Heart failure with preserved ejection fraction (HFpEF)

Overview and definition

Picture a 77-year-old woman with hypertension, AF and obesity who gets breathless climbing stairs. Her echo shows a normal ejection fraction. Her profile fits HFpEF epidemiology: in population samples nearly half of all heart failure is HFpEF, and patients tend to be older and more often female.^[1]

The 2026 ESC guideline defines heart failure as a clinical syndrome of signs and/or symptoms caused by structural and/or functional cardiac abnormalities. Those abnormalities result in elevated intracardiac pressures and/or inadequate cardiac output, at rest and/or during exercise.^[1]

This guideline updates and replaces the 2021 ESC version, so the 2021 three-band scheme is now history.^[1]

Classification

Two phenotypes in 2026

The 2021 ESC guideline used three phenotypes: HFrEF (LVEF 40% or below), HFmrEF (LVEF 41–49%) and HFpEF (LVEF 50% or above).^[1] The 2026 Task Force eliminated HFmrEF and reclassified heart failure on a pathophysiological basis into two distinct phenotypes.^[1]

Why? The ESC reasons that HFmrEF is similar to HFrEF in clinical characteristics, outcomes and response to treatments. Its pathophysiology is closer to HFrEF (reduced contractility and systolic dysfunction) than to HFpEF (increased stiffness and diastolic dysfunction).^[1] Many patients with LVEF 41–49% may experience a deterioration in LVEF.^[1]

^[1]

LVEF is a continuous measure, so the ESC gives the 50% value only as guidance. A single cut-point is "somewhat arbitrary", LVEF measurement varies with modality and interpreter, and LVEF can read falsely high, for example in mitral regurgitation.^[1]

Improved ejection fraction

Improved LVEF needs a rise of at least 10% (absolute) to a value above 40%, often into the normal range.^[1] These patients stay on FMT at the highest tolerated doses, including those who become asymptomatic or improve their LVEF, to reduce the risk of HF hospitalisation and death (Class I, level C).^[1] Gradual discontinuation under frequent clinical, laboratory and imaging surveillance may be considered only in highly selected asymptomatic patients with complete normalisation of LV function, LV volume and natriuretic peptides after specific treatment of a reversible cause, to accommodate patient preference (Class IIb, level C).^[1]

Stages A to D

The 2026 guideline adopts heart failure stages. Stage A is at risk without signs, symptoms or cardiac abnormalities. Stage B (pre-HF) has cardiac abnormalities without signs or symptoms. Stage C is symptomatic HF: cardiac abnormalities with current or prior signs and/or symptoms of HF.^[1] Stage D is advanced HF, with persistent severe symptoms and objective cardiogenic impairment despite optimal therapy.^[1]

Therapy labels

^[1]

Reading the class and level

The 2026 ESC Guidelines, including this one, are the first to use the revised level-of-evidence system.^[1] For therapy, level A is conclusive evidence, usually from 2 or more adequately powered RCTs free from major bias. Level B is now split into B1 and B2.^[24]

^[24]

Diagnostic tests have a separate scale. Level A needs at least two high-quality studies, level B one high-quality or at least two moderate-quality studies, and level C the rest, including expert consensus.^[25]

Epidemiology and risk factors

^[1]

The HFpEF share may be rising because coronary disease is treated better while hypertension, obesity and other HFpEF risk factors become more common.^[1] Diabetes, hypertension, obesity and smoking account for about 50% of all HF events at population level, and the share is similar for HFpEF and HFrEF.^[1] Sedentary behaviour and physical inactivity are strong risk factors, especially for HFpEF.^[1]

The comorbidity load is heavy. HFpEF patients tend to be older and more often female, with a high burden of comorbidities including AF, CKD, obesity, diabetes and non-cardiovascular comorbidities.^[1] Hypertension is common in HF, especially in HFpEF, and AF is slightly more prevalent in HFpEF than in HFrEF.^[1] Iron deficiency is also slightly more prevalent in HFpEF.^[1]

Several subtypes of HFpEF have been suggested, and their shares vary by region. Up to 50% have a metabolic/obesity subtype (younger, higher BMI) in some regions; up to 50% have an ageing, vascular subtype (older, stiff arteries, frequent CKD) in others.^[1]

Cohort data show the shift over time. Among Mayo Clinic Olmsted County HF discharges from 1987 to 2001 with ejection fraction data (4596 of 6076), 47% had preserved ejection fraction, the proportion rose over time, and it was higher in community than referral patients (55% vs 45%).^[19] In the Framingham and Cardiovascular Health Study cohorts aged 60 or over, HFrEF incidence fell between the 1990s and 2000s while HFpEF incidence rose.^[20]

Pathophysiology

Start with the ventricle. The ESC contrasts the two phenotypes directly: HFrEF is reduced contractility and systolic dysfunction, HFpEF is increased stiffness and diastolic dysfunction.^[1]

The Paulus–Tschöpe paradigm proposes how comorbidities lead to that stiffness. It proposes a systemic proinflammatory state induced by comorbidities as the cause of the myocardial changes.^[17]

1. A high prevalence of comorbidities such as overweight/obesity, diabetes, COPD and salt-sensitive hypertension induces a systemic proinflammatory state.
2. That state causes coronary microvascular endothelial inflammation.
3. Endothelial inflammation reduces nitric oxide bioavailability, cGMP content and protein kinase G (PKG) activity in adjacent cardiomyocytes.
4. Low PKG activity favours hypertrophy and raises resting tension through titin hypophosphorylation.
5. Stiff cardiomyocytes and interstitial fibrosis contribute to high diastolic LV stiffness and heart failure development.

