To diagnose heart failure with preserved ejection fraction (HFpEF), the following three criteria have to be fulfilled [1–8]:

(To fulfill criteon 3, the European Society of Cardiology asks for the following two to be present: elevated natriuretic peptides and either (I) proof of a relevant structural cardiac abnormlity (indicated by an enhanced LA size, LALVI, or a left ventricular muscle mass, LVMI, above the normal range), or/and (II) proof of abnormal diastolic properties, diastolic dysfunction [3]).

The criteria defining the syndrome used by authors based on the latest ACCP/AHA [4] and ESC guideline [3] have merged closer together, particularly the range of LV-EF. However, in the most recent, 2016 guideline, the ESC definition demands increased natriuretic peptide serum levels in addition to either diastolic dysfunction and/or signs of a structural heart disease [3], thus strengthening and appreciating the importance of biomarkers and structural abnormalities.

HFrEF (heart failure with reduced ejection fraction) is indicated by signs and symptoms typically present and constituting heart failure, and a LV-EF < 40% [2–4].

Furthermore, recently both the AHA/ACCP and the ESC introduced a mid-range (HFmEF) [3] or borderline [4] type, a group with a LV-EF between 40 and 49% (41–49% ACCP/AHA) but otherwise featuring all other HFpEF criteria.

Specific diagnostic criteria (read below, Sect. 5.5) delineate exactly the findings and parameters indicative for a structural heart disease and/or suggestive for diastolic dysfunction.

HFpEF is a considerably complex malady [8, 10, 11] of broad phenotypic heterogeneity [12, 13], and multi-facet pathophysiology [9, 13–15], may potentially afflict various organs [13, 15], and mostly goes without a specific etiology but with miscellaneous pathogenetic underlying causes [13, 16–18]. Its clinical spectrum typically varies from dyspnea on exertion to even acute pulmonary edema [19–21]. Since diastolic dysfunction (DD) is a dominant, if not the dominant [9] feature of this disorder [11, 22–24], taking a key role in HFpEF pathophysiology [22, 25, 26], HFpEF has frequently been referred to as “diastolic heart failure” (in contrast to “systolic heart failure” or HFrEF) in the past [8, 27]. Indeed, more than 2/3 of all patients with HFpEF display DD at rest [28–30], during stress even up to 80–90% are found to develop abnormal diastolic properties [31]. Hemodynamically, DD impairs ventricular filling [32, 33] with a higher LVEDP for any given end-diastolic volume [34].

However, meanwhile it is quite clear that DD is not a unique finding in patients previously classified suffering from diastolic heart failure, but also occurs in patients with “systolic” heart failure (heart failure where the ejection fraction is reduced) [2, 33, 35, 36], is present in many asymptomatic elderly (60–80%) suffering from hypertension [37–40], and even more, altered diastolic properties are a very common and arguably physiological observation in elderly individuals associated with the aging process [37–39, 41–44]. Consequently, HFpEF is known to predominantly afflict older hypertensive patients [45].

Traditionally, DD has been considered to be an important intermediate step in the development of HFpEF, notifying, if displayed, that hypertension (HTN)/hypertensive heart disease (HHD) may progress to heart failure with preserved ejection fraction [46, 47], and chronic hypertension was thought to potentially turn into HFpEF [48, 49]. Meanwhile, “hypertension is neither necessary nor sufficient for HFpEF development” as Desai writes [48]. Many clinical conditions, myocardial as well as non-myocardial ones, are known to be associated with and may predominantly cause (acute) heart failure with normal ejection fraction, including valvular heart diseases, congenital heart disease, pericardial disease and primary (isolated) right heart failure with basically normal systolic LV function [50–52], whereupon abnormal diastolic function is the most common pathophysiology applying in these cases [51, 52].

Also, quite a number of other features have been acknowledged to be present and contribute to the pathobiology of HFpEF such as impaired LA-function [53], chronotropic incompetence [54, 55], right ventricular dysfunction and pulmonary hypertension [56–58], and even limitations in LV systolic capabilities are present in patients with HFpEF [59, 60]. Moreover, modifications and abnormalities of “extracardiac” features may arise in HFpEF patients being crucially involved, including altered vascular properties affecting LV afterload and ventricular—vascular coupling conditions [53, 54, 61–63], changes in preload circumstances (circulatory volume overload) [64], neuroendocrine activation [65], as well inflammation/endothelial dysfunction [66, 67], and impaired peripheral vasodilator reserve [54, 68, 69].

As such, a very heterogeneous group of patients with different etiological features and several pathophysiological mechanisms applying and contributing may display the syndrome of HFpEF [70]. Consequently, recent findings and facts, recognizing that diastolic dysfunction is not the only underlying abnormality in this syndrome, have led to change the term diastolic heart failure, which implies a single operating pathophysiology [71].

