HFrEF — Heart Failure with Reduced Ejection Fraction
Contents. #Details · #Key Facts · #Pathophysiology · #Diagnosis · #HFimpEF · #Contradictions / Open Questions · #Connections · #Sources
Management. The treatment pathway — four pillars, additional agents, revascularisation, ICD, CRT and TEER — is on concepts/HFrEF-Management.
Details
HFrEF is the phenotype of heart failure defined by LVEF <50% with symptoms and/or signs — the 2026 ESC guideline moved the boundary up from 40% and absorbed the old HFmrEF band into it. It is characterised by reduced myocardial contractility, eccentric LV dilatation and neurohormonal activation, accounts for about 60% of HF in outpatient registries, and carries one-year mortality around 20% and 5-year mortality of 53–67%. It has by far the largest evidence base of the three phenotypes: the foundational medical therapy — SGLT2i, MRA, beta-blocker and ACE-I/ARNI/ARB — reduces estimated all-cause mortality by about 73% versus no treatment, and HFrEF is the only phenotype in which device therapy has randomised mortality evidence. HFimpEF, LVEF that recovers by ≥10% to above 40% on treatment, retains structural risk and requires indefinite continuation of therapy. (sources/HF-ESC-2026, rating: very high; sources/HF-AHA-2022)
Part of entities/Heart-Failure — the parent overview, which covers the syndrome definition, staging, decompensated HF, advanced HF and special populations.
Key Facts
- LVEF <50% now defines HFrEF in the 2026 ESC guideline. The old 41–49% band has no dedicated trial behind it — the recommendation is extrapolated from HFrEF populations. (sources/HF-ESC-2026, rating: very high)
- HFrEF is the commonest phenotype in outpatient registries at about 60%, with one-year mortality around 20% and five-year mortality 53–67%. (sources/HF-ESC-2021, rating: very high)
- Foundational medical therapy — SGLT2i, MRA, beta-blocker and ACE-I/ARNI/ARB — each carries Class I / COR 1A and together reduce estimated all-cause mortality by ~73%. SGLT2i and MRA are Class I at any ejection fraction; the beta-blocker and renin-axis inhibitor are HFrEF-specific, which is what separates this pathway from concepts/HFpEF-Management. (sources/HF-AHA-2022)
- HFrEF is the only heart failure phenotype in which device therapy — ICD and CRT — has randomised mortality evidence. (sources/HF-ESC-2026, sources/HF-AHA-2022)
- Do not stop treatment when the LVEF recovers. Withdrawal in HFimpEF produced 40% relapse within six months. (sources/HF-ESC-2026, sources/HF-AHA-2022)
Pathophysiology
- Neurohormonal activation. Reduced output triggers RAAS and sympathetic activation. Initially compensatory, chronic excess drives maladaptive remodelling — hypertrophy, apoptosis, interstitial fibrosis and progressive dilatation. Natriuretic peptides are released in proportion to wall stress. (sources/HF-ESC-2021, rating: very high)
- Eccentric remodelling. Volume-overload geometry: sarcomere replication in series produces chamber dilatation with proportional wall thinning, and the Frank-Starling curve shifts right and flattens. GDMT reverses this — the HFimpEF phenotype — but structural risk persists after LVEF normalises. (sources/HF-AHA-2022)
- Aldosterone escape despite ACE inhibition — the mechanistic basis for MRA. ACE inhibitors suppress aldosterone only transiently; it escapes through angiotensin-independent pathways, including direct potassium stimulation of the adrenal zona glomerulosa, ACTH, and locally generated angiotensin II within myocardium. Chronically elevated aldosterone then drives myocardial and vascular fibrosis, baroreceptor dysfunction, sympathetic activation and impaired norepinephrine reuptake — effects additive to and distinct from ACE inhibitor haemodynamics. This is why MRA benefit in RALES was independent of the ACE inhibitor effect. (sources/mra-hfref-rales-nejm-1999, rating: very high)