^[17]

The paradigm shifts emphasis from LV afterload excess to coronary microvascular inflammation. Remodelling also differs from HFrEF, where loss of cardiomyocytes drives it.^[17]

Obesity adds its own mechanics. Adipocyte mass drives systemic inflammation and a proinflammatory epicardial fat that may cause fibrosis. The ventricle's capacity to dilate in response to extra blood volume is impaired, so even modest volume overload may lead to overfilling and a disproportionate rise in filling pressures.^[1] Obese HFpEF patients are younger on average, with worse signs and symptoms, quality of life and functional capacity.^[1]

Clinical presentation

Typical symptoms (ESC Table 7) are dyspnoea, orthopnoea, paroxysmal nocturnal dyspnoea, reduced exercise tolerance, fatigue with longer recovery after exercise, and ankle oedema. Bendopnoea, a less typical symptom, is breathlessness on bending forward.^[1] Elderly patients may present atypically, and in the very elderly suspicion should be raised even with an atypical presentation.^[1]

Diagnosis starts with pre-test probability. Several clinical factors may contribute to or be associated with HFpEF; the more factors present, the higher the likelihood.^[1] Clinical suspicion is followed by natriuretic peptide testing and echocardiography.^[1]

^[1]

Differential diagnosis

The ESC asks you to exclude conditions that might mimic HFpEF, such as HCM and amyloidosis. It also flags conditions with similar symptoms that may coexist with it: pulmonary disease, anaemia, obesity and deconditioning.^[1]

^[1]

ATTR cardiomyopathy is prevalent in HFpEF, severe aortic stenosis, carpal tunnel syndrome, lumbar spinal stenosis and autonomic or sensory polyneuropathy (AHA/ACC/HFSA 2022).^[2]

Investigations

Natriuretic peptides

Natriuretic peptide measurement is recommended in suspected HF, interpreted in relation to age, obesity and other factors that affect the level (Class I, level C).^[1] The 2021 ESC single rule-out threshold (NT-proBNP below 125 pg/mL) lacks specificity in older people: more than 75% of a healthy population over 80 years exceeded it.^[1]

^[1]

Peptides also fall with ARNI/ACE-I/ARB, beta-blockers, MRAs, SGLT2 inhibitors, constrictive pericarditis and African ancestry, although ARNI may raise BNP levels.^[1] They rise with AF, kidney dysfunction, advanced age, pulmonary hypertension and many other conditions. The higher the level, the more likely heart failure is the diagnosis.^[1]

Echocardiography: ESC Table 10

Transthoracic echocardiography is recommended to confirm the diagnosis, distinguish phenotypes and seek the aetiology (Class I, level C).^[1] ESC Table 10 gives simplified echocardiographic criteria for supporting objective evidence of HFpEF:

^[1]

The probability of HFpEF rises with each parameter in the pathological range. E/e′ of 15 or more is more specific but less sensitive.^[1] AHA/ACC/HFSA 2022 likewise names an increased left atrial volume index and/or LV mass index as the key structural alterations.^[2]

Scores

The ESC wants a pragmatic diagnosis. Complex scores have given conflicting results in the community, though meta-analysis suggests acceptable accuracy. If used, they should be interpreted with caution and used primarily to raise clinical suspicion of HFpEF rather than to establish a definitive diagnosis.^[1]

^[2]

In a retrospective case-control study of outpatients with unexplained dyspnoea, HFpEF was confirmed by elevated exercise PCWP and controls had normal rest and exercise haemodynamics. There, H2FPEF had a higher AUC than HFA-PEFF (0.845 vs 0.710). Low-probability HFA-PEFF scores had a 55% false-negative rate, against 25% for H2FPEF.^[16]

When the diagnosis is uncertain

CPET, exercise stress echocardiography or invasive haemodynamic assessment may be considered.^[1] Diastolic stress testing is emerging as a tool to unmask diastolic dysfunction that may not be apparent at rest.^[1] In obese patients, read CPET with care: peak oxygen consumption adjusted for body weight underestimates exercise capacity, and adjustment for lean body mass should be used for risk stratification.^[1]

Invasive testing is indicated in only a minority of patients: those whose diagnosis is uncertain and in whom establishing the diagnosis or an alternative cause is likely to affect management.^[1] Right heart catheterisation (potentially including exercise right heart catheterisation) may be considered in selected patients with suspected HFpEF or structural heart disease to confirm the diagnosis, or when clinical examination and non-invasive testing are insufficient to understand the haemodynamic state (Class IIb, level C).^[1]

In suspected HF, laboratory screening for comorbidities is recommended: full blood count, eGFR and UACR, electrolytes, liver and thyroid function, HbA1c, lipids, and iron status (TSAT and ferritin) (Class I, level C).^[1]

Management: chronic HFpEF

Goals

The three goals are lower mortality, fewer recurrent HF hospitalisations, and better clinical status, functional capacity and quality of life.^[1] The ESC reports no sex differences in HF drug recommendations.^[1]

Foundational medical therapy

The ESC recommends MRAs and SGLT2 inhibitors for all symptomatic HF regardless of LVEF, because they reduce mortality and HF hospitalisation.^[1]

^[1]

Doses (ESC Table 11)

^[1]

Sacubitril–valsartan has an optional lower starting dose of 24/26 mg twice daily for patients with a history of symptomatic hypotension, ACE-I-naive patients, or eGFR 30–60 mL/min/1.73 m².^[1]

Spironolactone has an optional 12.5 mg start when kidney function or hyperkalaemia warrants caution. Finerenone's starting and target doses differ by eGFR: 10–20 mg if eGFR is 60 mL/min/1.73 m² or below, and 20–40 mg if eGFR is above 60.^[1]

