Fig. 13.1
Left ventricular pressure-volume loops in systolic and diastolic dysfunction (Adapted with permission from: Aurigemma GP, Gaasch WH. Clinical practice. Diastolic heart failure. N Engl J Med. 2004;351(11):1097–105)
However, problems with this terminology emerged during the early use of echocardiography. Some patients with presumed diastolic heart failure failed to manifest clear-cut echocardiographic evidence of impaired left ventricular (LV) filling. The situation became even more complex as clinicians began to realize that many, if not all, patients with systolic heart failure also manifest some echocardiographic features of impaired LV filling. Some patients were described as having classic diastolic heart failure in the absence of echocardiographic evidence of impaired LV filling. In fact, the most consistent echocardiographic finding of patients with HFpEF is an enlarged left atrium .
The typical place for diagnosing diastolic heart failure is no longer the cardiac catheterization laboratory. Echocardiography is now widely used to verify the diagnosis of heart failure, typically dividing it into HFpEF and HFrEF , depending on the clinical and echocardiographic findings. Numerous echocardiographic features help define impaired LV filling and diastolic heart failure (◘ Fig. 13.2) [12–17]. However, none of these features are always present in every patient with HFpEF, and none are entirely reliable with regard to predicting an elevated LVEDP [18–20].
Fig. 13.2
(a–c) Doppler criteria for classification of HFpEF (Adapted with permission from: Redfield MM, Jacobsen SJ, Burnett JC, Jr, Mahoney DW, Bailey KR, Rodeheffer RJ. Burden of systolic and diastolic ventricular dysfunction in the community: appreciating the scope of the heart failure epidemic. JAMA. 2003;289(2):194–202)
Although some investigators prefer the term heart failure with normal ejection fraction (HFnEF ) [8], the term HFpEF has more recently become widely used in the American literature. Most contemporary studies indicate that patients with HFpEF typically have a normal-sized or even small left ventricle internal chamber dimension in diastole [3]. Left ventricular concentric hypertrophy is commonly present, but in some cases, the hypertrophy is eccentric, and other patients do not manifest left ventricular hypertrophy (LVH ) at all.
Left atrial enlargement is very common in patients with HFpEF and is perhaps the first echocardiographic sign to herald the subsequent HFpEF. It often precedes the development of atrial fibrillation which is also common in HFpEF. Typically, the LV ejection fraction in patients with HFpEF is 50 % or greater, although some studies have defined it as 40 % or greater. Grade II or III diastolic dysfunction is usually observed by echocardiography, but this is highly variable and not evident in every case.
Invasive LV pressure monitoring in HFpEF indicates increased left ventricular end-diastolic filling pressure and a shift in pressure-volume relationship (◘ Fig. 13.1). That is, the left ventricular end-diastolic pressure (LVEDP) is higher relative to the LV end-diastolic volume. This change in the relationship between LV pressure and volume is highly characteristic of an increase in LV chamber stiffness (referred to as k, an engineering term for chamber stiffness).
The LV in patients with HFpEF is clearly less distensible, thus allowing for an increase in the LVEDP , despite only a small increase in end-diastolic volume. Not only is the increase in chamber stiffness at the level of the left ventricle but evidence also indicates that the cardiac myocytes themselves are thickened and stiffer and have increased microtubular density. These changes at the cardiomyocyte level may render the cell stiffer. Moreover, increased collagen content is well known to occur in the hypertrophied left ventricle, and this undoubtedly also contributes to the increased chamber stiffness [21].
Classically, patients with HFpEF manifest early exercise intolerance, chronotropic incompetence, micro-endothelial dysfunction, and an inability to increase LV ejection fraction in response to inotropic stimuli such as exercise or dobutamine [1, 22, 23]. In essence, patients with HFpEF show reduced cardiovascular inotropic and chronotropic reserve and an inability to dilate the peripheral microvasculature in response to exercise. Because LVEDP tends to increase in these patients at rest and rises further with exercise, the left atrium (LA) may remodel and increase in size prior to clinical manifestations of heart failure. Atrial fibrillation and flutter are common comorbid conditions in patients with HFpEF and likely further impair diastolic filling, sometimes leading to signs and symptoms of acute heart failure and the need for hospitalization.
