During the 1970s and 1980s HFpEF was most frequently associated with hypertension, but obesity is currently the most common associated comorbidity.17–19 Obesity is an independent risk factor for HFpEF, more markedly in females than in males, while older women with central obesity are at the highest risk.18,19

Central obesity is defined as increased visceral (intra-abdominal) adipose tissue (VAT).20 Intra-abdominal fat is highly metabolically active, and produces a chronic inflammatory state, as well as hemodynamic imbalance, both of which damage the heart and blood vessels.21

Intra-abdominal adipocytes produce leptin, neprilysin and aldosterone, and promote beta-2 adrenegic receptor and renal sympathetic nerve activation.22 Neprylisin degrades natriuretic peptides, while aldosterone promotes sodium retention, resulting in plasma volume expansion and hence congestion.22 Additionally, neprilysin and aldosterone, as well as adipokines (leptin and others), induce a low-grade chronic inflammatory state, promoting myocardial and vascular fibrosis.23,24 Sodium retention is further amplified by the direct action of beta-2 adrenergic activation, which stimulates mineralocorticoid receptors, leading to cardiac and vascular fibrosis, sodium retention, and consequent plasma volume expansion.22

In sum, adiposity-driven plasma volume expansion is coupled with reduced ventricular and vascular distensibility due to fibrosis, induced by the overstimulation of cardiac and vascular mineralocorticoid receptors. This leads to a marked increase in end-diastolic pressure (hemodynamic congestion), resulting in dyspnea and peripheral edema (clinical congestion).25

The above-mentioned persistent systemic inflammatory state results, in turn, in expansion and inflammation of epicardial adipose tissue. The latter creates a substrate for AF and induces coronary microvascular endothelial inflammation, contributing to myocardial dysfunction and remodeling.12,26,27 Cardiac fibroblasts are activated, leading to interstitial myocardial fibrosis.28

This chain of events leads to myocardial diastolic dysfunction and left ventricular (LV) concentric remodeling, both resulting, again, in a marked increase in end-diastolic pressure.14

As the sequence of events described above involves the left atrium, it can lead to AF and, by involving the left ventricle, promotes diastolic dysfunction, both of which are associated with HFpEF.12

The emergence of AF is very common and important in HFpEF patients, since in a small ventricle with diminished distensibility, the contribution of atrial kick is extremely important to maintain LV filling and cardiac output. Thus, the presence of AF reduces LV filling, leading to diminished output. Additionally, a stiff ventricle is more dependent on diastolic time for adequate filling. Rapid AF dramatically reduces LV diastolic time and thus may lead to flash pulmonary edema.29

Accordingly, the H2FPEF score attributes the greatest weight to the presence of AF among all variables considered when predicting the likelihood that HFpEF is present in an individual patient.30

Furthermore, ventricular fibrosis also occurs in HFpEF and may lead to an increased risk of ventricular arrhythmias and sudden cardiac death.31,32

Between 55% and 90% of HFpEF patients have arterial hypertension, which is the most prevalent comorbidity in HFpEF.33,34

Important mechanisms linking hypertension and HFpEF include LV hypertrophy, myocardial fibrosis, and subsequent diastolic dysfunction.

Myocardial fibrosis and reduced capillary density, in the presence of LV hypertrophy, result in myocardial ischemia, thus worsening LV diastolic function.33–35

Additionally, aortic stiffness consequent to hypertension is responsible for increased arterial wave reflection velocity, inducing late systolic load, thus contributing to LV diastolic dysfunction.