SGLT2 inhibitors: the evidence

^{[3] [4] [5]}

• DELIVER: 6263 patients with HF and LVEF above 40% were randomised to dapagliflozin 10 mg once daily or matching placebo, in addition to usual therapy. Over a median of 2.3 years, worsening HF or CV death occurred in 16.4% versus 19.5% (HR 0.82; 95% CI 0.73–0.92).^[3]
• DELIVER components: worsening HF HR 0.79 (95% CI 0.69–0.91); CV death HR 0.88 (95% CI 0.74–1.05). Results were similar above and below LVEF 60%, and with or without diabetes.^[3]
• EMPEROR-Preserved: in a double-blind trial, 5988 patients with class II–IV HF and LVEF above 40% were randomised to empagliflozin 10 mg once daily or placebo, in addition to usual therapy. Over a median of 26.2 months, the primary outcome occurred in 13.8% versus 17.1% on placebo (HR 0.79; 95% CI 0.69–0.90), mainly through fewer HF hospitalisations; effects appeared consistent with or without diabetes.^[4]
• Pooled meta-analysis: across 12 251 participants, CV death or first HF hospitalisation HR 0.80 (0.73–0.87); CV death 0.88 (0.77–1.00); first HF hospitalisation 0.74 (0.67–0.83).^[5]

Common side effects include genital and urinary infections. A mild early eGFR fall is expected, and long-term kidney function improves.^[1]

MRAs: steroidal and non-steroidal

The MRA recommendation for HFpEF rests on a meta-analysis of spironolactone and finerenone trials.^[1] Spironolactone was studied across LVEF ranges, eplerenone only in HFrEF (LVEF up to 35%), and finerenone in patients with LVEF >40% per the ESC (FINEARTS-HF enrolled LVEF of 40% or more).^[1]

• TOPCAT: in a randomised, double-blind trial, 3445 patients with symptomatic HF and LVEF of 45% or more received spironolactone 15–45 mg daily or placebo. Over a mean follow-up of 3.3 years, the primary outcome was 18.6% versus 20.4% (HR 0.89; 95% CI 0.77–1.04; P=0.14).^[7]
• TOPCAT signal and cost: of the primary-outcome components, only HF hospitalisation was significantly lower (12.0% vs 14.2%; HR 0.83; P=0.04). Hyperkalaemia doubled (18.7% vs 9.1%) and creatinine rose.^[7]
• FINEARTS-HF: in a double-blind trial, patients with HF and LVEF of 40% or more were randomised 1:1 to finerenone (maximum 20 mg or 40 mg once daily) or matching placebo, in addition to usual therapy. Total worsening HF events plus CV death: rate ratio 0.84 (95% CI 0.74–0.95; P=0.007).^[6]
• FINEARTS-HF detail: over a median follow-up of 32 months, CV death was 8.1% versus 8.7% (HR 0.93; 95% CI 0.78–1.11). Hyperkalaemia rose and hypokalaemia fell.^[6]
• Patient-level MRA meta-analysis (13 846 patients): CV death or HF hospitalisation HR 0.87 (0.79–0.95) in HFmrEF/HFpEF trials versus 0.66 in HFrEF. CV death was not reduced in HFmrEF/HFpEF (0.92; 0.80–1.05).^[8]

Finerenone and eplerenone are more mineralocorticoid-receptor specific and cause less gynaecomastia.^[1] The ESC makes no recommendation on potassium binders, because evidence of clinical benefit is insufficient.^[1]

Loop diuretics

^[1]

The ESC files loop diuretics under additional medical therapy for all LVEF ranges, so they are not part of FMT. They treat congestion with dynamic, individualised dosing (Class I, level A; the Task Force assigned level A for clear clinical effectiveness; dynamic dosing is so widely accepted as effective that adequately powered RCTs have not been, and for ethical reasons are unlikely to be, conducted). The ESC generally advises the lowest possible dose that maintains euvolaemia.^[1] Avoid chronic thiazide use in stable outpatients: it often induces severe electrolyte disturbances that can go undetected in the ambulatory setting.^[1]

ACE-I, ARB and ARNI

^[1]

• PARAGON-HF: 4822 patients with NYHA class II–IV HF, LVEF of 45% or more, elevated natriuretic peptides and structural heart disease were randomised to sacubitril–valsartan (target 97/103 mg twice daily) or valsartan (target 160 mg twice daily). Primary rate ratio 0.87 (95% CI 0.75–1.01; P=0.06).^[9]
• PARAGON-HF components: CV death 8.5% versus 8.9% (HR 0.95); total HF hospitalisation rate ratio 0.85 (0.72–1.00).^[9]
• Who might benefit: PARAGON-HF hinted at benefit with lower ejection fraction and in women.^[9] A PARADIGM-HF plus PARAGON-HF meta-analysis showed benefit below an LVEF of 60–65%, so the ESC says ARNIs may be considered, especially at lower LVEF.^[1]
• ARNI safety: more hypotension and angioedema, less hyperkalaemia.^[9]
• ARBs: a four-trial meta-analysis found no signal for benefit on CV death (HR 1.02), all-cause death (HR 1.01) or HF hospitalisation (HR 0.92).^[1]
• CHARM-Preserved: 3023 patients with NYHA class II–IV HF and LVEF above 40% were randomised to candesartan (target 32 mg once daily) or matching placebo. Over a median follow-up of 36.6 months, the primary outcome was 22% versus 24% (unadjusted HR 0.89; 95% CI 0.77–1.03). Fewer patients were admitted for heart failure (230 vs 279).^[23]

ACE-Is, ARBs and beta-blockers may be used particularly for comorbidities, because their effect on morbidity and mortality in HFpEF is uncertain. Many patients have coexisting indications for ACE-I/ARB therapy, such as hypertension, diabetes or CKD.^[1] Vericiguat is an HFrEF drug: it may be considered in symptomatic HFrEF with LVEF under 45% despite optimal FMT, to reduce the risk of HF hospitalisation and CV death (Class IIb, level B1).^[1]

