What Ejection Fraction Measures

Ejection fraction answers a narrow but useful question: what percentage of the blood present in a ventricle just before contraction leaves by the end of contraction? The left-ventricular ejection fraction, or LVEF, is reported most often because the left ventricle supplies the systemic circulation.

The number helps describe systolic performance, classify research populations, and guide some treatments, yet it can be preserved in symptomatic heart failure, vary with blood pressure and method, and obscure low forward flow when the chamber is small. A report is best read with volumes, rhythm, and valve findings. It is best read with wall motion, filling measures, and symptoms. Examination and other tests belong there too.

The calculation behind the percentage#

End-diastolic volume is the amount of blood in a ventricle at its fullest point, just before contraction. End-systolic volume is the amount remaining after contraction. Stroke volume is their difference:

stroke volume = end-diastolic volume - end-systolic volume

Ejection fraction is:

ejection fraction = stroke volume / end-diastolic volume x 100

If a left ventricle fills to 120 milliliters and 50 milliliters remain after contraction, stroke volume is 70 milliliters and LVEF is about 58%, so the chamber did not empty completely, and complete emptying is not expected.

Two people can have the same LVEF and different stroke volumes. A small ventricle with an end-diastolic volume of 70 milliliters and LVEF of 60% ejects 42 milliliters. A larger ventricle at 140 milliliters and the same LVEF ejects 84 milliliters. Heart rate then determines cardiac output: stroke volume multiplied by beats per minute. So LVEF is not interchangeable with cardiac output, blood pressure, contractility, or overall circulatory adequacy, because each of those describes a different feature.

How echocardiography estimates LVEF#

Transthoracic echocardiography is the usual first method because it is noninvasive, portable, widely available, and provides much more than LVEF; it can show chamber size, wall thickness and motion, valve structure, blood-flow velocities, pericardial fluid, right-heart features, and clues about filling pressure.

For two-dimensional imaging, professional chamber-quantification recommendations favor the biplane method of disks when image quality permits, and a reader traces the endocardial border in apical four-chamber and two-chamber views at end-diastole and end-systole. Software models the volume as a stack of disks, then calculates LVEF.

Foreshortening can make the ventricle look shorter and distort volumes. Poor border definition, body habitus, lung disease, rapid or irregular rhythm, and suboptimal positioning can reduce confidence. Contrast agents may improve cavity definition in selected studies. Three-dimensional echocardiography can avoid some geometric assumptions and may improve reproducibility when images are adequate.

Some reports use visual estimation. Experienced readers can make useful estimates, especially when tracing is difficult, but if you see a visual range such as 45% to 50%, do not treat it as an exact laboratory value. The report should indicate the method and limitations.

Other imaging methods#

Cardiac magnetic resonance, often abbreviated CMR, calculates volumes from a stack of cine images covering the ventricle; it is frequently treated as a reference method for ventricular volumes and can be more reproducible than two-dimensional echocardiography. It also characterizes tissue, scar, edema, congenital anatomy, and right-ventricular structure in appropriate protocols.

CMR is not automatically the right test for everyone. Availability, scan time, and implanted-device compatibility affect feasibility. So do claustrophobia, ability to lie still, rhythm, and the need for contrast or other sequences. The interpretation also depends on how borders, papillary muscles, and trabeculations are handled.

Gated nuclear blood-pool imaging, myocardial perfusion studies, cardiac computed tomography, and contrast ventriculography can estimate LVEF too. Each uses its own acquisition, temporal resolution, assumptions, and processing. Values from different modalities are correlated but not identical. Longitudinal follow-up is easiest to interpret when method and laboratory are consistent, provided the clinical question has not changed.

Preserved does not mean normal in every respect#

A preserved proportion can coexist with impaired filling. A stiff or thickened ventricle may accept less blood or require higher pressure to fill. It can still eject a normal percentage of the smaller filled volume. Symptoms such as breathlessness and fluid retention may occur when filling pressures rise, especially with exertion.

Heart failure with preserved ejection fraction therefore requires more than symptoms and an LVEF at or above a cutoff. Objective evidence of cardiac structural or functional abnormality and elevated filling pressures or natriuretic peptides may be needed, while alternative causes of symptoms must be considered.

Valve disease can also complicate the number. With significant mitral regurgitation, part of the ejected blood travels backward into the left atrium, so total stroke volume can look maintained while forward output is impaired. In severe aortic stenosis, afterload can alter performance. LVEF may remain apparently preserved until disease is advanced.

Regional wall-motion abnormalities from ischemia or scar may be clinically important even when the global average remains in a preserved range. Strain imaging, volumes, valve measures, and symptom testing may reveal information that one percentage misses.

Reduced LVEF is a finding, not the cause#

A reduced value means the left ventricle is ejecting a smaller fraction of its filled volume under the conditions of that study. It does not identify why. Coronary disease, prior myocardial injury, and myocarditis can produce similar numbers. So can genetic cardiomyopathy, toxins, and tachyarrhythmia. So can valve disease, severe pressure load, endocrine conditions, and other processes.