- Electrical remodelling — ↑INaLate (CaMKII-driven), ↑If (HCN4 upregulation in ventricular myocytes), ↑NCX1, ↓IKr/IKs/Ito/IK1 — raises arrhythmic susceptibility and sudden-death risk independently of LVEF, and lowers repolarisation reserve so that QT-prolonging drugs carry elevated torsades risk. This is the substrate that makes ICD therapy a disease-specific intervention rather than a generic one. Full ionic detail: concepts/Electrical-Remodeling. (sources/membrane-potential-physrev-2021, rating: very high)
Diagnosis
- LVEF <50% on echocardiography or CMR, in a patient meeting clinical HF criteria (symptoms, signs, and a cardiac structural or functional abnormality). The 2026 ESC cut-off replaced ≤40%. (sources/HF-ESC-2026, rating: very high; sources/HF-AHA-2022)
- Natriuretic peptides support the diagnosis and grade severity. In obesity, BNP is physiologically lower — a normal value does not exclude HFrEF in an obese patient. (sources/HF-ESC-2021, sources/HF-AHA-2022)
- HFimpEF (AHA 2022): prior LVEF ≤40% with follow-up LVEF >40% on GDMT. Distinct from "recovered EF" — structural abnormality and relapse risk persist, so GDMT continues indefinitely (COR 1, LOE B-R). (sources/HF-AHA-2022)
HFimpEF
- Definition: an absolute LVEF rise of ≥10% to above 40% after treatment. Deliberately framed as remission rather than recovery — structural abnormalities, biomarker abnormalities and relapse risk all persist. (sources/HF-AHA-2022, rating: very high; sources/HF-ESC-2026)
- Continue GDMT (COR 1, LOE B-R): withdrawal produced 40% relapse within six months, and only 50% of attempted withdrawals succeeded long term. The ESC adds one narrow exception — gradual withdrawal may be considered (IIb/C) in highly selected patients with complete normalisation of LV function, volume and natriuretic peptides after treatment of a reversible cause. (sources/HF-AHA-2022, sources/HF-ESC-2026)
- TTNtv-associated DCM shows the highest reverse-remodelling rate (~70%). See entities/TTN and entities/DCM.
Contradictions / Open Questions
- A June 2026 institutional handbook still treats LVEF 41–49% as HFmrEF. Version 9.1 of the Hospital Authority handbook carries the four-category AHA 2022 scheme, so the band this page now calls HFrEF remains a separate phenotype in the operational guidance of a health system. Its HFrEF section otherwise matches — four foundational drugs, ARNI after a 36-hour ACE-I washout, ICD at LVEF ≤35%, CRT-D at QRS ≥150 ms or LBBB 130–149 ms — which makes the classification the sole point of departure. (sources/Handbook-HA-2024, rating: high)
- The 41–49% band is now HFrEF without its own evidence. The 2026 Task Force absorbed HFmrEF into HFrEF on similarity of mechanism and treatment response, but states plainly that no large prospective RCT has been performed exclusively at LVEF 41–49%. The beta-blocker, ACE-I/ARNI and ARB recommendations carry a footnote to that effect. Clinically the extrapolation is reasonable; evidentially it is an extrapolation, and the previous ESC 2023 Class I SGLT2i grade for HFmrEF now survives only as an inference. (sources/HF-ESC-2026, rating: very high)
Connections
- Part of entities/Heart-Failure — the parent overview
- Management: concepts/HFrEF-Management — the four pillars, additional agents, device therapy and revascularisation; all therapy-level contradictions live there
- Related to entities/HFpEF — the preserved-EF counterpart; beta-blocker advice is inverted between the two
- Related to concepts/HFpEF-Management — the other pathway, which shares the SGLT2i and MRA pillars and nothing after them
- Related to entities/DCM, entities/TTN and entities/LMNA — nonischaemic aetiology, reverse remodelling, and ICD with LVEF >35%
- Related to concepts/Electrical-Remodeling — the ionic substrate linking reduced EF to arrhythmia and sudden death
- Related to concepts/Cardiorenal-Syndrome — MRA hyperkalaemia, diuretic resistance, the eGFR ceilings on quadruple therapy
- Related to entities/ICD, entities/CRT and concepts/Conduction-System-Pacing — device detail beyond the management summary
- Related to concepts/HF-COPD-Comorbidity — bisoprolol selection and diagnostic overlap