Guide détaillé à venir
Nous préparons un guide éducatif complet pour le LVEF Estimation Guide. Revenez bientôt pour des explications étape par étape, des formules, des exemples concrets et des conseils d'experts.
Left Ventricular Ejection Fraction (LVEF) is the percentage of blood ejected from the left ventricle with each heartbeat. It is the single most important quantitative parameter of cardiac systolic function and underpins diagnosis, risk stratification, and treatment decisions across a wide spectrum of cardiovascular conditions including heart failure, coronary artery disease, cardiomyopathy, valvular heart disease, and cardiotoxic chemotherapy monitoring. Visual estimation of LVEF — colloquially called the 'eyeball' method — is the rapid, intuitive assessment of LVEF performed by an experienced echocardiographer or cardiologist by examining the wall motion and cavity size change between systole and diastole on real-time 2D echocardiography or cardiac MRI. Whilst not as mathematically precise as the Biplane Simpson's method (which traces endocardial borders in two planes), visual estimation by an experienced observer is remarkably accurate — studies show that experienced echocardiographers' visual estimates correlate closely with quantitative measurements, often within 5–10 percentage points. EF classification follows internationally agreed thresholds from the ESC, ACC/AHA, and ASE guidelines. Heart failure with reduced ejection fraction (HFrEF) is defined as LVEF below 40%, and carries the greatest mortality burden; heart failure with mildly reduced ejection fraction (HFmrEF) covers 40–49%; and heart failure with preserved ejection fraction (HFpEF) is defined by LVEF at or above 50% with symptoms of heart failure and diastolic dysfunction evidence. The 35% threshold is critical in device therapy: patients with LVEF ≤ 35% who remain symptomatic (NYHA class II–III) despite optimal medical therapy for at least 3 months are eligible for implantable cardioverter-defibrillator (ICD) implantation and/or cardiac resynchronisation therapy (CRT) if left bundle branch block is present.
Visual LVEF Estimation (Qualitative Categories):
Hyperdynamic: LVEF > 70% — cavity nearly obliterates in systole
Normal: LVEF 55–70% — cavity reduces by approximately half
Mildly reduced: LVEF 45–54% — reduced but definite contraction visible
Moderately reduced: LVEF 35–44% — clearly impaired, walls move but cavity barely changes
Severely reduced: LVEF < 35% — minimal wall motion, large static cavity
Biplane Simpson's Method (quantitative reference standard):
LVEF (%) = [(EDV − ESV) / EDV] × 100
Where:
EDV = End-Diastolic Volume (mL) — traced in A4C and A2C views at mitral valve opening
ESV = End-Systolic Volume (mL) — traced at smallest cavity size
Fractional Shortening (M-mode approximation):
FS (%) = [(LVEDD − LVESD) / LVEDD] × 100
Normal FS: 25–45% (approximate LVEF correlation: FS × 2 ≈ LVEF for spherical ventricles)- 1Acquire standard echocardiographic views: parasternal long axis (PLAX), parasternal short axis (PSAX) at papillary muscle level, apical four-chamber (A4C), and apical two-chamber (A2C). Ensure frame rate is adequate (>50 frames/second) to capture end-systole accurately.
- 2For visual estimation, observe 3–5 cardiac cycles in real time. Focus on: (1) degree of endocardial inward motion during systole, (2) systolic wall thickening (normal myocardium thickens by >50% in systole), and (3) proportional change in cavity size from diastole to systole.
- 3Classify LVEF into one of five visual categories: hyperdynamic (>70%), normal (55–70%), mildly reduced (45–54%), moderately reduced (35–44%), or severely reduced (<35%). Report both the category and an estimated percentage range (e.g., 'moderately reduced, estimated 38–42%').
- 4For Biplane Simpson's method (the recommended quantitative technique per ASE/EACVI guidelines), trace the endocardial border in A4C at end-diastole (onset of QRS or maximum cavity size) and A2C at end-diastole, then repeat at end-systole (minimum cavity size, typically 1 frame after aortic valve closure). The software divides the ventricle into stacked discs and sums volumes.
- 5Consider the degree of regional wall motion abnormalities (RWMA). In segmental disease (e.g., post-MI), the visual LVEF estimate must account for non-contracting segments. Regional hypokinesia, akinesia, or dyskinesia in one territory with preserved function elsewhere can result in a ventricle that 'looks' near-normal globally but has a meaningfully reduced LVEF. Simpson's method is superior in this setting.