Moreover, the initial consideration, HFpEF may be a precursor of HFrEF, being part of the same disease process, which may potentially step forward to HFrEF, and in which HFpEF and HFrEF indicate the two extremes within a continuum of a single disease [72, 73], has been abandoned due to a lack of evidence [9, 13, 15, 33, 74–76]. Of course, there may be overlaps as some patients with HFpEF are shown to lose up to 5.8% of their EF per year finally ending up with an EF < 50% (40%), while those with reduced EF may show improvements [74]. It is assumed and very likely that a transition from HFpEF to HFrEF may, in turn, occur due to additional adverse events, particularly intercurrent myocardial ischemia and infarctions causing loss of cardiomyocytes [77].

Nevertheless, all available evidence strongly suggests to consider HFpEF as a separate, distinct entity which has to be distinguished from HFrEF: The two syndromes differ in elementary issues of the pathogenesis and pathophysiology, in their etiologies, clinical and demographic characteristics, structural (cardiomyocyte hypertrophy [78] and myocardial fibrosis of varying degree [79]) and functional (cardiomyocyte stiffness [78, 80]) features, time to clinically overt malady, neuroendocrine response and biochemical parameters, associated co-morbidities and, of great importance, in their response to therapy [11, 14, 27, 66, 78, 81–84]. While HFpEF is basically attributed to endothelial dysfunction, HFrEF has to be considered as a disorder of the cardiomyocytes [27].

At least 50% of all patients presenting signs and symptoms of heart failure have a normal or only minimally impaired global systolic LV function, thus suffer from HFpEF [34, 81, 85–89]. Moreover, Owan recognized that the occurence of HFpEF in all heart failure cases (HFpEF and HFrEF) increased from 38% to 54% within the last two decades [81]. Indeed, compared to HFrEF, the relative prevalence of HFpEF is increasing by 10% per decade [8, 81, 87, 90, 91], and the “true” prevalence of HFpEF in the general population is estimated at 1–5.5% [92].

HFpEF surely is a disorder of the elderly as its proportion is increasingly found with older ages [81, 85, 86, 88]. Although elderly women seem to be more afflicted in US [18] and Euopean surveys [93, 94], internationally a more balanced sex distribution appears to exist [95–97]. Comorbidities typically and highly prevalent in and associated with HFpEF (though also related to increasing age) include hypertension (60–80%), obesity (41–46%), diabetes mellitus (13–76%), coronary artery disease (20–76%), atrial fibrillation (15–41%), impaired renal function (40–55%), and hyperlipidemia (16–77%) [81, 85–87, 91, 98–101].

Readmission rates add up to nearly 30% within 60–90 days after discharge [102] and to roughly 50% within 1 year [103].

Mortality rates recently reported in the literature describe in short term (30–90 days) 5–9.5% deaths [86, 87] , 29% deceased patients after 1 year since diagnosed and 68% (55–74%) after 5 years [87, 88, 91]. As such, the prognosis of HFpEF is definitely similar to, and as grim as, those found in patients with HFrEF (32% after 1 and 68% after 5 years) [85–88, 104]. However, in contrary to patients with HFrEF, the reasons of mortality in HFpEF are more often due to non-heart failure cardiovascular issues [18, 105, 106], reaching 40% of the causes of death [107, 108].

Consequently, in the majority of patients with HFpEF, a specific etiology cannot be determined [9, 13, 16, 83], rather, “HFpEF occurs most commonly in the elderly who have one or more co-morbidities like hypertension, obesity, diabetes, metabolic syndrome, chronic kidney disease, atrial fibrillation, and/or anemia” [109]. As such, the co-morbidities exert a considerable impact on the pathogenesis of HFpEF [9, 13, 17, 42], and HFpEF may be considered to be the “identical” clinical result of different diseases with diverse and miscellaneous underlying pathophysiologies [7]. Nevertheless, in some cases a (more) specific cause, usually provoking diastolic dysfunction and concomitant/consecutively HFpEF, may be identified as in case of hypertrophic, restrictive, infiltrative, or genetically determined cardiomyopathies as well constrictive pericarditis or cardiac fibroelastosis [50, 51, 83].

Heart failure with preserved ejection fraction, accounting for more than 50% of all heart failure cases [81, 89], is henceforth recognized as a separate and discrete clinical syndrome rather than a “milder form” and/or precursor of HFrEF as growing evidence clearly indicates [73].