Patients with HFpEF manifest LVH, LA enlargement, and diastolic dysfunction. In many cases, LV mass is also increased. The increase in LA size is independently associated with increased morbidity and mortality in patients with HFpEF [24].
Some cardiology thought leaders believe that systolic and diastolic heart failures are different phenotypes of the same syndrome [25, 26]. It is possible that HFpEF and HFrEF are merely extremes in the spectrum of overlapping phenotypes, and therefore, neither LV ejection fraction nor LV cavity dimensions can capture the wide variety of remodeling that goes on during the progression of heart failure. Many investigators, however, consider HFpEF and HFrEF as distinct phenotypes within the heart failure spectrum and that these two phenotypes have different responses to therapy. The latter view is currently most widely held. Others believe there is no intrinsic diastolic property that can explain the occurrence of heart failure with normal EF [27].
Despite this lack of complete agreement regarding the definition of HFpEF, nearly all cardiologists recognize it clinically as a particular type of heart failure that is increasing in incidence and that it is relatively resistant to conventional neurohumoral blocking therapy such as renin-angiotensin-aldosterone inhibitors and beta-adrenergic inhibitors.
It is likely that HFpEF consists of multiple subgroups expressing distinctly different phenotypes and underlying pathophysiologies [6, 28]. This highly variable syndrome and its many clinical varieties are perhaps what lead to confusion about its definition. It is not a single phenotype but has great heterogeneity within itself. These different phenotypes may respond differentially to various therapies [28].
Epidemiology and Comorbid Conditions
In 2010, 1 in 9 deaths in the United States were related to heart failure. About 5.1 million Americans above the age of 20 years have heart failure [29]. Heart failure accounts for 35 % of cardiovascular disease deaths. The prevalence of heart failure will increase by 46 % from 2012 to 2030, resulting in more than eight million people above the age of 18 years with heart failure. Probably more than half of these cases will be patients with HFpEF. It is widely believed that the prevalence of HFpEF is increasing with the aging population [17, 30, 31].
It is also widely agreed that HFpEF is more common in women than in men. The incidence of HFpEF is 6.6 % in women age 65–69 years old and is increased to 14 % in women over 85 years. Risk factors for the development for HFpEF include older age, female gender, hypertension, obesity, obstructive sleep apnea, diabetes, coronary artery disease, and being African American [17, 30]. Normal aging is associated with an increase in LV stiffness, even when blood pressure is controlled and LV mass is reduced [32]. Some experts believe that HFpEF is simply the result of the aging process, so it may not be amenable to conventional heart failure therapy. The overall prevalence of LV diastolic dysfunction as measured by echo in a random sample of the general population is as high as 27.3 % [33]. The so-called preclinical diastolic dysfunction clearly can progress to symptomatic HFpEF and may eventually become a target for therapeutic intervention [34]. About 40 % of all hospital admissions for heart failure are for patients with HFpEF [35]. In one report from 2005 to 2010, the proportion of hospitalizations for HFpEF increased from 33 to 39 % [35]. Two-thirds of patients with HFpEF will develop atrial fibrillation , which greatly complicates the syndrome and leads to frequent hospitalizations [36]. In one report, about 70 % of patients with HFpEF had angiographically proven coronary artery disease (CAD) [37]. Therefore, CAD is common in patients with HFpEF and is associated with an increase in mortality and some deterioration in LV function [37]. Although the prevalence of concomitant CAD may be 70 %, only about 40 % of patients with HFpEF have angina pectoris [37, 38]. Revascularization of patients with HFpEF and coronary artery disease (CAD) appears to improve survival, at least in one observational study [37]. With this information, one needs to consider the use of diagnostic coronary angiography in patients with HFpEF, even though many of these patients are elderly, frail, and poor candidates for diagnostic coronary angiography.