Beta-blockers

Beta-blockers are FMT for HFrEF only. In HFpEF the ESC says they may be used particularly for comorbidities.^[1]

• Individual-patient meta-analysis of 11 RCTs: in sinus rhythm, beta-blockers reduced all-cause and CV mortality consistently across LVEF strata, except in the small subgroup with LVEF of 50% or more (244 patients). LVEF rose with beta-blockers in every sinus-rhythm group except LVEF of 50% or more.^[21]
• Same analysis, AF: in atrial fibrillation, beta-blockers did not improve prognosis.^[21]
• PRESERVE-HR: a randomised, investigator-blinded crossover trial in HFpEF with chronotropic incompetence. Peak VO₂ rose after beta-blocker withdrawal (14.3 vs 12.2 mL/kg/min).^[22]
• AHA/ACC/HFSA 2022: beta-blockers may treat hypertension after MI, with symptomatic CAD, or in AF with rapid ventricular response, balanced against the potential contribution of chronotropic incompetence to exercise intolerance in some patients.^[2]

Obesity: semaglutide and tirzepatide

^[1]

Lifestyle interventions (such as caloric restriction, exercise and patient education) have a significant impact on body weight and are recommended for all patients with HF.^[1]

• STEP-HFpEF: 529 patients with HFpEF and BMI of 30 or more were randomised to semaglutide 2.4 mg once weekly or placebo for 52 weeks. Placebo-adjusted KCCQ-CSS difference 7.8 points (mean change 16.6 vs 8.7 at 52 weeks); placebo-adjusted body-weight difference −10.7 percentage points (mean change −13.3% vs −2.6%).^[10]
• STEP-HFpEF function: placebo-adjusted 6-minute walk difference 20.3 m (mean change 21.5 vs 1.2 m), and the hierarchical win ratio was 1.72.^[10]
• STEP-HFpEF DM: 616 patients with HFpEF, BMI of 30 or more and type 2 diabetes were randomised to semaglutide 2.4 mg once weekly or placebo for 52 weeks. Placebo-adjusted KCCQ-CSS difference 7.3 points; placebo-adjusted body-weight difference −6.4 percentage points.^[11]
• SUMMIT: 731 patients with HF, LVEF of at least 50% and BMI of at least 30 were randomised 1:1 to tirzepatide (up to 15 mg subcutaneously weekly) or placebo. Over a median follow-up of 104 weeks, CV death or worsening HF occurred in 9.9% versus 15.3% on placebo (HR 0.62; 95% CI 0.41–0.95).^[12]
• SUMMIT components: worsening HF events HR 0.54; CV death 8 versus 5 patients (HR 1.58; 95% CI 0.52–4.83). At 52 weeks the mean KCCQ-CSS change was 19.5 versus 12.7 points (placebo-adjusted difference 6.9 points).^[12]
• Post-hoc pooled analysis of SELECT, FLOW, STEP-HFpEF and STEP-HFpEF DM (3743 of 22 282 participants with a history of HF with mildly reduced or preserved EF, which the paper calls HFpEF): CV death or HF events HR 0.69 (0.53–0.89); worsening HF events HR 0.59; CV death alone HR 0.82 (not significant).^[13]

The ESC notes that RCTs in HFpEF with obesity demonstrated favourable effects on body weight, QoL, symptoms and functional capacity. Pooled analyses and SUMMIT "suggested a potential benefit" on HF events, and morbidity and mortality effects remain an evidence gap.^[1]

In SUMMIT, adverse events, mainly gastrointestinal, led to stopping tirzepatide in 6.3% versus 1.4% on placebo.^[12]

Exercise

Personalised exercise training within multidisciplinary exercise-based cardiac rehabilitation is recommended for all stable patients, unless there are specific contraindications, to improve exercise capacity and QoL and reduce all-cause hospitalisation (Class I, level B1). Personalised exercise training outside rehabilitation is recommended for all stable patients on a long-term basis, unless there are specific contraindications, to improve exercise capacity and QoL and reduce all-cause hospitalisation (Class I, level B2).^[1] In HFpEF it is safe and improves exercise capacity and QoL. Data on death and admissions are insufficient.^[1] A meta-analysis of 7 RCTs (346 participants) found that peak VO₂ improved (weighted mean difference 2.57), with no significant change in diastolic function.^[26]

Devices

^{[1] [14] [15]}

In RELIEVE-HF, shunting in the preserved-LVEF stratum (LVEF >40%) raised the annualised CV event rate to 60.2% versus 35.9% (RR 1.68).^[14]

Decompensated heart failure

The 2026 guideline replaces "acute HF" with decompensated HF (DHF). DHF is acute or gradual onset of HF symptoms and/or signs severe enough to deserve urgent medical attention, leading to an unplanned hospital admission, emergency department visit or ambulatory visit, and requiring initiation or intensification of treatment. It may be worsening of pre-existing HF or de novo HF.^[1] Treatment addresses any associated triggers urgently, with diuretics for congestion plus vasodilators for high blood pressure, and inotropes and/or temporary MCS for hypoperfusion.^[1]

• Adding short-term IV acetazolamide or oral hydrochlorothiazide should be considered in fluid overload previously treated with loop diuretics, to reduce congestion (Class IIa, level B1).^[1]
• In-hospital initiation of an SGLT2 inhibitor is recommended in DHF after initial stabilisation, to improve QoL and congestion symptoms and reduce HF hospitalisation (Class I, level B1).^[1]
• Exclude persistent congestion before discharge, start or optimise FMT, and book early post-discharge visits.^[1]