Exercise intolerance with often severe dyspnoea on exertion and acute pulmonary edema are the key clinical pictures HFpEF patients present [19–21]. 2/3 of all HFpEF patients feature LV diastolic dysfunction at rest [28, 110], however up to 80–90% may display abnormal diastolic properties during stress [31]. Accordingly, LV diastolic dysfunction, as a central factor in the pathobiology and a patho-physiological hallmark of HFpEF [22, 24, 25], evokes, either alone or in combination with other pathophysiological features [1, 22, 25], the phenotypic, clinical appearances and the elevated filling pressures (a general finding in any kind of heart failure [111]) present in this syndrome [22]. The other features include combined ventricular-vascular stiffening (notably enhanced central aortic stiffening and (consecutively) blunted ventriculo–arterial coupling) [55, 61, 62, 68], impaired systemic vasodilator reserve [24, 54], systolic limitations [49, 112, 113], and extra-cardial causes like volume overload [114] and pulmonary hypertension [56, 58, 115] with subsequent ventricular, mostly diastolic interactions [41].

LV diastolic dysfunction underlying HFpEF is, in the absence of pericardial and endocardial disease [116], attributed to abnormal diastolic myocardial stiffness [8, 116, 117]. Diastolic myocardial stiffness is determined by (a) the composition, functional status and the amount of the extracellular matrix (ECM) and by (b) the cardiomyocytes, accurately the cardiomyocyte tension, respectively the cardiomyocyte stiffness which is largely defined by the functional and structural properties of the cytoskeletal giant protein titin [8, 118]. While originally the diastolic passive myocardial and the overall diastolic chamber stiffness have primarily been assigned to be predominantly determined by the collagen quantity and quality of the ECM [14] and by collagen crosslinking [119, 120], most recent study results revealed that cardiomyocyte stiffness alone has the capability to induce HFpEF without any involvement of the ECM [121]. This is in line with data demonstrating that 1/3 of HFpEF show normal collagen volume fraction although similar LV stiffness and end-systolic wall stress [80]. Meanwhile, several studies on HFpEF patients clearly relate enhanced diastolic LV stiffness to elevated cardiomyocyte stiffness [122–124].

Cardiomyocyte tension and stiffness are largely modulated by titin [125]. Changes in cardiomyocyte properties are reported to possibly occur in the acute setting attributed to alterations in phosphorylation status of titin (relative hypo-phosphorylation) and intramolecular disulfide bridging (both energy-consuming processes), associated and in conjunction with acute energy deficits [55]. As a result, an acute increase in passive LV diastolic stiffness ensues [126] causing acute cardiac failure [127]. In contrast, the collagen turnover and thus modification may take considerably longer with a known collagen half-life of 80–120 days [128]. Accordingly, increased myocardial stiffness and tension, predominantly caused by cardiomyocyte properties, may arise acutely, whereas alterations of the ECM indicate long-term and chronic changes.

The majority of individuals with DD will never develop symptoms [129], however, worsening diastolic function is identified to decisively contribute to the onset of clinical heart failure symptoms [130]. The transition from compensated conditions to overt HFpEF is reported to be related to profound myocardial stiffening [131, 132]. Drazner [133] recently illustrated in his paper on “the progression of hypertensive heart disease”, that both, (a) the progressive and adverse change of ECM composition and amount [106, 134, 135] enhancing myocardial stiffness [136] in patients suffering from hypertensive heart disease and (b) the (accompanying) increase in LV filling pressures [53, 130, 137], are causally responsible for the transition from HHD to HFpEF—indeed, ventricular passive stiffness substantially impacts LV filling pressures [22, 25]. However, other factors affecting LV-filling pressure such as PH and (subsequently influencing) ventricular interdependence, (consecutive) atrial dysfunction and vascular components, notably enhanced central aortic stiffness [21, 48, 62], may decisively contribute as well [56, 62, 138].

Intermittent or permanent increases in LVEDP potentially facilitating left-atrial dilatation and atrial fibrillation (thus atrial dysfunction) [138], and elevated pulmonary pressures are indicative for clinically relevant DD [31].

Traditionally, DD has been considered to be an important intermediate step in the development of HFpEF, occurring in patients with hypertension/hypertensive heart disease developing heart failure [46, 47], and chronic HTN was supposed to potentially turn into HFpEF [48, 49]: Hypertension has been viewed as being the “predominant factor in the development and the progression to and of HFpEF” [139]. HTN is found to be present in 60–80% of all patients diagnosed with HFpEF [81, 98]. Cellular and extracellular structural and functional changes as well as adaptions are demonstrated in the myocardial tissues and in cardiac function of HTN patients subsequently developing DD [78, 140] and HFpEF [98, 141, 142]. Even mild hypertension can result in DD [143]. As such, chronic pressure overload (e.g. HTN) is recognized to be a leading risk factor and cause of DD [92, 144] and of HFpEF [141, 145].

