functional parameters suggestive of diastolic dysfunction (figure 2) are included in the H2FPEF score (E/e’) and HFA-PEFF score (E/e’ and other parameters). Methodology for measuring parameters of diastolic function and limitations of these measures are discussed separately (figure 2). (See "Echocardiographic assessment of the right heart", section on 'Pulmonary artery pressure' and "Echocardiographic evaluation of left ventricular diastolic function in adults".)
•Morphologic parameters for left atrial volume, LV mass, and LV wall thickness are included in the HFA-PEFF score. A pattern of concentric remodeling (figure 3) is commonly seen in patients with HFpEF. Measurement of these parameters and their limitations are discussed separately. (See "Echocardiographic evaluation of the atria and appendages", section on 'LA volume'.)
If morphologic parameters cannot be adequately measured by echocardiography, measurement by CMR is suggested, although slight differences in measures among imaging modalities have been reported [78].
●Echocardiography is also helpful in identifying causes of HF with an LVEF ≥50 percent other than HFpEF, including valvular and pericardial disease, as discussed below. (See 'Differential diagnosis' below.)
Functional and morphologic parameters provide complementary information. While Doppler and tissue Doppler measurements can be used together to estimate LV diastolic filling pressures (see "Echocardiographic evaluation of left ventricular diastolic function in adults", section on 'Estimation of left atrial pressure' and "Echocardiographic assessment of the right heart", section on 'Pulmonary artery pressure'), these estimates reflect ambient pressures at the time of measurement, and accuracy is imperfect [88,89]. By contrast, in the absence of AF, left atrial size (best measured as left atrial volume) reflects the degree of chronic LV diastolic pressure elevation (ie, left atrial volume integrates the chronically elevated LV diastolic pressure over time). (See "Echocardiographic evaluation of the atria and appendages", section on 'Left atrium' and "Echocardiographic evaluation of left ventricular diastolic function in adults".)
Other cardiac imaging — Echocardiography is generally the preferred method of assessing cardiac structure and function. However, if left atrial volume, LV mass, and LV wall thickness cannot be adequately assessed by echocardiography, then CMR (or computed tomography [CT]) is suggested, although slight differences in measures among imaging modalities have been reported [78]. If the LVEF cannot be adequately assessed by echocardiography, alternative methods include CMR, cardiac radionuclide ventriculography, and cardiac CT. The choice among these tests is discussed separately. (See "Tests to evaluate left ventricular systolic function".)
Natriuretic peptide level — Serum levels of the natriuretic peptides BNP or NT-proBNP have prognostic value in patients with HFpEF and may aid in diagnosis of HFpEF in selected patients, particularly in those with intermediate H2FPEF scores in whom other causes of elevated levels (such as significant lung disease) have been excluded. However, natriuretic peptide levels must be interpreted with care given the limited sensitivity and specificity of results, and normal levels should not be considered to exclude disease. (See "Natriuretic peptide measurement in heart failure".)
BNP and NT-proBNP levels should be interpreted with attention to potential confounding factors. Factors limiting the sensitivity and specificity of natriuretic peptide levels include the effect of renal failure (which causes elevated BNP levels and even greater elevation in NT-proBNP levels), the effect of obesity (which tends to depress BNP and NT-proBNP levels), as well as conditions other than left-sided HF that cause elevations in levels. As an example, causes of PH result in right heart strain, which is associated with elevations in BNP and NT-proBNP.
A normal natriuretic peptide level does not exclude the diagnosis of HFpEF [57,90]. This is because natriuretic peptide levels are lower in obesity (common in HFpEF) [22], filling pressures are often normal at rest in HFpEF, and ventricular wall stress (which stimulates natriuretic peptide release) is often mitigated by LV hypertrophy in HFpEF. As an example, in a study, 18 percent of patients with invasively confirmed HFpEF had NT-proBNP levels below the threshold considered to exclude disease (≤125 picograms/mL) [91]. Therefore, a normal BNP or NT-proBNP level does not rule out the diagnosis of HFpEF [57,92,93]. (See "Natriuretic peptide measurement in heart failure".)