Specific scenarios

Atrial fibrillation. AF occurs in more than half of HF patients.^[1] Oral anticoagulation is recommended in clinical AF at elevated thromboembolic risk by CHA₂DS₂-VA, to prevent ischaemic stroke and thromboembolism (Class I, level A); DOACs are recommended in preference to VKAs in HF to prevent stroke and thromboembolism (Class I, level B1). The exception is moderate or severe mitral stenosis or a mechanical prosthetic valve, where VKAs are recommended.^[1] There is no evidence on drug strategies for rate control in AF with HFpEF. Lenient rate control, with a resting heart rate under 110 b.p.m., should be considered as the initial target, with re-evaluation based on symptoms.^[1]

The catheter ablation recommendation (Class IIa, level C) covers selected patients with symptomatic AF and HFrEF, to improve QoL and reduce the risk of HF hospitalisation or death. All of these are required: high-burden AF, continuous persistent AF of less than 1 year, and a clear cause–effect relationship between AF and HF. The ESC says ablation should be considered only in carefully selected patients with HFrEF and symptomatic AF. Data in HFpEF are limited.^[1] AV node ablation with CRT beat drug therapy in severely symptomatic permanent AF regardless of LVEF.^[1]

Hypertension. The best antihypertensive strategy in HFpEF is uncertain. Start or uptitrate an MRA; ACE-Is, ARBs and calcium channel blockers regress LV mass. Keep the patient normotensive and avoid hypotension, because preload reserve is limited.^[1]

Diabetes. SGLT2 inhibitors are recommended in all HF with type 2 diabetes, whatever the HbA1c.^[1] Semaglutide or tirzepatide should be added if not contraindicated in HFpEF with type 2 diabetes, especially with obesity. Saxagliptin and glitazones are contraindicated in HF.^[1]

Iron deficiency. IV iron is recommended in symptomatic HFrEF with iron deficiency, but data in HFpEF are scarce.^[1]

Complications and pitfalls

• Labelling a ventricle with any previous LVEF under 50% as HFpEF. The HFpEF definition requires LVEF never previously under 50%.^[1]
• Trusting a low NT-proBNP in obesity. Obesity is associated with lower peptide cut-off values, so interpret with caution.^[1]
• Missing amyloidosis. Increased wall thickness with red flags should raise suspicion of cardiac amyloidosis, particularly over 65 years; initial testing is serum and urine immunofixation, serum free light chains and DPD/PYP/HMDP bone scintigraphy.^[1]
• Stopping FMT for a creatinine bump. A creatinine rise under 50% above baseline, with eGFR above 15 mL/min/1.73 m², is acceptable after MRA or SGLT2 inhibitor initiation.^[1]
• Expecting a mortality benefit. SGLT2 inhibitors and MRAs reduce HF hospitalisation in HFpEF without proven CV death reduction.^[1]
• Over-treating blood pressure. Limited preload reserve makes hypotension harmful.^[1]

Prognosis

Five-year survival after HF onset is still under 60%.^[1] HFpEF mortality may be lower than in HFrEF, but the difference is small because HFpEF patients are much older, and hospitalisation risk is similar.^[1]

Risk climbs with repeated admissions: patients with repeated HF hospitalisation have particularly high mortality.^[1] Primary care data from England show that NT-proBNP above 2000 pg/mL carries a two-fold risk of early HF hospitalisation, so expedited specialist assessment is warranted.^[1]

^{[18] [20]}

After an HF admission, the outlook is poor across the ejection fraction range. In GWTG-HF linked to Medicare, risk-adjusted 5-year mortality was 75.7% in HFpEF and 75.3% in HFrEF. CV and HF readmission rates were higher in HFrEF and borderline EF than in HFpEF.^[18] In Olmsted County, survival improved over time for reduced EF but not for preserved EF.^[19]

Special populations

Older adults. The suggested outpatient NT-proBNP levels above which HF is likely rise with age, and elderly patients may present with more atypical symptoms of HF.^[1]

Chronic kidney disease. In a FINEARTS-HF secondary analysis (HF with LVEF of 40% or more), eGFR fell acutely by 2.9 mL/min/1.73 m² versus placebo over the first 3 months, yet finerenone reduced the risk of new-onset micro- and macroalbuminuria by 24% and 38%.^[1] Semaglutide and tirzepatide may lower eGFR acutely but protect kidney function long term. Severe CKD remains an evidence gap.^[1]

Women. Patients with HFpEF are more often female (55% in the Swedish HF registry).^[1] In PARAGON-HF, women showed a possible benefit from sacubitril–valsartan.^[9]

Evidence, guidelines and regional differences

^{[1] [2]}

The US document keeps HFmrEF as a separate class, with patients usually on a dynamic trajectory, either improving from HFrEF or deteriorating to HFrEF.^[2] It proposes adding evidence of spontaneous (at rest) or provokable (e.g. exercise, fluid challenge) increased LV filling pressures to the HFmrEF and HFpEF classifications.^[2] It notes that post hoc TOPCAT analyses have limitations. They suggest a possibility of benefit in appropriately selected symptomatic patients with LVEF of 45% or more, raised BNP or HF admission within 1 year, eGFR above 30, creatinine below 2.5 mg/dL and potassium below 5.0 mEq/L.^[2]

Open questions in the ESC list include optimal HFpEF drug therapy, GLP-1 RA and tirzepatide effects on morbidity and mortality, and validation of exercise PCWP cut-offs.^[1]

Australia and New Zealand (NHFA/CSANZ 2018)