- 6Compare with prior studies to assess trajectory. A LVEF that has dropped from 65% to 48% (mildly reduced) carries different clinical urgency than a stable LVEF of 48% of unknown duration. Trajectory is especially important in chemotherapy-related cardiomyopathy monitoring (cardio-oncology protocols define a >10% absolute fall as significant even if still within 'mildly reduced' range).
- 7Apply clinical thresholds: LVEF < 40% = HFrEF (guideline-directed medical therapy with ACE inhibitors/ARNi, beta-blockers, MRA, SGLT2 inhibitors); LVEF ≤ 35% after 3+ months optimal therapy = assess ICD/CRT eligibility; LVEF < 50% in valvular disease may trigger intervention even if asymptomatic (e.g., severe aortic regurgitation or mitral regurgitation).
No device therapy threshold met
Symmetric wall thickening and a cavity that reduces to approximately half its size in systole is the hallmark of a normal LVEF in the 55–70% range. No intervention is needed based on LVEF alone. Diastolic function should also be assessed in hypertensive patients as they are at risk of HFpEF.
LVEF ≤ 35% after ≥3 months optimal therapy — ICD + CRT (if LBBB) indicated
A visually severely reduced LVEF with a large, barely contracting ventricle should always be quantified with Biplane Simpson's before device therapy decisions. At 28%, this patient is well below the 35% ICD threshold and has been on optimal medical therapy for 4 months. LBBB assessment with 12-lead ECG is needed to determine CRT eligibility (QRS >150 ms LBBB pattern).
High LVEF in shock = distributive (septic) not cardiogenic cause
A hyperdynamic LVEF in the context of shock indicates preserved (or enhanced) myocardial contractility, pointing toward distributive or obstructive shock rather than cardiogenic shock. This has critical management implications — vasopressors are appropriate; inotropes would be harmful. Point-of-care echo (POCUS) for visual LVEF assessment is central to shock haemodynamic phenotyping in the ICU.
Recheck LVEF in 3 months on GDMT before device therapy decision
Post-MI LVEF assessment at 5 days represents an acute measurement that may underestimate long-term function due to myocardial stunning (reversibly dysfunctional but viable myocardium). Guidelines recommend repeat LVEF assessment after 3 months of guideline-directed medical therapy (GDMT) before making ICD implantation decisions, as stunning recovery may raise LVEF above the 35% threshold.
Emergency and ICU POCUS (point-of-care ultrasound) — rapid visual LVEF categorisation (hyperdynamic/normal/reduced) to phenotype shock and guide resuscitation decisions within minutes
Cardiology outpatient clinics — serial LVEF monitoring in heart failure patients to assess response to guideline-directed medical therapy and determine device therapy eligibility
Cardio-oncology programmes — baseline and serial LVEF surveillance in cancer patients receiving anthracyclines, trastuzumab, or other cardiotoxic agents per chemotherapy monitoring protocols
Pre-operative risk assessment — LVEF below 35–40% significantly increases perioperative cardiac risk; anaesthetic and surgical planning must account for reduced contractile reserve
Cardiac device clinics — LVEF measurement is a prerequisite for ICD and CRT implantation decisions and is reassessed at device generator changes and in response to clinical deterioration
Post-MI LVEF — timing matters for device decisions
LVEF measured within the first 48–72 hours of STEMI may be falsely low due to myocardial stunning — reversible contractile dysfunction of viable myocardium that recovers over days to weeks after reperfusion. Guidelines recommend waiting at least 40 days post-MI before reassessing LVEF for ICD eligibility, and at least 3 months of optimal medical therapy for non-ischaemic cardiomyopathy. Implanting an ICD based on the acute post-MI LVEF risks unnecessary device implantation in patients whose LVEF will recover.
LVEF in valvular heart disease — thresholds differ
In chronic valvular heart disease, the LVEF thresholds for intervention are different from those used in heart failure guidelines. For severe aortic regurgitation or mitral regurgitation, surgical intervention is recommended at LVEF < 60% (not < 40%) because the volume-overloaded ventricle has supra-normal preload that artificially inflates apparent LVEF. Similarly, for severe aortic stenosis, LVEF < 50% is the threshold for intervention, and 'paradoxical low-flow low-gradient' AS (preserved LVEF but low stroke volume) requires careful haemodynamic assessment.