Diabetes mellitus occurs in about 40 % of patients with heart failure. It is common in patients with HFpEF and tends to occur in younger, more obese patients. Patients with HFpEF and diabetes mellitus are more often males and tend to have concomitant hypertension, renal dysfunction, pulmonary disease, and peripheral vascular disease [39]. Patients with diabetes also have more LVH and higher LVEDP than patients with HFpEF and no diabetes. Patients with HFpEF and diabetes have less exercise capacity compared to those who have no diabetes and are also more likely to be hospitalized.
It now seems clear that HFpEF is a syndrome with various phenotypes and is far from being a single clinical entity. It has different etiologies and presents as different subtypes with a wide spectrum of signs and symptoms. Some patients go on to develop pulmonary hypertension and some do not. Some develop symptomatic angina , and some do not. Some manifest concentric LV remodeling, while others do not. Patients appear to progress back and forth across the spectrum, making it unlikely that a single form of therapy will ever emerge as uniformly effective.
The mode of death in patients with HFpEF is cardiovascular in about 60 % of cases, with sudden death and heart failure death being most common [40]. This percentage of cardiovascular death (60 %) is less than what is typically reported in patients with HFrEF. This observation is in keeping with the notion that patients with HFpEF are older and have more non-cardiovascular (non-CV) deaths. The large number of non-CV deaths in patients with HFpEF is likely due to cancer and other maladies of old age. Performing an adequately powered randomized controlled trial of therapy for patients with HFpEF using CV death as an end point would require an unrealistic, large sample size. This may partly explain why clinical trials in HFpEF have failed to uncover successful therapy. It is possible that the lower rate of CV death in the HFpEF population in therapeutic trials to date has largely led to treatment neutrality. Because CV end points are less common in this population of patients, most trials of HFpEF therapy to date likely have been underpowered (◘ Table 13.1).
Table 13.1
Key randomized controlled trials of HFpEF
Trial | Year | N | Ejection fraction | Primary outcome hazard ratio (95 % confidence interval) | Comments |
|---|---|---|---|---|---|
CHARM-PRESERVED | 2003 | 3023 | >40 % | Composite of CV death and HF hospitalization 0.86 (0.74–1.0); P = 0.051 | Significant reduction in HF hospitalization |
Candesartan vs. placebo | |||||
PEP-CHF | 2006 | 850 | Wall motion index of <1.4 equivalent to EF 40 % | All-cause mortality or unplanned HF hospitalization 0.69 (0.47–1.01); P = 0.055 at 12 months | Post hoc analysis showed a trend toward benefit with perindopril at 12 months |
Perindopril vs. placebo | |||||
I-PRESERVE | 2008 | 4128 | >45 % | All-cause mortality or hospitalization for CV cause 0.95 (0.86–1.05); P = 0.35 | None |
Irbesartan vs. placebo | |||||
TOPCAT | 2014 | 3445 | >45 % | Composite of death from CV causes, aborted arrest, or hospitalization for HF 0.89 (0.77–1.04); P = 0.14 | Overall, the trial was considered neutral. However, patients treated in the United States did appear to benefit from this therapy. There appears to be international variation in the response to this therapy |
Spironolactone vs. placebo |
Pathophysiology
General Concepts
HFpEF is no longer considered a single, unitary, pathophysiologic entity that will simply respond to therapies designed to improve increased LV filling pressures [41]. Rather, new paradigms are emerging for treating HFpEF that incorporate a myriad of mechanisms including coronary microvascular inflammation, reduced nitric oxide availability, reduced cyclic guanosine monophosphate (cGMP), reduced protein kinase G, and increased myocyte hypertrophy [42]. The hope is that uncovering new mechanisms will lead to improved therapies for patients with HFpEF. To date, we do not have an effective specific therapy for HFpEF, and the treatment still largely consists of diuretics and management of comorbidities. However, new molecules to treat HFpEF are under consideration and in early stages of clinical trials.