By contrast, an elevated NT-proBNP or BNP value is more meaningful and can serve as evidence that HFpEF is present, unless caused by another mechanism. An NT-proBNP >450 picograms/mL has been shown to be 85 percent specific for HFpEF [75]. BNP >100 picograms/mL and NT-proBNP >300 picograms/mL are independent predictors of adverse cardiovascular events in patients with HFpEF [94]. For a given BNP level, the prognosis in patients with HFpEF is similar to that in patients with HF with reduced EF (HFrEF) [95]. (See "Natriuretic peptide measurement in heart failure".)
Hemodynamic exercise test — A hemodynamic exercise test (right heart catheterization with the PCWP assessed at rest and with exercise) is not required for diagnostic evaluation in all patients with suspected HFpEF. However, in selected patients with suspected HFpEF who have intermediate H2FPEF and HFA-PEFF scores, cardiology consultation and right heart catheterization for assessment of cardiac filling pressures at rest and exercise is useful as the clinical gold standard to make or exclude the diagnosis of HFpEF [3,57,91,92]. A hemodynamic exercise test is also helpful for selected patients with low H2FPEF and HFA-PEFF scores with undetermined causes of symptoms despite evaluation for noncardiac causes. (See 'Approach to diagnosis' above.)
On right heart catheterization, PCWP ≥15 mmHg at rest or ≥25 mmHg during exercise is diagnostic for HFpEF. Pressures are measured at end-expiration. Exercise is performed during right heart catheterization with cycle ergometry (in patients with internal jugular venous access) or arm abduction with weights (in those with femoral venous access), though the latter provides a less robust exercise stress [57].
The role of coronary angiography in patients with HFpEF is discussed below. (See 'Tests for coronary artery disease' below.)
Tests for coronary artery disease — CAD is common in HFpEF, seen in approximately two-thirds of patients in angiographic and autopsy studies [46,96]. Mortality is higher in patients with HFpEF and CAD compared with those without.
CAD is a potentially reversible cause of HFpEF [46]. Findings suggesting the possible presence of myocardial ischemia include angina, dyspnea on exertion (such effort symptoms may be anginal equivalents), the presence of multiple risk factors for atherosclerosis, regional wall motion abnormalities on echocardiography or other cardiac imaging, and evidence of ischemia on exercise testing. Evidence of ischemia can be detected by radionuclide myocardial perfusion imaging or stress echocardiography, though false-positive and false-negative test results are not uncommon in people with HFpEF [46]. (See "Overview of stress radionuclide myocardial perfusion imaging" and "Overview of stress echocardiography" and "Microvascular angina: Angina pectoris with normal coronary arteries".)
Stress testing is recommended for patients with HFpEF with standard indications for stress testing (eg, patients with stable angina with intermediate or high pretest likelihood of CHD). In addition, an imaging stress test is suggested in patients presenting with HFpEF who have known CAD and no angina, unless the patients are not eligible for revascularization [76]. However, a study reported high rates of false-positive and false-negative stress tests in people with HFpEF, suggesting that these modalities may be of limited value in this patient population [46]. (See "Stress testing for the diagnosis of obstructive coronary heart disease" and "Selecting the optimal cardiac stress test".)
Coronary angiography is recommended when noninvasive evaluation suggests that ischemia may be contributing to HF or other indications for coronary angiography are present (see "Chronic coronary syndrome: Overview of care", section on 'Identifying patients for angiography and revascularization'). While revascularization may improve symptoms in patients with HFpEF with significant CAD, some patients with HFpEF may have persistent symptoms, including pulmonary edema requiring recurrent hospitalizations despite coronary revascularization [97]. A number of patients with HFpEF develop ischemia with exercise even in the absence of epicardial CAD, manifested by elevation in serum troponin levels [67].