The National Heart Foundation of Australia and the Cardiac Society of Australia and New Zealand (NHFA/CSANZ) published their Guidelines for the Prevention, Detection, and Management of Heart Failure in Australia in 2018; they were intended to replace the 2011 update.^[27] This section draws on the full guideline text published in Heart, Lung and Circulation, which gives guidance on the clinical care of adult patients with heart failure in Australia.^[27] Its recommendations were formulated with GRADE methodology, which gives the strength of a recommendation for or against an intervention; each row of the NHFA/CSANZ 2018 recommendation tables below condenses one selected recommendation and keeps its printed strength and quality of evidence.^[27] For topics with a limited evidence base, or where the impact of interventions on clinical outcomes was considered modest, the guideline includes comments in its practice advice sections.^[27] The guideline received final approval in June 2018, so it predates the AHA/ACC/HFSA 2022 and ESC 2026 guidelines used in the rest of this topic.^[27] Comparison rows set selected 2018 rows beside the matching ESC 2026 rows; any 2018 row that differs from a newer recommendation elsewhere in this topic should be read as history rather than current practice.^[27]

Defining and diagnosing HFpEF

NHFA/CSANZ 2018 defines HFpEF as all of the following: clinical symptoms with or without signs of heart failure; a measured EF of at least 50%; and objective evidence of either relevant structural heart disease or diastolic dysfunction.^[27] The structural heart disease or diastolic dysfunction must be without an alternative cause, for example significant valvular heart disease.^[27] Relevant structural heart disease means LV hypertrophy (increased LV wall thickness, or LV mass index above 115 g/m² in men or above 95 g/m² in women) or left atrial enlargement (left atrial volume index above 34 mL/m²).^[27] Diastolic dysfunction means high left-sided filling pressure documented by any of the following:^[27]

• invasive measurement, for example PCWP of 15 mm Hg or more, or LV end-diastolic pressure above 16 mm Hg;^[27]
• echocardiography with at least three of: mitral annular e′ below 7 cm/s (septal) / below 10 cm/s (lateral); average E/e′ above 14; left atrial volume index above 34 mL/m²; tricuspid regurgitation velocity above 2.8 m/s;^[27]
• biomarker analysis using the natriuretic peptide rule-in cut-offs (Table 4 of the guideline).^[27]

Because evidence of high filling pressure might only be present on exertion, exercise testing may be considered when clinical suspicion of HFpEF remains despite not meeting these criteria.^[27] With exercise, filling pressure is considered positive invasively if PCWP is above 25 mm Hg.^[27] On echocardiography it is positive when all of the following occur: average E/e′ above 14 or septal E/e′ above 15 and peak tricuspid regurgitation velocity above 2.8 m/s during or immediately after exercise, and septal e′ below 7 cm/s at baseline.^[27] As practice advice, noting that precise cut-offs vary between trials and with patient characteristics such as age, weight and renal function, the guideline proposes BNP below 100 ng/L and NT-proBNP below 300 ng/L as a pragmatic rule-out guide.^[27] Its Table 4 gives rule-in values of BNP above 400 ng/L and NT-proBNP above 450, 900 or 1800 ng/L for ages under 50, 50–75 and over 75 years.^[27] A footnote to that table says defining cut-off values, particularly for rule-in, is complicated and somewhat limited in accuracy because multiple factors influence natriuretic peptide levels.^[27] The guideline’s practice advice also warns that BNP and NT-proBNP levels are generally lower in HFpEF than in HFrEF, so their rule-out reliability is significantly weaker in suspected HFpEF.^[27] Other guidelines have used lower cut-offs in the ambulatory setting, but the trade-off is more false positives and unnecessary downstream testing.^[27]

^{[27] [27] [1]}

For ESC 2026 Table 10, the probability of an HFpEF diagnosis increases with the number of parameters in the pathological range.^[1]

LVEF 41–49%

NHFA/CSANZ 2018 did not recommend a separate mid-range (HFmrEF) category at that time: an LVEF of 41–49% is HFrEF with a mild reduction.^[27] In that range, additional criteria are required, for example signs of heart failure, or diastolic dysfunction with high filling pressure demonstrated by invasive means, echocardiography or biomarker testing.^[27] ESC 2026 has likewise removed HFmrEF and expanded HFrEF to include LVEF up to 50%, whereas AHA/ACC/HFSA 2022 keeps HFmrEF at LVEF 41–49%.^[1][2] The 2018 drug rows for a mild reduction in LVEF are all weak recommendations:^[27]

^{[27] [27] [1]}

These 2018 rows are superseded where newer guidelines differ: in ESC 2026, LVEF 41–49% is HFrEF, for which foundational medical therapy includes beta-blockers, ACE-Is/ARNIs/ARBs, MRAs and SGLT2-Is.^[1] ESC 2026 notes that no large prospective RCT has been performed exclusively at LVEF 41–49%, and that these patients appear to respond to FMT similarly to patients with LVEF below 40%.^[1] AHA/ACC/HFSA 2022 gives SGLT2 inhibitors a Class of Recommendation 2a in HFmrEF and weaker recommendations (Class 2b) for ARNi, ACEi, ARB, MRA and beta blockers.^[2]

The 2022 consensus statement says the 2018 NHFA/CSANZ heart failure guidelines were not scheduled for review.^[28] Given this and high clinician interest, an academic group (Evidence to Practice) facilitated a working group of clinicians with expertise in the diagnosis and management of heart failure in Australia to develop a consensus statement on drug studies published since the 2018 guidelines.^[28] The 2022 statement uses the label HFmrEF for LVEF 41–49%; its new HFmrEF recommendations, with the strength and quality of evidence it prints, are:^[28]

^[28]