This prevailing mechanistic view of the syndrome of HFpEF based on classical, traditional knowledge and perceptions (mechanical/neuroendocrine model of heart failure [146]) received even more support by recent analyses and study results enlarging the existing concept by, notably, central and peripheral vascular and v-a-coupling issues (“HFpEF is recognized as a disease of abnormal v-a-coupling” [147]), consecutive and associated PH and ventricular interactions, all potentially influencing and contributing to the pathophysiology and pathobiology of acute heart failure [14, 21, 41, 56, 61, 148]. Furthermore, these features fit very well into the recently provided concept by Cotter, assigning acute heart failure either to a predominantly acute vascular or to a prevailing cardiac, acutely decompensating disorder [149, 150]. However, often both conditions are contributing with only one prevailing [149, 150]. These findings emphasize that the pathophysiology of heart failure is heterogeneous, the syndrome of acute heart failure complex and the disorder obviously of systemic dimension [18].

Anyhow, in recent years, a bundle of considerable evidence, strongly linking HFpEF to systemic inflammation, has been established [66, 67, 151, 152]. Significantly elevated, high levels of pro-inflammatory cytokines and other markers of activated inflammation including tumor necrosis factor alpha (TNFα), several interleukins such as IL-1, IL-6, monocyte chemotactic/chemoattractant protein 1 (MCP1), adhesion molecules such as intercellular adhesion molecule-1 (ICAM1) and vascular cell adhesion molecule-1 (VCAM1), and CRP, at least hsCRP (high sensity), released by immune-competent cells (neutrophil granulocytes, monocytes, macrophages, T cells), but endothelial cells and even vascular smooth muscle cells as well [141], are consistently laboratory-confirmed assured in blood samples (and thus within the systemic, peripheral circulation) of heart failure patients [153–156]. Being further of substantial prognostic relevance, these inflammatory mediators, and thus inflammation as such, are considered as being crucially implicated in the disease process [152]. Indeed, increased levels of inflammatory features are independently associated with asymptomatic diastolic dysfunction [157], and repetitive and progressive inflammatory episodes are demonstrated to be strongly associated with the progression of ventricular diastolic dysfunction to HFpEF [154, 158]. Furthermore, a recently published study provides distinct evidence that systemic inflammatory conditions are predictive of incident HFpEF [151], a strong sign of a causal impact of inflammation on the aetiopathogenesis of HFpEF [141, 159].

Moreover, HFpEF, a disease of the elderly [89, 144], is typically accompanied by a range of comorbidities including arterial hypertension, obesity, diabetes (as a rule type II), metabolic syndrome, coronary artery disease, chronic kidney disease, and COPD [85, 86, 142, 160]. All these disorders have been identified as being risk factors for, and precursors of, incident heart failure [161–164]. Furthermore, these maladies are independently associated with early development of diastolic LV dysfunction [165–168]. All these pathologies deploy low grade systemic inflammation [66, 141, 151, 169, 170]. “HFpEF is, compared to asymptomatic patients although as well suffering from obesity, diabetes, HTN, etc., characterized by an increase in cardiac inflammation” [66]. Moreover, metabolic risk factors are not only strongly associated with inflammation, but also with endothelial dysfunction, oxidative stress, impaired myocardial energetics, abnormal cardiomyocyte Ca-handling, reduced NO bioavailability, and maladaptive cardiac remodelling [171–173].

Inflammation per se is a protective response to physiological and unphysiological stimuli, injuries and insults of any kind, e.g. infection, and applies by interactions between cell surfaces, extracellular matrix, and pro-inflammatory mediators [174]. It is basically a vascular answer to any stimulation or threat [175, 176]. Although traditionally considered to be a local process, inflammation may potentially enlarge to a systemic condition [177]. Janeway and Travers state: “The inflammatory response has to be recognized as a systemic process rather than “purely” a local reaction” [178].

Inflammation is inevitably associated with endothelial activation and dysfunction: Endothelial cells are recognized to considerably participate in the initiation, maintenance, and amplification of inflammatory processes [179, 180] and as such, endothelial cells are an integral component of the early innate immune response (conditional innate immune cells) to injury of any kind [181]. The distinct and very close correlation between inflammation and endothelial dysfunction is well established [182]. Inflammation causes endothelial dysfunction [112, 183, 184], subsequently, the dysfunctional endothelial cells display a number of features contributing to and, in turn, amplifying the inflammatory process [181].