Other tests
Cardiopulmonary exercise test — Noninvasive cardiopulmonary exercise testing has a limited role in differentiating patients with HF from those with lung disease and deconditioning, and such testing cannot reliably differentiate HFrEF from HFpEF [42]. Peak VO2 less than 80 percent predicted on cardiopulmonary exercise testing was formerly considered to be suggestive of HFpEF, but a large study with invasive confirmation of diagnosis revealed that cardiopulmonary exercise testing lacks sufficient sensitivity and specificity to differentiate HFpEF from noncardiac causes of dyspnea [60]. In this study, peak VO2 of less than 14 mL/min/kg was highly specific for HFpEF, and patients with values higher than 20 mL/min/kg were unlikely to have HFpEF. However, patients in the intermediate range of 14 to 20 mL/min/kg showed a high degree of overlap. (See "Approach to the patient with dyspnea", section on 'Cardiopulmonary exercise testing'.)
Exercise echocardiography — Some studies suggest that there may also be a role for exercise echocardiography in the evaluation of HFpEF, but there are conflicting data on its accuracy [91,98-100]. (See "Echocardiographic evaluation of left ventricular diastolic function in adults", section on 'Tissue Doppler imaging'.)
The role of stress testing (including exercise echocardiography) in patients with HFpEF with suspected CAD is discussed below. (See 'Tests for coronary artery disease' above.)
DIFFERENTIAL DIAGNOSIS — It is important to exclude possible mimics of HFpEF. These include non-HF conditions as well as causes of HF with an LVEF ≥50 percent that are not HFpEF.
Non-HF conditions — Among patients who have an LVEF ≥50 percent and symptoms or signs suggestive of HF (such as shortness of breath, ankle edema, or paroxysmal nocturnal dyspnea), some do not have HF but have one or more other conditions such as obesity, deconditioning, advanced age, venous insufficiency, lung disease, or myocardial ischemia [101]. However, many of these conditions also increase the risk that HFpEF is present, so it is important to avoid assuming that one of these conditions is the sole cause of symptoms and pursue workup based upon the pretest probability that HFpEF is present, as described above. (See 'Approach to diagnosis' above.)
While echocardiographic evidence of diastolic dysfunction is a key feature of HFpEF, it is important to distinguish asymptomatic diastolic dysfunction (included in American College of Cardiology/American Heart Association [ACC/AHA] stage B HF) from HFpEF (included in ACC/AHA stages C and D HF) (table 1). The presence of diastolic dysfunction and an LVEF ≥50 percent does not indicate HFpEF unless the clinical syndrome of symptomatic HF is present as described above. (See 'Echocardiography' above and 'Prevalence and demographics' above and 'Diagnosis' above and "Determining the etiology and severity of heart failure or cardiomyopathy", section on 'Stages in the development of HF'.)
HF not caused by HFpEF — Causes of HF in patients with an LVEF ≥50 percent other than HFpEF (table 2) include the following conditions. The distinction among causes is important, because some are responsive to specific therapies directed at the underlying disease:
Cardiomyopathy — Cardiomyopathies include a variety of myocardial disorders that manifest with various structural and functional phenotypes that may or may not be familial; valvular, ischemic, or hypertensive heart disease are excluded from the formal definition of cardiomyopathy. Some patients with HF with an LVEF ≥50 percent have a cardiomyopathy such as hypertrophic or restrictive cardiomyopathy, or cardiac amyloidosis (table 4A-B). In addition, although many or most patients with HF and LV noncompaction have an LVEF <50 percent, some have an LVEF ≥50 percent. Cardiomyopathies should be distinguished from HFpEF, as etiology, prognosis, and management differs. (See "Definition and classification of the cardiomyopathies".)
Cardiomyopathies are generally suspected based upon clinical presentation, family history, and echocardiography. Additional cardiac imaging (eg, CMR imaging) is indicated to evaluate specific causes. Endomyocardial biopsy may be helpful in selected settings in establishing alternative causes of HF, such as identifying specific causes of restrictive cardiomyopathy (eg, amyloidosis). The indications for this procedure are discussed separately. (See "Endomyocardial biopsy".)
●Restrictive cardiomyopathy is characterized by nondilated ventricles with impaired ventricular filling, which is caused by a variety of familial or nonfamilial conditions. LV hypertrophy is typically absent, particularly with idiopathic disease, although increased LV wall thickness may occur with infiltrative disease (such as amyloidosis or iron overload) or storage disease (such as Fabry disease) (table 4A-B). (See "Restrictive cardiomyopathies".)