Vascular fibrosis and stiffening also play an important role in HFpEF, since they reduce arterial compliance, increasing LV diastolic dysfunction and producing left and right ventricular-arterial uncoupling.36,37

The association between diabetes and HFpEF may be mediated by hyperinsulinemia, which can suppress autophagy and promote mitochondrial dysfunction. Increased sympathetic activation and cardiac hypertrophy are important consequences of hyperinsulinemia.27 It can also activate Na-H exchangers in renal tubules and cardiomyocytes, and cause expansion of epicardial fat, which is associated with microcirculatory dysfunction and myocardial fibrosis.12

Additionally, the myocardium can become insulin-resistant with obesity, and gender and obesity interact in predicting myocardial glucose uptake and insulin sensitivity.38

Chronic kidney disease is present in about half of patients with HFpEF. Pathophysiological mechanisms implicated in HFpEF are linked to the association of CKD with hypertension and metabolic abnormalities, which activate a systemic inflammatory reaction, endothelial dysfunction, and myocardial fibrosis, including increased central venous and intra-abdominal pressures, renin-angiotensin system activation, oxidative stress, and chronic inflammation.33,34

Considering the overlap between the cardiorenal and cardiometabolic syndromes, it has been suggested that a wider concept of a cardiovascular-kidney-metabolic (CKM) syndrome should be adopted.39 The molecular mechanisms of CKM syndrome include a variety of interconnected factors; however, what links risk factors for CKM to cardiovascular disease and CKD remains unclear.39

Impaired cellular calcium regulation, loss of muscle mass and decline in muscle strength, among many other factors, may play a role in the pathogenesis of HFpEF.40,41

Chronotropic incompetence, defined as a significantly lower peak heart rate during exercise compared to that of normal individuals of the same age group, has also been consistently associated with exercise intolerance in HFpEF, by limiting these patients’ ability to increase cardiac output.42

It is still unclear why HFpEF is more prevalent in women. Women, particularly when older and obese, tend to have greater epicardial and intramyocardial fat volume and higher levels of adipocyte-associated proinflammatory mediators than men.18 Additionally, women with increased visceral adipose tissue (VAT) present 33% greater exercise pulmonary capillary wedge pressure than those without, whereas this was not observed in men.17 Moreover, women tend to show greater arterial stiffness, and a smaller, thicker and stiffer left ventricle than men.43

Both the 2021 ESC and the 2022 American Heart Association/American College of Cardiology/Heart Failure Society of America (AHA/ACC/HFSA) heart failure guidelines recommend the identification and treatment of HFpEF etiology and comorbidities with a class I indication1,33,45,46 and consider them core elements of HFpEF management.1,33,45,46 They also recommend increased physical activity and reduction of body weight in overweight and obese patients.1,45

Congestion is central in HFpEF, and thus loop diuretics are a crucial part of therapy.1,45 Elevated LV filling pressures, as measured by LV end-diastolic pressure (LVEDP), are needed to ensure appropriate diastolic filling, which is essential to maintain cardiac output. However, increased LVEDP, if above a certain level, leads to pulmonary congestion. On the one hand, diuretics are key to reducing congestion and dyspnea in HFpEF by lowering LVEDP. On the other hand, in the context of a stiff ventricle, this may compromise LV filling, impairing cardiac output and causing hypotension.56,57 This is why in HFpEF the window between volume overload and hypovolemia is very narrow; thus, diuretics must be managed with caution. Additionally, considering the preload dependence of HFpEF, congestion correction requires smaller doses of diuretics than in HFrEF.57 Despite the paucity of evidence from randomized controlled trials, diuretics have a class I indication for HFpEF treatment in both the European and American guidelines.1,45

Sodium-glucose co-transporter-2 inhibitors (SGLT2is) were initially shown to reduce HF hospitalizations in patients with type 2 diabetes, and subsequently demonstrated a reduction in HF hospitalizations and cardiovascular and all-cause mortality in HFrEF patients.58,59

In patients with HFpEF, the EMPEROR-Preserved60 and DELIVER61 studies showed that SGLT2is, regardless of the presence of diabetes, reduced the combined risk of cardiovascular death/HF hospitalization (EMPEROR-Preserved) and the risk of cardiovascular death/worsening HF (DELIVER). The combined results of these two studies met the criteria for a class I indication, level of evidence A, from the ESC1 and a II-A recommendation from the AHA/ACC/HFSA guidelines.45 Currently, SGLT2is are the only class of drugs with this level of indication in HFpEF. All patients with HFpEF should be under an SGLT2i (empagliflozin or dapagliflozin), unless not tolerated or contraindicated.