Cardio-oncology monitoring — different LVEF thresholds
For patients receiving cardiotoxic chemotherapy (anthracyclines, trastuzumab, tyrosine kinase inhibitors), cardio-oncology guidelines define cardiotoxicity as a fall in LVEF of >10 percentage points to below 53% (European guideline) or >5 percentage points to below 55% (some North American protocols). Serial echo monitoring at defined intervals is mandated. A fall meeting cardiotoxicity criteria may necessitate holding chemotherapy, initiating cardioprotective therapy (ACEi, beta-blocker), and urgent cardiology review — even if the absolute LVEF remains above 40%.
Technically difficult echocardiographic windows
In patients with severe obesity, COPD (hyperinflated lungs displacing cardiac axis), post-cardiac surgery (pericardial effusion, pericardial scar), or mechanical ventilation, acoustic windows may be insufficient for reliable LVEF estimation. In these cases, echocardiographic contrast agents (ultrasound contrast) can be used to enhance endocardial border definition and dramatically improve LVEF measurement accuracy. If contrast echo is inadequate, cardiac MRI or radionuclide ventriculography (MUGA scan) are appropriate alternatives.
Biventricular function — RVEF also matters
LVEF assessment should always be accompanied by evaluation of right ventricular (RV) size and function, as RV dysfunction carries important independent prognostic information. RV dysfunction in the setting of HFrEF is associated with significantly worse outcomes and may preclude isolated left-sided mechanical circulatory support. Quantitative RV parameters include TAPSE (tricuspid annular plane systolic excursion, normal >17 mm), fractional area change (FAC, normal >35%), and RV free wall longitudinal strain on speckle tracking.
| LVEF Category | LVEF Range | HF Classification | Key Clinical Actions |
|---|---|---|---|
| Hyperdynamic | > 70% | Not HFrEF — consider obstructive/distributive pathology | Exclude HOCM, high-output states (sepsis, thyrotoxicosis, AV fistula) |
| Normal | 55–70% | HFpEF if symptomatic with raised filling pressures | Assess diastolic function; treat underlying hypertension/AF |
| Mildly reduced | 45–54% | HFmrEF | Initiate HFrEF-type GDMT; reassess in 3–6 months |
| Moderately reduced | 35–44% | HFrEF | Full GDMT; reassess LVEF at 3 months before ICD decision |
| Severely reduced | < 35% | HFrEF — device eligible | ICD ± CRT if LBBB and NYHA II–III; full GDMT mandatory |
| Critical / Cardiogenic shock | < 20% | End-stage / refractory HF | MCS (IABP, Impella, ECMO), transplant/VAD referral |
What is a normal LVEF?
A normal LVEF is generally considered 55–70% by most cardiology society guidelines including the ASE, ESC, and ACC/AHA. Some guidelines define the lower limit of normal as 52% in men and 54% in women based on gender-specific reference ranges. An LVEF above 70% may be normal in young, fit individuals or hyperdynamic states (sepsis, anaemia, thyrotoxicosis), but can also reflect hypertrophic cardiomyopathy with dynamic outflow obstruction.
What LVEF qualifies a patient for an ICD?
A LVEF of 35% or less (≤35%), measured after at least 3 months of optimal guideline-directed medical therapy in a patient with symptomatic heart failure (NYHA class II or III), is the primary criterion for primary prevention ICD implantation. This threshold is derived from landmark trials including MADIT-II (LVEF ≤30%), MADIT-CRT (LVEF ≤30%), and SCD-HeFT (LVEF ≤35%). Importantly, the LVEF must be assessed at 90+ days from any acute event (MI, new diagnosis) to exclude transient stunning.
What is the Biplane Simpson's method and why is it the reference standard?
The Biplane Simpson's method (modified Simpson's rule) calculates LVEF by dividing the left ventricular cavity into a stack of thin discs in two orthogonal apical views (4-chamber and 2-chamber). The summed disc volumes for end-diastole and end-systole are used to calculate LVEF. It is the reference standard because it makes no geometric assumptions about ventricular shape — unlike linear M-mode measurements that assume the ventricle is a prolate ellipsoid. This matters especially in patients with regional wall motion abnormalities, aneurysms, or remodelled ventricles where shape assumptions fail.
How accurate is visual LVEF estimation?
In experienced hands, visual LVEF estimation correlates highly with quantitative measurements, typically within 5–10 percentage points. Studies comparing experienced echocardiographers' visual estimates to cardiac MRI (the gold standard) show mean absolute errors of around 5–8%. Visual estimation is faster, does not require good endocardial border definition, and is more robust in technically difficult images. However, it is more operator-dependent and less reproducible than quantitative methods, particularly for borderline LVEF values (e.g., distinguishing 38% from 42%).