•Cardiac amyloid represents the most common form of restrictive cardiomyopathy and may be familial or nonfamilial. Cardiac amyloid typically manifests as LV hypertrophy with marked reduction in basal longitudinal strain on echocardiogram. A granular, sparkling appearance is seen in some but not all cases. Low limb lead voltage on the surface ECG with a pseudoinfarction pattern (loss of precordial R wave progression in leads V1 to V6) can suggest an infiltrative process such as amyloidosis, but this feature has low sensitivity. Series have reported that amyloidosis may be a more common etiology of the clinical syndrome of HF with normal LVEF than previously recognized [102,103]. For individuals with suspected immunoglobulin light-chain (AL) cardiac amyloid, CMR is a key test. For patients with suspected transthyretin amyloid (ATTR) cardiomyopathy, bone scintigraphy or CMR is helpful. Given the availability of specific treatments for cardiac amyloid, it is important to differentiate this condition from HFpEF. (See "Cardiac amyloidosis: Epidemiology, clinical manifestations, and diagnosis" and "Amyloid cardiomyopathy: Treatment and prognosis".)
•Familial causes of restrictive cardiomyopathy include an unknown gene mutation, sarcomeric gene mutations, familial amyloidosis (TTR or apolipoprotein mutation), familial causes of iron overload (eg, hereditary hemochromatosis, hereditary anemias), Fabry disease, glycogen storage disease, desminopathy, and pseudoxanthoma elasticum. (See "Cardiac amyloidosis: Epidemiology, clinical manifestations, and diagnosis" and "Clinical manifestations and diagnosis of hereditary hemochromatosis", section on 'Cardiac iron overload'.)
Cardiac Fabry disease should be suspected in patients with LV hypertrophy of unknown etiology with or without other clinical manifestations of Fabry disease, such as severe neuropathic or limb pain, telangiectasias and angiokeratomas, renal manifestations, and cerebrovascular involvement. In men with suspected Fabry disease, the diagnosis is generally confirmed by measurement of leukocyte alpha-Gal A activity. In women with suspected Fabry disease (and men with marginal levels of alpha-Gal A activity), genetic testing is recommended. (See "Fabry disease: Cardiovascular disease" and "Fabry disease: Clinical features and diagnosis".)
•Nonfamilial causes of restrictive cardiomyopathy include amyloid (AL or wild-type ATTR), systemic sclerosis, endomyocardial fibrosis (idiopathic or caused by hypereosinophilic syndrome or drugs), carcinoid heart disease, metastatic cancer, radiation, nonfamilial iron overload (eg, acquired iron-loading anemia, high dietary intake) drug toxicity (anthracycline). (See "Endomyocardial fibrosis" and "Hypereosinophilic syndromes: Clinical manifestations, pathophysiology, and diagnosis" and "Carcinoid heart disease" and "Cardiotoxicity of radiation therapy for breast cancer and other malignancies" and "Cardiotoxicity of radiation therapy for Hodgkin lymphoma and pediatric malignancies", section on 'Heart failure'.)
●Hypertrophic cardiomyopathy is most commonly caused by a mutation in one of several sarcomeric genes. The term hypertrophic cardiomyopathy is also used in a broader sense to include a variety of familial and nonfamilial conditions with increased ventricular wall thickness or mass not caused by pathologic loading conditions, such as hypertension or valve disease (table 4A-B). (See "Hypertrophic cardiomyopathy: Clinical manifestations, diagnosis, and evaluation" and "Definition and classification of the cardiomyopathies", section on 'Hypertrophic cardiomyopathy'.)
•Familial causes in addition to sarcomere gene mutations include unknown mutations, glycogen storage disease, lysosomal storage disease (including Fabry disease), syndromic HCM (eg, Noonan syndrome, LEOPARD syndrome [lentigines, ECG abnormalities, ocular hypertelorism, pulmonic stenosis, abnormal genitalia, retardation of growth, and sensorineural deafness], Friedreich ataxia), and familial amyloidosis (TTR or apolipoprotein mutation).