The pattern can help. A regional abnormality suggests a different differential diagnosis from global dysfunction. Chamber dilation, wall thickness, and right-ventricular function may be relevant. So may valve findings, electrocardiography, and biomarkers. So may symptoms, family history, and coronary or tissue assessment.

LVEF also does not map directly to symptom severity. Some people with markedly reduced values have modest symptoms, while others with preserved values are very limited. Congestion, rhythm, and skeletal muscle influence function. So do lungs, kidneys, and anemia. So do deconditioning and comorbid conditions.

Loading conditions can change the number#

The ventricle contracts against afterload, broadly the resistance and pressure it must overcome. A sudden increase in blood pressure can reduce the fraction ejected even without a new loss of intrinsic muscle function. Lower afterload can increase it.

Preload, the filling condition before contraction, also matters. Diuresis, dehydration, fluid administration, ventilation, and valve lesions can alter volumes and LVEF. Heart rate and rhythm affect beat-to-beat filling. With atrial fibrillation, measurements may average several representative cycles.

These dependencies do not make LVEF meaningless. They explain why timing and context belong in interpretation. A value taken during shock, severe hypertension, rapid arrhythmia, acute valve failure, or immediately after an intervention may not match your stable outpatient measurement.

What counts as reduced or preserved is changing#

The 2022 AHA/ACC/HFSA clinical practice guideline used specific categories that remain important for treatment evidence: HFrEF at LVEF 40% or less, HFmrEF at 41% to 49%, HFpEF at 50% or greater, and improved LVEF when a prior value of 40% or less later rises by at least 10 percentage points to above 40%.

The Second Universal Definition of Heart Failure, released in 2026, moves away from one rigid set of global LVEF cutoffs. Its consensus framework uses reduced, preserved, and improved categories while recognizing that reference limits can vary. They can vary with sex, age, ancestry, modality, and laboratory standards.

Those documents answer related but different questions. The 2026 document is a consensus definition, not a replacement treatment guideline. Trials, drug labels, device indications, coverage rules, and the 2022 guideline may still specify numeric LVEF criteria. Your report should state which framework it is using rather than silently translating one into another.

Measurement variation and meaningful change#

LVEF is usually reported as a number, but the true measurement has uncertainty. Reader tracing, image plane, and border visibility all contribute. So do equipment, software, rhythm, and loading conditions. Differences of several percentage points can occur without a biological change.

A change is more credible when it is larger than expected method variation, obtained with comparable technique, accompanied by consistent volume or strain changes, and fits the clinical course. A shift from 39% to 42% can cross a category boundary without establishing recovery. Conversely, a modest numerical change may matter if it is reproducible and connected to a treatment or device criterion.

Reports should include left-ventricular volumes indexed to body size when available, method, study quality, comparison date, and important technical limitations. The original images may deserve review when a high-stakes decision for you rests on a borderline value.

Improvement is not the same as cure#

LVEF can improve after treatment of ischemia, inflammation, or arrhythmia. It can improve after treatment of valve disease, pressure load, or cardiomyopathy. Improvement is encouraging and often carries prognostic information. It may represent reverse remodeling and more effective pump performance.

Yet relapse can occur, and the underlying susceptibility may persist. The 2022 guideline specifically recognizes heart failure with improved LVEF rather than relabeling every person as never having had heart failure. Evidence and guideline recommendations about continuing therapy depend on the condition and clinical context. Symptoms, natriuretic peptides, exercise capacity, congestion, arrhythmia risk, valve disease, and ventricular structure can all remain abnormal after LVEF rises, so follow-up should not be reduced to chasing a single percentage.

Reading an imaging report well#

When you have the report in hand, first identify which ventricle is described. Right-ventricular ejection fraction is distinct and has different geometry and reference values. Next identify modality, quantitative method, rhythm, and image quality. Look at end-diastolic and end-systolic volumes, not only the ratio.

Then read global and regional wall motion, chamber sizes, and wall thickness. Read valve findings, right-heart measures, estimated pressures, and filling information. Compare with prior studies using the same modality where possible. Note whether the clinical state was similar.

Finally, connect the imaging finding to the question you started with. Diagnosing heart failure, evaluating a murmur, assessing cardiotoxic treatment, investigating chest pain, and deciding about a device require different supporting evidence. LVEF is one input within that pathway, not a standalone verdict.

Sources#

1. 2022 AHA/ACC/HFSA guideline for the management of heart failure
2. ASE and EACVI recommendations for cardiac chamber quantification
3. American Heart Association explanation of ejection fraction
4. Society for Cardiovascular Magnetic Resonance post-processing standards
5. Second Universal Definition of Heart Failure, 2026
6. American College of Cardiology summary of the 2026 definition

Imaging findings should be interpreted with the full clinical picture.*