Beside the 2022 SGLT2 inhibitor row, ESC 2026 recommends an SGLT2-I (dapagliflozin or empagliflozin) in symptomatic HF independent of LVEF, to reduce the risk of HFH or CV death (I, A).^[1] Beside the 2022 ACE inhibitor, ARNI and ARB row, ESC 2026, which counts LVEF 41–49% as HFrEF, recommends an ACE-I or ARNI (I, A), or an ARB if an ACE-I or ARNI is not tolerated (I, A), in symptomatic HFrEF.^[1] ESC 2026 also recommends intravenous iron in symptomatic HFrEF with iron deficiency to alleviate HF symptoms and improve QoL (I, B1), and says it should be considered to reduce the risk of HFH (IIa, B1).^[1]

Managing HFpEF

In HFpEF, the 2018 guideline sets the main aims of treatment as improving symptoms and quality of life and decreasing hospitalisation.^[27] It notes that none of the major RCTs of drug therapy in HFpEF conducted to that date had achieved their primary endpoint, although promising signals of fewer heart failure hospitalisations had been reported for ARBs and MRAs (CHARM-Preserved and TOPCAT).^[27] The 2022 Australian consensus statement confirms that the 2018 guidelines gave no specific treatment recommendations for HFpEF, because none of the major HFpEF trials had shown a significant benefit for their primary endpoints.^[28] Instead, the 2018 guideline gives practice advice:^[27]

• Diuretics are usually required to manage congestion, with careful attention to avoid over-diuresis.^[27]
• Loop diuretics are generally preferred, although thiazide diuretics are an alternative, especially if the patient is hypertensive.^[27]
• Comorbidities including hypertension, ischaemic heart disease, diabetes and AF should be identified and actively managed.^[27]
• The evidence for neurohormonal antagonists is less robust, but these agents are often used to manage comorbidities, and low-dose spironolactone may be considered to decrease hospitalisations for heart failure.^[27]
• In infiltrative cardiomyopathies such as cardiac amyloidosis, consider referral to specialised centres with expertise in this area.^[27]

^[27]

Where newer guidelines differ

Where the 2026 ESC guideline gives a different recommendation, the table sets the two side by side; read the 2018 column as history.^[27][1]

^{[27] [28] [1]}

The 2022 Australian consensus statement says an SGLT2 inhibitor (empagliflozin) should be considered in HFpEF to decrease cardiovascular mortality or hospitalisation for heart failure (strong recommendation for; moderate quality of evidence).^[28] AHA/ACC/HFSA 2022 gives SGLT2 inhibitors in HFpEF a Class of Recommendation 2a.^[2]

High-yield summary

• Classification: 2026 ESC has two phenotypes; HFmrEF is gone and LVEF 41–49% is HFrEF.^[1]
• Definition (all required): current or prior symptoms and/or signs + LVEF ≥50% (never previously under 50%) + objective structural and/or functional abnormalities consistent with LV diastolic dysfunction/raised LV filling pressures, supported by raised natriuretic peptides.^[1]
• Echo (Table 10): LV mass index ≥95 g/m² (female) or ≥115 g/m² (male), or RWT above 0.42; LAVI above 34 (sinus) or 40 (AF) mL/m²; E/e′ above 9 at rest; PASP above 35 mmHg or TR velocity at rest above 2.8 m/s.^[1]
• Gold standard: invasive haemodynamic exercise testing; PCWP ≥15 mmHg at rest, ≥25 mmHg on exercise, or LVEDP ≥16 mmHg at rest is generally considered diagnostic.^[1]
• FMT: SGLT2-I and MRA (sMRA for HFrEF; sMRA/nsMRA for HFpEF), Class I, level A in symptomatic HF to reduce HF hospitalisation or CV death; in HFpEF the benefit is driven by fewer HF hospitalisations.^[1]
• Trials: DELIVER HR 0.82; EMPEROR-Preserved HR 0.79; FINEARTS-HF rate ratio 0.84; TOPCAT HR 0.89 (P=0.14); PARAGON-HF rate ratio 0.87 (P=0.06).^{[3][4][6][7][9]}
• Obesity: semaglutide/tirzepatide Class IIa, level B1 for symptomatic HF with LVEF ≥45% and BMI ≥30 kg/m², to reduce weight and improve exercise capacity and QoL. Separately, SUMMIT reported HR 0.62 for CV death or worsening HF; the ESC calls the HF-event data a suggested potential benefit.^[1][12]
• Shunts: RELIEVE-HF showed more CV events with a shunt in its preserved-LVEF stratum (LVEF >40%).^[14]
• Beta-blockers: in the 11-trial individual-patient analysis, in sinus rhythm the mortality reduction did not extend to the small subgroup with LVEF of 50% or more; withdrawal raised peak VO₂ in PRESERVE-HR.^[21][22]
• Prognosis after admission: risk-adjusted 5-year mortality 75.7% in HFpEF versus 75.3% in HFrEF (GWTG-HF linked to Medicare).^[18]
• Evidence levels: B1 means suggestive evidence, usually from at least one adequately powered RCT free from major bias, or a meta-analysis of such RCTs, with some evidence against chance (e.g. P <.05 for superiority); B2 means limited evidence from 2 or more adequately powered non-randomised studies with careful bias control, or a meta-analysis of small, underpowered RCTs.^[24]