Endothelial dysfunction (ED) refers to an “activated” endothelium denoting a maladaptive response to pathological stimuli [185]. Thus, systemic inflammation potentially affects the whole body, more accurately is likely to activate the endothelium of the whole body including the coronary microvasculature and central cardiac endothelium, e.g. endomyocardium [66, 146].

Indeed, cumulating evidence indicates that the inflammatory condition and the endothelial dysfunction [182, 186] must be central and crucial features in the pathobiology of HFpEF [24, 66, 67]. Endothelial dysfunction is associated with cardiovascular diseases, e.g. coronary artery disease, hypertension, diabetes, chronic renal disease, and noticed as a systemic disorder [187–190]. As a result of accumulated co-morbidities, the unifying affection acknowledged and with considerable implication in the pathobiology of HFpEF is endothelial dysfunction (ED) [48]: Comorbidities present in HFpEF lead to ED [117].

Compared to age-matched controls, patients with HFpEF display ED, and ED is related to adverse outcome [67]. Thus, the endothelium takes a central position in the (inflammatory) response, coordinating and “orchestrating” the reply and the reactions to the metabolic, biomechanical, and chemical threats provoked by the co-morbidities [179, 180, 191].

The cardiac endothelial tissue encompasses the endocardium, the intramyocardial capillaries, and the endothelial cells of the coronary microvasculature [117]. The central endothelium, comprising the vessel network of heart and the pulmonary blood flow path, constitutes the largest endothelial surface of the body [192], decisively contributing to the development of heart failure with preserved EF [192]. Endothelial cells are capable to communicate bidirectionally [193, 194]. The cardiac endothelium is demonstrated to modulate cardiac performance [195] since it affects, by autocrine/paracrine signalling (by releasing factors such as NO, ET-1, and natriuretic peptides), the contractile properties [196]. The acute cardiomyocyte function decisively depends on cardiac endothelial cell condition and function [195, 196]. Accordingly, the influence of the endothelial cells on different cardiac cells emphasizes the importance of ED in and the impact of ED on the pathobiology of HFpEF [192, 197]. The “systemic” ED and especially the coronary microvascular endothelial inflammation (see below the new concept by Paulus and Tschoepe, see Fig. 5.4) are not only important bystanders of HFpEF, but play a pathophysiologic relevant and causative role in that syndrome [159, 192].

Hence, the comorbidities commonly seen in patients suffering from HFpEF induce a systemic inflammatory state and as such will afflict the central endothelial cells of the coronary microvasculature and of the endocardium causing ED as clearly evidenced by histologic-bioptical studies [66, 123]: The systemic inflammation is suggested to gradually affect (inflame) the cardiac microvasculature [66], causing ED [197, 198] and subsequently impacts on the interaction between cardiac endothelium and the cardio-myocytes [66, 117, 199, 200], so that finally the myocardium may be inflamed [66, 199]. The expression of adhesion molecules [66] facilitates the recruitment, activation, and transendothelial migration of inflammatory cells into the vessel walls and the myocardium [66]. The conversion of fibroblasts into myofibroblasts, which significantly affect ECM composition, collagen synthesis and collagen deposition in the interstitial cardiac tissues, promoting myocardial fibrosis, is stimulated [201] and accompanied by DD [202, 203]. The amount of cardiac ECM and the collagen quantity and composition influence and co-determine chamber stiffness [103], and a correlation between both, myocardial collagen and the amount of inflammatory cells, and diastolic dysfunction could be established [66]. Activated myofibroblasts, for their part, provoke and maintain inflammation by producing chemokines and cytokines stimulating the inflammatory cell recruitment and ED, thus contribute to establish a vicious cycle maintaining and even fuelling the inflammatory and associated processes [204].

The most important biological consequence of ED certainly is the impaired NO bioavailability [117, 159]. Particularly caused by oxidative stress, hyperglycemia following insulin resistance (IR), components of the activated RAAS (namely A II) and by TNFα [117, 205], the limited NO availability will lead to substantial consequences: The dysfunctional endothelial cells can offer the adjacent cardiomyocytes only a markedly diminished NO supply, this results in disrupted NO-cGMP-PKG signalling (more detailed in the paragraph on pathophysiology), leaving titin hypophosphorylated [78, 122, 205, 206] and facilitates disulfide bridging within the titin molecule [207]. Histologic-bioptic samples of patients suffering from HFpEF revealed reduced PKG activity and low cGMP concentrations in their myocardial tissues, associated with markedly enhanced cardiomyocyte stiffness [205]. Titin decisively determines the elastic properties of the heart [78]: Myocardial and chamber passive diastolic stiffness, crucially determining LVEDP, are largely shaped and assigned to the properties of the giant sacromeric cytoskeleton protein titin [125, 208], notably in normal sized heart chambers as typical in HFpEF [121, 209–211]. Elevated diastolic LV stiffness causing DD is basically attributed to elevated intrinsic cardiomyocyte stiffness as numerous studies reported [80, 122–124, 212]. We have substantial evidence indicating that “stiffened” titin alone is able to induce DD and HFpEF [210, 213], independent of ECM and thus myocardial fibrotic state [210].