•Nonfamilial causes include nonfamilial amyloidosis (AL or wild-type ATTR). (See "Cardiac amyloidosis: Epidemiology, clinical manifestations, and diagnosis".)
●LV noncompaction (LVNC) is a cardiomyopathy characterized by prominent LV trabeculae and deep intertrabecular recesses, resulting in a thickened myocardium consisting of a noncompacted layer and a thin compacted layer. Although LVNC was traditionally considered a developmental anomaly, acquired cases have been reported. Although many/most patients with HF and LVNC have a reduced LVEF, some have an LVEF ≥50 percent. (See "Isolated left ventricular noncompaction in adults: Clinical manifestations and diagnosis".)
Other causes of HF — Other non-HFpEF causes of HF with an LVEF ≥50 percent include valve disease, pericardial disease, high-output HF, obstructive lesions, and right HF.
●Valve disease (severe stenosis or regurgitation, or at least moderate mixed stenosis and regurgitation) – This is generally diagnosed and evaluated by echocardiography. Of note, many patients with HFpEF also display some evidence of valvular heart disease, but not in the severe range. Mild to moderate valve lesions in patients with HFpEF are generally considered to be "bystanders" rather than causal of HF symptoms. In particular, nonsevere mitral and tricuspid regurgitation are both very common in HFpEF. The presence of mild to moderate mitral regurgitation in HFpEF is an indicator of more severe left atrial myopathy and is also related to the presence of atrial fibrillation (AF) [104]. Tricuspid regurgitation is also commonly associated with AF and also suggests the presence of right HF [47].
●Pericardial disease such as constrictive pericarditis – Echocardiography is helpful in identifying pericardial disease. Further evaluation by computed tomography [CT], CMR imaging, and/or invasive hemodynamic evaluation may be required to confirm the diagnosis of constrictive pericarditis. (See "Constrictive pericarditis: Diagnostic evaluation and management".)
●High-output HF, which may be suggested by the history and physical examination findings – The diagnosis can be confirmed by cardiac catheterization [105]. (See "Causes and pathophysiology of high-output heart failure".)
●Obstructive lesion in great vessel or heart, such as an intracardiac mass or pulmonary vein stenosis – Echocardiography is generally the initial test for evaluation of cardiac masses. Pulmonary stenosis is suggested by symptoms and history (eg, history of catheter ablation near or within pulmonary veins) and generally confirmed by CT or CMR. (See "Cardiac tumors" and "Atrial fibrillation: Catheter ablation", section on 'Pulmonary vein stenosis'.)
●Right HF due to non-HFpEF causes (including right ventricular [RV] infarction, arrhythmogenic RV cardiomyopathy, and pulmonary arterial hypertension not due to left heart disease). These are suggested by echocardiography and some cases require invasive hemodynamic evaluation. (See "Right ventricular myocardial infarction" and "Arrhythmogenic right ventricular cardiomyopathy: Anatomy, histology, and clinical manifestations" and "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults".)
However, many patients with HFpEF display substantial concurrent RV dysfunction, tricuspid insufficiency, and right HF [64,106], or go on to develop right HF that may be difficult to distinguish from the above causes of right HF when followed longitudinally [65]. The development of right HF is associated with worse outcomes. A majority of patients with HFpEF (>80 percent) also display pulmonary hypertension (PH) secondary to chronic elevation in left heart pressures [87], and RV dysfunction associated with PH [64,106] is also common. It is important to distinguish PH caused by HFpEF from other types of PH. (See "Pulmonary hypertension due to left heart disease (group 2 pulmonary hypertension) in adults" and "Clinical features and diagnosis of pulmonary hypertension of unclear etiology in adults".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Heart failure in adults".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Heart failure with preserved ejection fraction (The Basics)")
●Beyond the Basics topic (see "Patient education: Heart failure (Beyond the Basics)")
SUMMARY AND RECOMMENDATIONS
●Heart failure with preserved ejection fraction (HFpEF) is a clinical syndrome in which patients have symptoms of HF, a left ventricular ejection fraction (LVEF) ≥50 percent, and evidence of cardiac dysfunction as a cause of symptoms (eg, abnormal LV filling and elevated filling pressures). (See 'Diagnosis' above.)