Several hypothetical mechanisms of action for SGLT2is in HFpEF have been advanced to explain their cardioprotective role. These include a modest direct decrease in blood pressure and an increase in diuresis/natriuresis. The latter results in a more selective reduction in extravascular rather than intravascular congestion. Additional possibilities include decreased epicardial fat mass, inflammation prevention, reduction of neurohormonal activation, increase in autophagy and lysosomal degradation, decreased oxidative stress, and the prevention of ischemia/reperfusion injury, among others.62–65 Although the blood pressure-lowering effects of SGLT2is were initially thought to be secondary to diuresis and natriuresis, they have been shown to persist even after a decline in GFR. This suggests that additional mechanisms of action (possibly involving reduced arterial stiffness, changes in sympathetic nervous system activity or improved endothelial function) may play a role as well.63 One of the most recent hypotheses to explain the effects of SGLT2is on cardiac function involves the possible modification of cardiomyocyte metabolic pathways induced by these agents (specifically, the use of ketogenic metabolism rather than fatty acid and glucose metabolism), resulting in improved cardiac efficiency.63,66 It has been proposed that SGLT2 acts as a nutrient surplus sensor, and so its inhibition enhances nutrient deprivation signaling and activates an autophagic flux that reduces cellular stress and promotes cellular survival.67 Another hypothesis involves SGLT2i-induced improvement in myocardial ionic homeostasis.63,66

Renin-angiotensin-aldosterone system inhibitors (RAASis) include angiotensin-converting enzyme inhibitors (ACEis), angiotensin receptor blockers (ARBs), angiotensin receptor/neprilysin inhibitors (ARNIs) and mineralocorticoid receptor antagonists (MRAs). None of the large trials involving RAASis in HFpEF met the composite endpoint of reducing cardiovascular mortality/HF hospitalizations. These include PEP-CHF (with perindopril), CHARM-Preserved68 (with candesartan), PARAGON-HF69 (with sacubitril/valsartan) and TOPCAT70 (with spironolactone). Nevertheless, although cardiovascular mortality was not reduced in the active arm in any of the above studies, candesartan and spironolactone reduced HF hospitalizations and sacubitril/valsartan showed a trend with no statistical significance.

The TOPCAT study deserves some additional comments. A pre-specified subgroup analysis suggested that the primary outcome was significantly reduced by spironolactone in patients with elevated natriuretic peptide levels at baseline. A subsequent subanalysis restricted to patients enrolled in North and South America showed a significant reduction in the primary composite endpoint (time to cardiovascular death, aborted cardiac arrest, or hospitalization for management of heart failure) in the intervention arm compared to placebo, whereas this was not observed in the European cohort.71,72 It is noteworthy that among the Russian cohort, a higher percentage of participants showed undetectable levels of canrenone (an active metabolite of spironolactone) compared to American and Canadian subjects.73

Considering the importance of vascular and cardiac fibrosis in HFpEF, it is plausible that antagonizing the mineralocorticoid receptor may have a positive impact on prognosis.74

ARNIs, ARBs and MRAs received a class II-B indication in the AHA/ACC/HFSA guidelines for HFpEF patients, with the disclaimer of higher benefit demonstrated in patients with LVEF closer to 50%.1,45