As such, acutely altered titin stiffness as in energy deficit [55] following acute (myocardial) ischemia with subsequent increase in LVEDP [53, 126, 130, 137] causing acute cardiac failure [127], may be likewise understood as a (predominantly) cardiac reason for acute heart failure in terms of Cotter’s concept [149, 150]. On the other hand, acute elevations of blood pressure predominantly acting on loading conditions [61, 62, 214] and consecutive (sometimes disproportionate) increases in LVEDP [61–63] may also precipitate acute heart failure, but as a result of primarily acutely changed vascular properties provoking an acute afterload mismatch [61, 149, 150].

Reduced NO bioavailability and disrupted NO-mediated signalling pathways and the increased formation of oxidative stress associated with the features activated, are well implicated in the pathobiology of heart failure [215, 216]. Oxidative stress of the coronary microvasculature reduces NO bioavailability, cGMP content, and PKG activity in the adjacent cardiomyocytes [17].

The metabolic syndrome, a cluster of metabolic factors, notably obesity, but even the principally physiologic aging process [217] are all strongly related to insulin resistance (IR) and enhanced oxidative stress, provoking adverse synergistic effects on myocardial structure and function [218]. Obesity, diabetes (type II), and IR are all reported to exert direct adverse effects on the myocardium independently of confounders like HTN or coronary artery disease [171–173]. These co-morbidities present in HFpEF are independently associated with early DD [165–168] and have been prospectively identified as precursors of incident heart failure [161–164]. Hence, metabolic disorders may contribute via enhanced myocardial inflammation, oxidative stress, downregulated NO bioavailability affecting the very important signalling NO-cGMP-PKG pathway, and limited bioenergetics to DD and HFpEF development [55, 127].

The joint detection of soluble ST2¹ and PTX ₃ ² within the blood stream, indicative for a systemic vascular inflammation in the presence of myocardial wall stress, is reported to correlate well with DD and HFpEF, hence substantiating that indeed inflammation is potentially a causal feature of HFpEF [221].

The association between the soluble TNFα type 1 receptor, a marker of systemic inflammation, and incident HFpEF found in elderly individuals further contributes to assume a causal role of inflammation in that type of heart failure [222]. High grade evidence comes from a study by Kalegeropoulos [151] since the results verify that systemic inflammation, induced by the co-morbidities observed in HFpEF, reflected by high levels of inflammatory markers in the circulation including the classical agents TNFα and IL-6, is predictive for incident HFpEF (but not for HFrEF and as such likewise indicating that both disorders are different entities). As the correlation demonstrated persists even after correcting for known heart failure risk factors (co-morbidities, etc.), these study findings are highly suggestive for a direct, causal role of inflammation in the pathogenesis of HFpEF [151].

Hence, it has been inevitably and necessary that Paulus and Tschoepe implemented a novel paradigm of the pathobiology of HFpEF: Their concept applies systemic inflammation as fundamental in the pathophysiology of HFpEF [159, 199]. The common co-morbidities including HTN, diabetes, and obesity associated and observed with HFpEF, cause a marked systemic inflammatory state, thereby also severely affecting the endothelial layers of the cardiac vessel system and even the endocardium, and thus provoke coronary microvascular, endocardial and (consecutively) myocardial inflammation and dysfunction [66, 123]. ED ensues and as a result of inflammation [123, 180, 197, 198], cardiomyocyte stiffening with subsequent DD develops [80, 123, 126, 210, 213] and ECM remodelling arises, leading to myocardial fibrosis, accompanied by DD [66, 120, 202, 203]. Accordingly, ED and microvascular, especially coronary microvascular disease are not only important bystanders of HFpEF but play a pathophysiologic relevant and causative role [159]. For further details of this concept, please see paragraph on special pathophysiology. The results of several animal studies nicely fit and support this new view of inflammation-induced HFpEF [223, 224].