References28ShowHide

1. [1]Køber L, Adamo M, Ruwald AC, et al. 2026 ESC Guidelines for the management of heart failure. Eur Heart J, 2026.PMID 42661420
2. [2]Heidenreich PA, Bozkurt B, Aguilar D, et al. 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation, 2022.PMID 35363499
3. [3]Solomon SD, McMurray JJV, Claggett B, et al. Dapagliflozin in Heart Failure with Mildly Reduced or Preserved Ejection Fraction. N Engl J Med, 2022.PMID 36027570
4. [4]Anker SD, Butler J, Filippatos G, et al. Empagliflozin in Heart Failure with a Preserved Ejection Fraction. N Engl J Med, 2021.PMID 34449189
5. [5]Vaduganathan M, Docherty KF, Claggett BL, et al. SGLT-2 inhibitors in patients with heart failure: a comprehensive meta-analysis of five randomised controlled trials. Lancet, 2022.PMID 36041474
6. [6]Solomon SD, McMurray JJV, Vaduganathan M, et al. Finerenone in Heart Failure with Mildly Reduced or Preserved Ejection Fraction. N Engl J Med, 2024.PMID 39225278
7. [7]Pitt B, Pfeffer MA, Assmann SF, et al. Spironolactone for heart failure with preserved ejection fraction. N Engl J Med, 2014.PMID 24716680
8. [8]Jhund PS, Talebi A, Henderson AD, et al. Mineralocorticoid receptor antagonists in heart failure: an individual patient level meta-analysis. Lancet, 2024.PMID 39232490
9. [9]Solomon SD, McMurray JJV, Anand IS, et al. Angiotensin-Neprilysin Inhibition in Heart Failure with Preserved Ejection Fraction. N Engl J Med, 2019.PMID 31475794
10. [10]Kosiborod MN, Abildstrøm SZ, Borlaug BA, et al. Semaglutide in Patients with Heart Failure with Preserved Ejection Fraction and Obesity. N Engl J Med, 2023.PMID 37622681
11. [11]Kosiborod MN, Petrie MC, Borlaug BA, et al. Semaglutide in Patients with Obesity-Related Heart Failure and Type 2 Diabetes. N Engl J Med, 2024.PMID 38587233
12. [12]Packer M, Zile MR, Kramer CM, et al. Tirzepatide for Heart Failure with Preserved Ejection Fraction and Obesity. N Engl J Med, 2025.PMID 39555826
13. [13]Kosiborod MN, Deanfield J, Pratley R, et al. Semaglutide versus placebo in patients with heart failure and mildly reduced or preserved ejection fraction: a pooled analysis of the SELECT, FLOW, STEP-HFpEF, and STEP-HFpEF DM randomised trials. Lancet, 2024.PMID 39222642
14. [14]Stone GW, Lindenfeld J, Rodés-Cabau J, et al. Interatrial Shunt Treatment for Heart Failure: The Randomized RELIEVE-HF Trial. Circulation, 2024.PMID 39308371
15. [15]Shah SJ, Borlaug BA, Chung ES, et al. Atrial shunt device for heart failure with preserved and mildly reduced ejection fraction (REDUCE LAP-HF II): a randomised, multicentre, blinded, sham-controlled trial. Lancet, 2022.PMID 35120593
16. [16]Reddy YNV, Kaye DM, Handoko ML, et al. Diagnosis of Heart Failure With Preserved Ejection Fraction Among Patients With Unexplained Dyspnea. JAMA Cardiol, 2022.PMID 35830183
17. [17]Paulus WJ, Tschöpe C A novel paradigm for heart failure with preserved ejection fraction: comorbidities drive myocardial dysfunction and remodeling through coronary microvascular endothelial inflammation. J Am Coll Cardiol, 2013.PMID 23684677
18. [18]Shah KS, Xu H, Matsouaka RA, et al. Heart Failure With Preserved, Borderline, and Reduced Ejection Fraction: 5-Year Outcomes. J Am Coll Cardiol, 2017.PMID 29141781
19. [19]Owan TE, Hodge DO, Herges RM, et al. Trends in prevalence and outcome of heart failure with preserved ejection fraction. N Engl J Med, 2006.PMID 16855265
20. [20]Tsao CW, Lyass A, Enserro D, et al. Temporal Trends in the Incidence of and Mortality Associated With Heart Failure With Preserved and Reduced Ejection Fraction. JACC Heart Fail, 2018.PMID 30007560
21. [21]Cleland JGF, Bunting KV, Flather MD, et al. Beta-blockers for heart failure with reduced, mid-range, and preserved ejection fraction: an individual patient-level analysis of double-blind randomized trials. Eur Heart J, 2018.PMID 29040525
22. [22]Palau P, Seller J, Domínguez E, et al. Effect of β-Blocker Withdrawal on Functional Capacity in Heart Failure and Preserved Ejection Fraction. J Am Coll Cardiol, 2021.PMID 34794685
23. [23]Yusuf S, Pfeffer MA, Swedberg K, et al. Effects of candesartan in patients with chronic heart failure and preserved left-ventricular ejection fraction: the CHARM-Preserved Trial. Lancet, 2003.PMID 13678871
24. [24]Jüni P, Antoniou S, Arbelo E, et al. 2024 Revision of the level of evidence grading system for ESC clinical practice guideline recommendations I: therapy and prevention. Eur Heart J, 2025.PMID 40116721
25. [25]Di Angelantonio E, Pennells L, Abdelhamid M, et al. 2024 Revision of the level of evidence grading system for ESC clinical practice guideline recommendations II: diagnostic tests and prediction models. Eur Heart J, 2025.PMID 40116738
26. [26]Sebastian SA, Padda I, Johal G Supervised exercise training in heart failure with preserved ejection fraction: A systematic review and meta-analysis of randomized controlled trials. Curr Probl Cardiol, 2024.PMID 38311273
27. [27]Atherton JJ, Sindone A, De Pasquale CG, et al. National Heart Foundation of Australia and Cardiac Society of Australia and New Zealand: Guidelines for the Prevention, Detection, and Management of Heart Failure in Australia 2018. Heart Lung Circ, 2018.PMID 30077227
28. [28]Sindone AP, De Pasquale C, Amerena J, et al. Consensus statement on the current pharmacological prevention and management of heart failure. Med J Aust, 2022.PMID 35908234