Patients with HFpEF are often elderly women who manifest hypertension, diabetes, and coronary artery disease. However, multiple phenotypes are possible and may include such comorbidities as disproportionate pulmonary hypertension, tricuspid regurgitation, right heart failure, and infiltrative/restrictive cardiomyopathy. It may be because of the wide spectrum of phenotypes that no therapy to date has proven uniformly effective for all patients with HFpEF [6].
Geometric Changes in Chamber Size and Wall Thickness
HFpEF is typically associated with significant LV remodeling that affects the LV and LA chambers, the cardiomyocytes, and the extracellular matrix. When pulmonary hypertension is present, the RV may hypertrophy and dilate, and tricuspid insufficiency may eventually ensue. LV remodeling is associated with normal or decreased LV end-diastolic volume, concentric chamber hypertrophy, increased wall thickness, and increased ratio of myocardial mass to cavity volume—all of which result in a steep diastolic pressure-volume relation (◘ Fig. 13.1) [43–46]. Fifty to 66 % of HFpEF patients have increased wall thickness and mass that is concentric in type [47, 48]. However, LVH is not essential to make the diagnosis of HFpEF, since patients with diabetes, CAD, and advanced age may develop HFpEF in the absence of LVH [11].
Other mechanisms of diastolic dysfunction include deficient early elastic LV recoil, blunted LV lusitropic response, and low LV preload reserve [45]. In normal individuals during diastole, the rapid pressure decay associated along with the “untwisting” and elastic recoil of the LV produces a suction effect that promotes ventricular filling by increasing the left atrium (LA) to LV pressure gradient. This tends to pull blood into the left ventricle. The suction process is augmented during exercise to compensate for the reduced diastolic filling period induced by the associated increase in heart rate. Systolic longitudinal and radial strain, systolic mitral annular velocities, and apical rotation are lower in patients with HFpEF, and these measurements fail to increase normally during exercise. In diastole, patients with HFpEF have reduced and delayed untwisting and reduced LV suction at rest and during exercise. Such changes can be measured using three-dimensional (3D) echo and strain calculations, but special software is required and the calculation can be time-consuming.
Other potential mechanisms contributing to the pathophysiology of HFpEF include [46, 49, 50]:
Asynchrony of relaxation due to regional myocardial infarction
Myocardial ischemia
Chamber hypertrophy
Myocardial and atrial fibrosis
Conduction disease
Geometric changes involving the LV chamber
Microvascular coronary disease
Atrial systolic dysfunction
Atrial diastolic dysfunction
Chronotropic incompetence
Atrial fibrillation
Supraventricular tachycardia
Diastolic dyssynchrony with or without electrical dyssynchrony has been observed in patients with HFpEF [51]. Whether or not dyssynchrony is an important contributor to the pathophysiology of HFpEF remains uncertain. It should be noted that cardiac resynchronization therapy (CRT) in patients with HFrEF and QRS duration of less than 130 ms does not reduce the rate of death or hospitalization and may increase mortality [52].
Cardiomyocyte and Extracellular Matrix
Patients with HFpEF have thicker cardiomyocytes with little or no change in cardiomyocyte length. This may result in decreased length/width ratio. Borbély et al. [44] have demonstrated thicker and shorter myocytes with increased myofibrillar density in patients with HFpEF compared to those with HFrEF. These myocyte changes correspond to the increase in LV wall thickness but appear not to alter LV volume. Although there is an increase in the amount of collagen in hearts of both patients with HFpEF and HFrEF, the thickness of the collagen bundles and the continuity of the fibrillar components of the extracellular matrix surrounding the cardiomyocyte are greater in HFpEF [46, 53].
Hemodynamic Abnormalities
Classically, patients with HFpEF manifest increased filling pressure either at rest, with exercise, or volume challenge. The elevated LVEDP is likely responsible for some of the dyspnea that the patients experience either at rest or with exertion, but this is not the complete story. Changes in active and passive relaxation of the LV also occur, which lead to alterations in negative dP/dt, or tau. The change in the pressure-volume relationship is a product of the increased stiffness of both the LV chamber and the abnormal myocyte structure and function.