However, HFpEF is not merely a conglumerate of co-morbidities [75, 225]. A study by Mohammed revealed that HFpEF patients, compared to healthy and hypertensive controls, feature more cardiovascular abnormalities than the individuals in either control group (healthy individuals and hypertensive persons), even after adjusting for comorbidities, sex and age [75], a result which is comprehensible and coherent. Furthermore, the outcome of HFpEF is demonstrated to be worse compared to patients with various comorbidities but with no evidence for heart failure: The mortality rates in the HFpEF group added up to 53–76 per 1000 patient-years while in the matched (correcting for age, sex and comorbidity allocation) control groups without HF, the mortality rate ranged between 11 and 47 per 1000 patient-years [226]. However, that difference was present although the burden of co-morbidities was lower in the HFpEF cohort [226]. Hence, those findings strongly suggest that HFpEF is not simply a collection of co-morbidities, but rather an independent entity [82]. Moreover, the transition to and deterioration in symptomatic HFpEF is related to additional pathobiological issues affecting the functional and structural myocardial status, including v-a-coupling disorders, neuroendorine activation, energy deficits (deficits of high energy phosphates), PH and ventricular interaction, and likewise ischemia [14, 41, 55, 147, 227, 228]. Ischemia caused by coronary ED potentially causes angina symptoms and may affect systolic and diastolic heart function [228, 229].

The development of HFpEF is strongly influenced by aging, a systemic, basically physiological process principally affecting all organs [230, 231]. LV diastolic stiffness rises with increasing age, even when BP and LV-mass are in physiological ranges [232–234]. With aging, diastolic relaxation is blunted attenuating the effect of diastolic suction [235, 236] and subsequently potentially increases LVEDP. NO-dependent vasodilation is compromised [237, 238], and low–grade systemic inflammation with associated impaired NO bioavailability [199] potentially provoking myocardial fibrosis are typical findings. Chronotropic incompetence, limited systolic function, and shortened cardiac output response to exercise [239, 240] further characterize normal aging. Accordingly, aging predisposes for HFpEF, and comorbidities present substantially aggravate the typical “abnormalities” ensuing with increasing age [68]. Aging and hypertension are considered to be the main risk factors for the development of HFpEF [38, 103], as they are a sufficient cause of HFpEF [48, 75]. Moreover, the presence of HTN/HHD was until recently thought to be inevitable for transitioning from asymptomatic DD to HFpEF [139, 199]. Indeed, HFpEF may, in some cases, “simply” reflect predominantly synergistic effects of the risk factors of elderly individuals [48]. As such, if diabetes and HTN coexist, cardiac abnormalities are demonstrated to be more severe and profound than characteristic for and typically seen in each disorder alone [241]. However, obesity, diabetes, HTN, and chronic kidney disease are each associated with unique structural and functional alterations in the heart and vasculature of HFpEF patients [75]. Metabolic disorders like obesity, diabetes, and insulin resistance directly display adverse effects on myocardial structure and function and this independently of confounders like HTN or CAD, referred to as “obesity” [171], “diabetic” [173], and “insulin-resistant” [172] cardiomyopathy. In HFpEF related to diabetes, increased LV diastolic stiffness is reported to be primarily attributed to enhanced cardiomyocyte stiffness and to the hypertrophy of cardiomyocytes [123, 126]. As those diabetic patients did not suffer from HTN, cardiomyocyte hypertrophy was definitely not due to pressure overload, but rather a specific effect of the diabetes [146]. In diabetes and insulin resistance, oxidative stress, generated via several pathways including the accumulation of advanced glycation end products (AGE), is markedly enhanced [146], further coupled with reduced oxidative defence, thus, an inflammatory milieu ensues [242]. Subsequently, NO bioavailability is substantially diminished (AGEs quench NO [243]) and endothelial function will be considerably afflicted and microvascular inflammation of the coronary vessel network and the endocardium occurs [159]. As a consequence of the critically limited NO bioavailability, hypophosphorylation of titin arises as Heerebeek demonstrated, displaying and/or contributing to cardiomyocyte stiffening [123] and cardiomyocyte hypertrophy—the latter typically eccentric [244]. Comparatively, in chronic pressure overload as in HTN and HHD, myocardial abnormalities, typically including concentric hypertrophy [133, 245], and excessive forms of collagen deposition, which will result in a marked increase in myocardial stiffness, are contributing to DD [136]. In obesity, the relative thickness of cardiomyocytes, indicative for concentric hypertrophy, increases [246].

Worsening DD is clearly shown to be independently related to incipient HFpEF development [129, 247]. DD is a prominent manifestation of diabetes [248], and in asymptomatic diabetic patients developing overt HF, worsening diastolic function was definitely related to subsequent incident HF [247]. Moreover, diabetic patients with DD have a significantly higher mortality rate [247]. The Relax-study results further emphasize the adverse role of diabetes in the progression to HFpEF [249].