Additionally, following the FIDELIO-DKD and FIGARO-DKD trials, finerenone, a non-steroidal MRA with high binding affinity, has a class I level A recommendation for the prevention of HF hospitalizations in patients with albuminuric CKD and type 2 diabetes.75 Pre-specified subgroup analyses of these two studies suggested a potential beneficial effect of finerenone in HF patients without symptomatic HFrEF.76,77 This effect was recently confirmed in the FINEARTS-HF study, which included 6016 NYHA class II–IV HFmrEF/HFpEF (LVEF ≥40%; LVEF ≥60% capped at 20%) patients with structural cardiac abnormalities, NT-proBNP ≥300 pg/ml (≥900 pg/ml if AF), K+ ≤5.0 mmol/l, estimated GFR ≥25 ml/min/1.73 m2, and on diuretic treatment for ≥30 days. This study met the primary endpoint with a statistically significant and clinically meaningful reduction in the composite of cardiovascular death and total HF events vs. placebo.78,79

There is no prospective evidence of a positive impact of beta-blockers on HFpEF. Some studies suggest a deleterious effect, with a higher risk of hospitalizations.80

Iron supplementation with IV administration of ferric carboxymaltose (FCM) has a clear beneficial effect on HFrEF patients.81,82 However, there are scarce data on the potential of iron supplementation in patients with HFpEF. Small-sample randomized clinical trials and retrospective studies have shown that supplementation with FCM significantly improves cardiac performance of HFpEF patients.83,84 Results from the ongoing PREFER-HF and FAIR-HFpEF trials may add evidence on this subject.85,86

Although they reduce epicardial fat and inflammation, GLP-1 receptor agonists pose challenges with sodium retention in HFpEF.87 As stated above, in the STEP-HFpEF trial, the GLP-1 agonist semaglutide demonstrated significant benefits in non-diabetic obese HFpEF patients, including reduced symptoms, enhanced exercise capacity, and greater weight loss compared to placebo across obesity categories.50,51 Notably, the degree of benefit correlated directly with weight reduction, emphasizing its pivotal role as a treatment strategy in the obesity phenotype of HFpEF.

Metformin reduces epicardial tissue expansion and inflammation, as well as myocardial fibrosis.87 In observational studies, it appeared to reduce the incidence of AF and HFpEF.87

Thiazolidinediones and dipeptidyl peptidase-4 inhibitors reduce epicardial adipose tissue dysfunction and could potentially reduce the risk of AF and HFpEF.87 However, although thiazolidinediones have favorable effects on experimental HFpEF and AF, their antinatriuretic properties make them inappropriate for use in HFpEF patients.87

VAT is a highly active endocrine organ that secretes cytokines and other mediators. As pointed out above, increased VAT seen in central obesity leads to epicardial fat expansion and inflammation, a major step in the genesis of HFpEF. Thus, targeting VAT may be a future approach to HFpEF through dietary, drug and/or surgical weight loss strategies.88

Epicardial adipose tissue expansion and associated inflammatory activity may be reduced by statins and anticytokine agents.87 These agents could, in theory, thereby reduce the incidence of AF and HFpEF.87

The ongoing EMPA-PRED and ENRICH-PEF clinical trials are expected to provide further evidence on HFpEF management. The EMPA-PRED study aims to assess the safety and efficacy of empagliflozin in patients with end-stage renal disease and HFpEF.89 The effects of enavogliflozin on exercise performance, diastolic dysfunction, and quality of life in HFpEF patients will be investigated in the ENRICH-PEF trial.90 Additionally, trials with aldosterone synthase inhibitors, such as baxdrostat, for treatment-resistant hypertension are ongoing.91

Atrial pacing, atrial shunting and AF ablation are three non-pharmacological alternatives that are being studied in HFpEF, although the evidence is not consistent. In the RAPID-HF trial, atrial pacing increased early and peak exercise heart rate, but no improvement in exercise performance or quality of life was observed.92 The CABA-HFPEF trial will assess whether catheter-based AF ablation can improve cardiovascular outcomes compared to conventional treatment in these patients.93 Lastly, the first large trial of an atrial shunt device, designed to lower left atrial pressure, presented neutral overall results. Ongoing trials are expected to add evidence in this field.94