There may be a molecular basis for the increased myocyte stiffness that is related to phosphorylation and dephosphorylation of the large intracellular cardiac myocyte molecule called titin [44]. A number of factors, including titin isoform switches (to a less compliant N2B isoform) and titin phosphorylation state, can affect passive myofiber stiffness, and perturbation in both has been described in HFpEF. Interaction of titin with other signaling molecules and with ion channels may also contribute to the effect of titin on diastolic stiffness. It is believed that alterations in titin may occur in concert with changes in the extracellular matrix, but this interaction is not well defined. Unfortunately, our understanding of the role of alteration in titin and titin interactions in patients with HFpEF is limited, and much remains to be clarified [44, 54].
Neurohumoral Abnormalities
The critical role of neurohumoral factors in regulating circulation and volume status in normal people and in patients with heart failure and reduced ejection fraction has been known for more than a century. The importance of neurohumoral activation in HFrEF pathophysiology was identified in the middle and latter part of the twentieth century and formed the basis for the neurohumoral hypothesis of heart failure [55, 56]. Major contributions to this understanding came from observational data collected in large, well-characterized populations of patients participating in systolic heart failure treatment trials [55, 57]. In many of these studies, plasma levels of vasoactive hormones including norepinephrine, renin, angiotensin II, aldosterone, vasopressin, and atrial natriuretic peptide were measured and found to be significantly increased compared with normal controls. These observations allowed for important insight into both the pathophysiology and the prognosis of patients with HFrEF [55, 57–59].
It is therefore well known that neurohumoral activation plays a fundamental role in the progression of HFrEF. Activation of the sympathetic nervous system, aldosterone, and the natriuretic peptide system is also known to occur in patients with HFpEF [60–62]. Although natriuretic peptide levels are lower in patients with HFpEF compared to patients with HFrEF, the plasma levels of catecholamines and aldosterone may be similarly increased. Whether other counter-regulatory hormones such as renin, angiotensin II, and endothelin are playing an active role in the pathophysiology of HFpEF remains to be determined. To date, clinical trials using angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers, and mineralocorticoid receptor blockers have failed to demonstrate a survival benefit in patients with HFpEF (◘ Table 13.1).
Risk Factors for HFpEF
Gender
Along with age, female gender is a potent risk factor for the development of HFpEF. Indeed, there appear to be important age-gender interactions, such that the prevalence of HFpEF increases more sharply with age in women than the prevalence of HF with a reduced EF [63, 64]. The reasons for the female predominance in HFpEF are not entirely clear, but women have higher vascular, LV systolic, and diastolic stiffness than men. Vascular and ventricular stiffness increases more dramatically with age in women [64]. Women develop greater wall thickness than men in response to an afterload stress such as aortic valve stenosis [65]. That is, for the same degree of aortic stenosis or hypertension, women may have more LVH than men.
Unique coronary vascular functional changes in women may also play a role in the pathophysiologic process of HFpEF. Microvasculature dysfunction, including inability to dilate the microvasculature in response to exercise, is an important feature of HFpEF. It is of interest that HFpEF is more common in women, as is coronary microvascular disease.
Age
Although cardiovascular disease may contribute to diastolic dysfunction in older people, studies have also suggested that the ability to fill the left ventricle during diastole deteriorates with normal aging [64]. The speed of left ventricular relaxation declines with age in men and women, even in the absence of cardiovascular disease.
Vascular, LV systolic, and LV diastolic stiffness increases with aging [64, 66]. Increases in vascular stiffness are likely related to effort intolerance in patients with HFpEF [23]. Structural cardiac changes with aging (e.g., increased cardiomyocyte size, increased apoptosis with decreased myocyte number, altered growth factor regulation [67], focal collagen deposition) and functional changes at the cellular level involving blunted beta-adrenergic responsiveness, excitation-contraction coupling, and altered calcium-handling proteins may also contribute to diastolic dysfunction with normal aging. At least one study has suggested that prolonged, sustained endurance training may preserve LV compliance with aging and help prevent HF in the elderly [68].