Hence, HFpEF may be seen as a cardiometabolic disorder [146, 199]. Likewise, chronic pressure load as in HTN/HHD is associated with (1) substantial collagen deposition and changes in collagen composition of the ECM, stiffening the heart muscle [136], and (2) considerable enhanced passive cardiomyocyte tension, both verified in HTN patients who subsequently display DD [78, 140]. Further deteriorating diastolic function (which is usually associated with a (further) rise in LVEDP since abnormal diastolic properties require rising filling pressures to ensure appropriate LV filling [53, 137]) may lead to overt HF symptoms reflecting HFpEF [130, 133, 134]. Thus, various features are involved in the process with a transition from a asymptomatic pre-clinical condition (with likewise enhanced inflammatory markers including IL-6 and TNFα [130, 250, 251]) to overt HFpEF [49, 53, 130, 252].

As such, HTN and consecutively HHD have lost their accentuated role in the group of co-morbidities being necessarily present for the transition from asymptomatic DD to overt HF [75, 199]: Paulus and Tschoepe [199] view HTN as “merely one of many comorbidities fuelling systemic inflammation, oxidative stress, and endothelial dysfunction in this syndrome”, and, “HTN is neither necessary nor sufficient for HFpEF development” as Desai writes [48] interpreting Paulus and Tschoepe.

However, even this example underlines the prominent heterogeneity of aetiologic factors and patho-mechanisms able to contribute to or even to induce HFpEF. The strong association between HFpEF and systemic inflammatory markers is well explained by (a) the inflammation created and induced by the co-morbidities verified and (b) by the hemodynamic-mechanistic features related to increased LVEDP, both causing inflammatory discharge, and as such further substantially supports the diversity of reasons and mechanisms (inflammation may be seen as a vascular response to any threat) found in and characteristic for this type of heart failure [151, 251, 253].

Thus, HFpEF is a very complex disorder with considerable phenotypic heterogeneity, multifactorial pathophysiological pathways, miscellaneous potential etiological factors and multiorgan involvement [13, 117]. Various features are involved in the process of transition from the asymptomatic pre-clinical condition (with likewise enhanced inflammatory markers including IL-6 and TNFα [250, 251]) to overt HFpEF [49, 53, 130, 252]. Accordingly, a “simple” paradigm shift from the traditional mechanistic-hemodynamic (namely afterload excess and vascular failure) approach, which is accompanied by neuroendocrine activation [48, 146], to an inflammatory cardiometabolic disease as suggested by Paulus and Tschoepe [199] will not meet and represent all the facets present, typically assigned to and denoting the syndrome of HFpEF.

Correspondingly, Butler [83] and Tschoepe and vanLinthout [117] point out: HFpEF is a highly complex disorder caused by various etiological features, potentially interacting each other, and as such involves multifactorial patho-physiological pathways. Cardio-metabolic, inflammatory conditions (precipitated by physiological aging possibly amplified by a range of comorbidities commonly accompanying HFpEF) essentially go along with altered mechanical cardio-vascular properties, incited neuroendocrine activity, and altered pulmonary hemodynamics, thereby predispose ensuing overt heart failure.

However, even Butler’s and Tschoepe’s and vanLinthout’s characterisation probably does not describe explicitly enough the wide spectrum of etiological and pathophysiological features verified to potentially contribute to the entity of HFpEF as their annotation does not literally refer to the most essential issue: Analyzing hemodynamic data at rest and when exposing patients with HFpEF to stress, HFpEF is precipitated by a bundle of cardiovascular disorders with heterogeneous underlying pathophysiologies [14, 25, 114]. These include diastolic dysfunction [22, 24, 25, 38] as the central and most common (but not exclusively [68]) pathophysiological hallmark, altered structural and functional systolic myocardial (impaired contractile function, particularly limited contractile reserve) and vascular properties (vascular stiffening and consecutively modified v-a-coupling), blunted (peripheral) vasodilatory response (largely a result of endothelial dysfunction), chronotropic and lusiotropic abnormalities, and the (consecutively) affected pulmonary circulation/RV-PA-unit [14, 21, 25, 56, 58, 62, 78, 114, 117]. Altered LV filling mechanics are the characteristic pathophysiological feature present in all HFpEF patients [254, 255]. They are the result of both “intrinsic structural and molecular alterations” [254], on the one hand attributed to cardio-metabolic, inflammatory aberrations thereby stiffening the left ventricle (heart muscle), and on the other hand assigned to an “increased vascular load imposed by a stiffened arterial vessel system” [254]. A stiffened ventricle and/or an altered vascular load affect ventricular–arterial coupling, and since the pulmonary circulation/RV-PA-unit is generally also afflicted, mainly through the elevated left ventricular filling pressures [254], HFpEF may indeed be considered as a coupling malady [254] (Fig. 5.1).