What is the difference between HFrEF, HFmrEF, and HFpEF?
These are LVEF-based classifications of heart failure: HFrEF (heart failure with reduced ejection fraction) is LVEF < 40% and has the strongest evidence base for pharmacological therapy (ACEi/ARNi, beta-blockers, MRA, SGLT2i). HFmrEF (mildly reduced, 40–49%) has emerging evidence supporting HFrEF-type medications. HFpEF (preserved, LVEF ≥ 50%) requires symptoms plus evidence of structural heart disease or elevated filling pressures — SGLT2 inhibitors (empagliflozin, dapagliflozin) have now demonstrated mortality benefit in HFpEF based on EMPEROR-Preserved and DELIVER trials.
When is cardiac MRI preferred over echo for LVEF measurement?
Cardiac MRI (CMR) is considered the gold standard for LVEF measurement due to its superior spatial resolution, lack of acoustic windows issues, and ability to characterise myocardial tissue (fibrosis via late gadolinium enhancement, oedema, infiltration). CMR is preferred when: (1) echocardiographic image quality is poor (obesity, COPD, post-surgical), (2) a precise LVEF is needed for device therapy decisions near the threshold (e.g., estimated 33–37% on echo), (3) cardiomyopathy aetiology is uncertain, or (4) arrhythmogenic right ventricular cardiomyopathy is suspected.
Can LVEF improve after treatment?
Yes — LVEF recovery is well documented and clinically significant. In ischaemic cardiomyopathy, myocardial stunning (reversible after revascularisation) and hibernation (chronically under-perfused but viable muscle that recovers with PCI/CABG) can both contribute to LVEF recovery. In dilated cardiomyopathy, guideline-directed medical therapy (ARNi, beta-blockers, MRA, SGLT2i) produces LVEF recovery in up to 40% of patients within 6–12 months — a phenomenon called 'reverse remodelling'. Chemotherapy-induced cardiomyopathy can also recover with cardioprotective therapy, particularly if detected early.
What is 3D echocardiography and is it more accurate than 2D?
Three-dimensional (3D) echocardiography uses a matrix transducer to acquire a full volumetric dataset of the left ventricle, allowing LVEF measurement without geometric assumptions and without dependence on two-plane orthogonality. Studies comparing 3D echo to cardiac MRI show lower bias and better reproducibility than 2D Biplane Simpson's, particularly in distorted ventricles. However, 3D echo requires higher image quality (good acoustic window), specialised software, and expertise. ASE/EACVI guidelines now recommend 3D echo as the preferred echocardiographic method when image quality is sufficient, especially for serial LVEF monitoring.
Conseil Pro
When a LVEF is borderline (estimated visually at 33–38%), always quantify it with Biplane Simpson's method and request cardiac MRI before making irreversible device therapy decisions. At the 35% threshold, the difference between 33% and 37% is clinically decisive — an ICD will be implanted for one and not the other. Do not rely solely on visual estimation for threshold-level decisions. Document the method used for every LVEF measurement.
Le saviez-vous?
The concept of ejection fraction was introduced by Arnie Katz and Edmund Cobbs in the 1960s using cardiac catheterisation and contrast ventriculography. For decades, LVEF could only be measured invasively. The advent of echocardiography in the 1970s and 1980s made non-invasive LVEF measurement routine, and the Biplane Simpson's method — adapted from Simpson's 1828 mathematical rule for estimating area under a curve — became the echocardiographic standard. Today, artificial intelligence-based automated LVEF measurement from echo loops performs comparably to experienced human echocardiographers and is increasingly embedded in clinical echo software.
Références
- ›Lang RM et al. Recommendations for Cardiac Chamber Quantification by Echocardiography in Adults — JASE/EACVI 2015
- ›McDonagh TA et al. 2021 ESC Guidelines for the Diagnosis and Treatment of Acute and Chronic Heart Failure — European Heart Journal 2021
- ›Al-Khatib SM et al. 2017 AHA/ACC/HRS Guideline for Management of Patients With Ventricular Arrhythmias and Prevention of Sudden Cardiac Death — JACC 2018
- ›Pellikka PA et al. Guidelines for Performance, Interpretation, and Application of Stress Echocardiography — JASE 2020
- ›Lyon AR et al. 2022 ESC Guidelines on Cardio-Oncology — European Heart Journal 2022