Growth differentiation factor 11 (GDF 11) has recently been shown to circulate in young mice and reverse age-related cardiac hypertrophy when there is cross circulation with an older mouse (parabiosis) [67]. This striking observation suggests that aging plays a prominent role in the development of cardiac hypertrophy and stiffness and that this process is potentially reversible [69, 70].
Diabetes Mellitus
Nearly half the patients with heart failure have diabetes mellitus , usually type II, and patients with HFpEF are no exception [39]. Diabetes is a potent risk factor for the development of heart failure . The prevalence of diabetes is seemingly similar in patients with HFpEF and HFrEF, suggesting that diabetes contributes to the pathophysiology of both [39]. Although diabetes predisposes to coronary artery disease, renal dysfunction, and hypertension, numerous studies suggest that there is a direct effect of diabetes and hyperglycemia on myocardial structure and function [71–73]. Myocardial contractile dysfunction in patients with diabetes is likely related to worsening cardiac mitochondrial function and mitochondrial dynamics [74]. These changes in contractile function observed in diabetic patients are not seen in obese patients at an early stage of insulin resistance [74].
Mechanisms responsible for increased diastolic stiffness of the diabetic heart are perhaps different than those changes found in patients with HFrEF. Altered cardiomyocyte resting tension is more important in HFpEF, whereas fibrosis and advanced glycation end products are more important in HFrEF [75]. Increases in passive trans-mitral LV inflow velocity measured by Doppler in diabetic patients are associated with the development of HFpEF and increased mortality, independent of hypertension or coronary artery disease [76]. Diabetic patients may develop cardiac steatosis , which can precede the onset of glucose intolerance and left ventricular dysfunction [73]. Lipid overstorage of human cardiomyocytes can be an early manifestation of type 2 diabetes mellitus and can be evident before the development of heart failure [73].
Other morphologic changes in the diabetic heart include myocyte hypertrophy, increased extracellular matrix (fibrosis), and intramyocardial microangiopathy. Functional changes, which may represent a continuum, include impaired endothelium-dependent and endothelium-independent vasodilation, impaired LV relaxation, increased passive diastolic stiffness, and contractile dysfunction.
Atrial Fibrillation and Frequent Premature Atrial Contractions
Atrial fibrillation is recognized as a frequent precipitant of acute heart failure in patients with HFpEF. Whereas atrial fibrillation may cause decompensated HF in patients with diastolic dysfunction, it is also true that diastolic dysfunction can provoke atrial fibrillation [77]. Thus, diastolic dysfunction, atrial fibrillation, and HFpEF are common and interrelated conditions that probably share common pathogenic mechanisms in the elderly. High-density premature atrial contractions, previously considered a benign entity, are now understood to likely be a precursor to the development of atrial fibrillation and may signal the forthcoming onset of atrial fibrillation and perhaps even stroke [78]. Excessive supraventricular ectopic activity is defined as ≥30 premature atrial contractions per hour or any episode of runs of ≥20 premature atrial contractions. To date we have insufficient data to know if treating this risk factor is appropriate or not.
Coronary Artery Disease
The reported prevalence of coronary artery disease or myocardial ischemia in patients with HFpEF varies widely, although a recent report of unselected patients using diagnostic coronary angiography indicated that 70 % of patients with HFpEF have significant underlying CAD [37, 79]. Although acute ischemia is known to cause transient diastolic dysfunction , the role of chronic coronary artery disease and myocardial infarction in patients with HFpEF is not precisely defined. Clearly, increased myocardial fibrosis from previous infarctions could contribute to the diastolic dysfunction. Patients with HFpEF and severe CAD may benefit from revascularization [37, 80].
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
