Heart failure with preserved ejection fraction

Abstract

As part of this series devoted to heart failure (HF), we review the epidemiology, diagnosis, pathophysiology, and treatment of HF with preserved ejection fraction (HFpEF). Gaps in knowledge and needed future research are discussed.

Fig. 1

Doppler criteria for classification of diastolic function: reproduced with permission from [105]. E, peak early filling velocity; A, velocity at atrial contraction; DT, deceleration time; Adur, A duration; ARdur, AR duration; S, systolic forward flow; D, diastolic forward flow; AR, pulmonary venous atrial reversal flow; e′, velocity of mitral annulus early diastolic motion; a′, velocity of mitral annulus motion with atrial systole; DT, mitral E velocity deceleration time

Fig. 2

Impaired LV relaxation increases diastolic filling pressure in the presence of tachycardia: reproduced with permission from [50]. Modeling experiments defining the impact of variation in the time constant of isovolumic relaxation on pressure–volume loops at heart rate of 60 bpm (a and c). Increases in τ values from 20 to 80 ms cause little change in the mean and end-diastolic LV pressure. At a heart rate of 120 bpm (b and d), increases in τ values cause increases in the mean and end-diastolic LV pressure

Fig. 3

Increased left ventricular diastolic stiffness in HFpEF demonstrated by invasive measurement of the diastolic pressure–volume relationships. Panel (a) from Zile et al. (reproduced with permission from [135]) shows a significant increase in the passive stiffness of the left ventricle in patients with diastolic heart failure (HFpEF) as compared to controls. Panel (b) from Borlaug et al. (reproduced with permission from [14]) shows summary data for diastolic pressure–volume relationships of HFpEF patients at rest (black) and with exercise (red) plotting raw (solid) and relaxation-corrected (dashed) data. With exercise, the position of the diastolic pressure–volume relationship shifts upward with increasing diastolic pressure at similar volumes. The diastolic stiffness constants calculated from these relationships increased significantly with exercise

Fig. 4

Differences in LV mechanics drive differential response to therapy in HFpEF and HFrEF: reproduced with permission from [112]. Observed and predicted end-systolic pressure volume changes with acute vasodilation in HFpEF and HFrEF. Patients were administered similar doses of nitroprusside with invasive hemodynamic monitoring. Observed changes in the end-systolic pressure–volume coordinates (diamonds) in HFpEF (black) compared to HFrEF (red) with reduction in arterial elastance (E_a, dashed lines) were similar to those predicted based on resting left ventricular end-systolic elastance (E_es, solid lines) and the observed changes in E_a with nitroprusside. Reduction in afterload produces greater blood pressure drop and less increment in stroke volume in HFpEF than HFrEF due to the difference in E_es

Fig. 5

Reduced left atrial compliance resulting in giant atrial “v” waves in HFpEF: reproduced with permission from [8]. Hemodynamic tracings from a HFpEF patient are shown. a Severe pulmonary arterial hypertension is noted at baseline. There is also elevation in the pulmonary capillary wedge pressure with a prominent V wave (36 mmHg). Right atrial pressure is elevated. b Sodium nitroprusside is administered, reducing pulmonary artery and pulmonary capillary wedge pressure and the V wave. c After re-equilibration, inhaled nitric oxide (iNO) is administered to increase pulmonary venous flow. Despite pulmonary arterial vasodilatation, pulmonary arterial pressures do not decline due to increased flow, but a marked rise in mean pulmonary capillary wedge pressure and in the V-wave amplitude is seen. Simultaneous echocardiography confirmed the absence of mitral regurgitation. Similar increases in V-wave amplitude are seen with exercise in HFpEF and are a hemodynamic hallmark of HFpEF

Fig. 6

Integrative HFpEF pathophysiology: patients with HFpEF have diastolic dysfunction. While arterial and LV systolic elastance (stiffness) are increased in HFpEF, resting contractile function is subtly impaired, as is the ability to appropriately dilate the peripheral vasculature and enhance chronotropic and LV systolic and diastolic performance with exercise (impaired reserve function). Chronic elevation of LV filling pressures leads to left atrial remodeling and dysfunction, mixed pulmonary hypertension, and, ultimately, right ventricular (RV) remodeling and dysfunction. Atrial remodeling leads to atrial fibrillation (AF), which contributes prominently to HFpEF pathophysiology. The HF state leads to neurohumoral (NH) activation and aggravates underlying age- and comorbidity-related renal dysfunction. These perturbations express first as effort intolerance and progress to lead to advanced HF

Fig. 7

Cardiomyocyte signaling pathways involved in regulating cardiac titin stiffness. Cardiomyocyte stiffness can be modulated by reversible phosphorylation (P) either within the titin N2-B element by factors activating cAMP/PKA, cGMP/PKG, MEK1/ERK2, and CaMKIIδ or within the titin PEVK element by factors activating CaMKII and PKCα. Titin stiffness can also be modulated by altering the N2BA/N2B titin isoform expression ratio. AC adenylyl cyclase, ATP adenosine triphosphate, βAR β-adrenergic receptor, CaMKII Ca²⁺/calmodulin-dependent protein kinase-IIδ, cAMP cyclic adenosine monophosphate, cGMP cyclic guanosinemonophosphate, ERK2 extracellular signal-regulated kinase-2, G small G-protein, GPCR G-protein-coupled receptor, GTP guanosine triphosphate, MEK1/2 MAPK/ERK kinase-1 and kinase-2, mTOR mammalian target of rapamycin, pGC particulate guanylyl cyclase, PI3K phosphatidylinositol-3-kinase, PKA protein kinase-A, PKCα protein kinase-Cα, PKG cGMP-dependent protein kinase-G, PLC phospholipase-C, Raf rat fibrosarcoma protein, Ras rat sarcoma protein, RBM20 RNA-binding motif protein-20, sGC soluble guanylyl cyclase, T3 triiodo-l-thyronine, TR thyroid hormone receptor. Titin segments: Ig, immunoglobulin-like domain; N2-B, cardiac-specific I-band element; PEVK, disordered sequence (>70 % P, E, V, and K residues); FNIII, fibronectin type III-like domain

Fig. 8

Titin isoform expression, titin isoform phosphorylation, and cardiomyocyte F_passive in aortic stenosis (AS), HFpEF (HFNEF), and HFrEF patients: reproduced with permission from [10]. a Representative sequential gel electrophoresis with SYPRO Ruby and ProQ Diamond phosphoprotein stain in left ventricular samples from patients with AS, HFrEF, and HFpEF showing phosphorylation of titin isoforms. b Higher N2BA/N2B ratio in AS, HFpEF, and HFrEF patients than in the control (CON) group (†P<0.05 vs. CON; ‡P<0.001 vs. CON). c Higher left ventricular end-diastolic wall stress (LVEDWS) in HFrEF patients than in CON, AS, and HFpEF groups (‡P<0.001 vs. CON, AS, or HFpEF). d Higher cardiomyocyte stiffness (F_passive) in HFpEF patients than in CON, AS, and HFrEF patients (‡P<0.001 vs. CON or AS; †P<0.01 vs. HFREF). After in vitro administration of PKA, F_passive fell in HFpEF (#P=0.0001 vs. HFNEF) and HFrEF cardiomyocytes (##P=0.0001 vs. HFREF). e Parallel trends in AS, HFREF, and HFNEF patients of the N2B phosphorylation deficit (P-N2BA/P-N2B, open squares) and the PKA-induced fall in F_passive (filled circles)

Fig. 9

Novel unifying paradigm for HFpEF pathophysiology: reproduced with permission from [96]