●Among all patients with HF worldwide, nearly half have a LVEF ≥50 percent (including those with HFpEF), nearly half have an LVEF ≤40 percent (HF with reduced ejection fraction [HFrEF]), and the remaining 10 to 24 percent have HF with mid-range ejection fraction [HFmrEF; LVEF 41 to 49 percent]). (See 'Prevalence and demographics' above.)
●Patients with HFpEF and those with HFrEF have similar symptoms and signs. Dyspnea and fatigue are by far the most common symptoms (table 3). Signs of HF (such as elevated jugular venous pressure, pulmonary rales, and lower extremity edema) may or may not be present. (See 'Clinical manifestations' above.)
●Evaluation of patients with suspected HFpEF includes identifying and evaluating comorbidities, as these affect management and prognosis. Comorbidities commonly associated with HFpEF include aging, systemic hypertension, coronary artery disease, diabetes and metabolic syndrome, obesity, sleep-disordered breathing, chronic obstructive lung disease, and kidney disease. Occult coronary heart disease is a common and potentially reversible cause of HFpEF. (See 'Associated comorbidities' above.)
●We suggest the following approach for diagnosis of HFpEF (algorithm 1):
•Identification of patients with suspected HFpEF based upon clinical evaluation including history, physical examination, and echocardiography. HFpEF is suspected in patients with symptoms of HF with an LVEF ≥50 percent and no apparent cause of symptoms other than HFpEF. (See 'When to suspect HFpEF' above.)
•Calculation of H2FPEF (Heavy, Hypertensive, atrial Fibrillation, Pulmonary hypertension, Elder, Filling pressure) and Heart Failure Association Pretest assessment, Echocardiography and natriuretic peptide, Functional testing, Final etiology (HFA-PEFF) scores in patients with suspected HFpEF to estimate the probability of HFpEF versus other causes of symptoms [45,78,79]. (See 'H2FPEF score' above and 'HFA-PEFF score' above.)
•Score interpretation. The probability that HFpEF is the cause of symptoms increases with increasing H2FPEF score (range 0 to 9) and increasing HFA-PEFF score (range 0 to 6):
-Low probability – A low H2FPEF score of 0 or 1 and an HFA-PEFF score of 0 or 1 is associated with a low probability of HFpEF. A low score suggests that symptoms are most likely due to a noncardiac cause, and such causes should be investigated. However, if the cause of symptoms remains uncertain after evaluation for noncardiac causes, cardiology consultation and a hemodynamic exercise test is suggested to determine if HFpEF is present. (See 'Hemodynamic exercise test' above.)
-Intermediate probability – H2FPEF and/or HFA-PEFF score ≥2 and neither score is high probability. In this setting, we suggest cardiology consultation and a hemodynamic exercise test. (See 'Hemodynamic exercise test' above.)
-High probability – An H2FPEF score of 6 to 9 or an HFA-PEFF score of 5 or 6 is associated with a high probability of HFpEF and is therefore considered diagnostic for HFpEF.
•On hemodynamic exercise testing (performed in intermediate-risk and only selected low-risk patients), pulmonary capillary wedge pressure ≥15 mmHg at rest or ≥25 mmHg during exercise is diagnostic for HFpEF. (See 'Hemodynamic exercise test' above.)
●When diagnosing HFpEF, it is important to exclude mimics, including non-HF conditions with similar symptoms (such as lung disease) as well as other causes of HF with an LVEF ≥50 percent, such as valvular heart disease, pericardial disease, cardiomyopathies, cardiac amyloidosis, and high-output HF (table 2). (See 'Differential diagnosis' above.)
ACKNOWLEDGMENTS
The UpToDate editorial staff acknowledges Michael R Zile, MD, and William C Little, MD, who contributed to earlier versions of this topic review.
The UpToDate editorial staff also acknowledges William H Gaasch, MD (deceased), who contributed to earlier versions of this topic.
