MUGA and LVEF: Count-Based Ejection Fraction

Introduction

The gated blood pool scan — MUGA, or equilibrium radionuclide angiography (ERNA) — measures left ventricular ejection fraction (LVEF) from counts rather than geometry, and that single physical difference is why it remains the most reproducible LVEF tool for catching small serial changes in patients on cardiotoxic chemotherapy. Because the counts detected from technetium-99m-labeled blood are proportional to the volume of blood in the chamber, MUGA computes ejection fraction with no assumption about the shape of the ventricle — the property that makes it operator-independent and highly repeatable. ^{1, 2}

In an era of ever-faster echocardiography and cardiac MRI, it is fair to ask why a decades-old planar nuclear medicine test still holds a place in cardiology. The answer is precision. When an oncologist needs to know whether a patient's LVEF has truly dropped by a few points on trastuzumab or an anthracycline — a decision that can pause life-saving cancer therapy — the reproducibility of the measurement matters as much as its accuracy. MUGA's count-based physics delivers exactly that reproducibility. ^{2, 3}

This guide explains the physics that makes MUGA count-based, how the LVEF formula handles background, how the study is acquired and labeled, what dose it delivers, and where MUGA still outperforms echocardiography. DRPS supports these programs through its PET/CT and nuclear medicine physics services across Florida, Maryland, Virginia, Washington DC, California, and beyond.

Topic Explanation

What MUGA measures, and why "counts, not geometry" matters

MUGA measures the volume of blood in the left ventricle indirectly, by counting the gamma photons emitted from technetium-99m bound to the patient's own red blood cells. The camera records the heart through many hundreds of cardiac cycles, sorting the counts into a series of frames (bins) spanning one representative average cycle synchronized to the ECG R-wave. The result is a cine loop of the beating blood pool and, crucially, a time–activity curve for the left ventricle. ¹

The foundational principle is that, within a fixed acquisition geometry, detected counts are proportional to the amount of labeled blood in the region of interest. ² End-diastole is the frame with the most counts (largest blood volume); end-systole is the frame with the fewest. Every other LVEF method — 2D and 3D echocardiography using area-length or disk-summation models, or contrast ventriculography — must assume a geometric model of ventricular shape to convert an image into a volume. MUGA does not. It reads volume changes directly from count changes, which removes a whole class of geometric error and observer subjectivity. ²

For the QC foundations that keep the underlying camera and dose trustworthy, see our companion guides on gamma camera performance testing and dose calibrator quality control.

Labeling the blood pool

The tracer is not injected as a free agent; the patient's own erythrocytes are labeled with Tc-99m in one of three ways, in ascending order of labeling efficiency: in vivo (stannous pyrophosphate injected first, then pertechnetate), modified in vivo (a hybrid drawn into a shielded syringe), and in vitro (cells labeled in a kit outside the body and reinjected). Labeling efficiency directly affects image quality: unbound pertechnetate accumulates in the stomach, thyroid, and background, degrading the target-to-background ratio the LVEF calculation depends on. ⁹ In vitro labeling gives the highest, most consistent efficiency and is preferred when image quality is paramount, such as serial cardiotoxicity monitoring. ^{1, 9}

Key Technical Principles

The count-based LVEF formula

The left ventricular ejection fraction is the fraction of end-diastolic blood volume ejected each beat. In count terms, with a background region of interest subtracted from both the end-diastolic and end-systolic counts: ¹

LVEF = \frac{( C _{E D} - C _{bk g} ) - ( C _{E S} - C _{bk g} )}{C _{E D} - C _{bk g}} \times 100%

The background term cancels in the numerator, giving the familiar teaching form:

LVEF = \frac{C _{E D} - C _{E S}}{C _{E D} - C _{bk g}} \times 100%

where $C_{E D}$ and $C_{E S}$ are the counts in the left-ventricular ROI at end-diastole and end-systole, and $C_{bk g}$ is the counts in a background ROI. The background is critical: photons from blood in overlying tissue and the chamber's surroundings add counts that are not from ejected volume. The background ROI is classically placed as a crescent just outside the left ventricle at end-systole, in a region of minimal cavity activity, and its per-pixel activity is scaled and subtracted before the ejection fraction is computed. ¹ An incorrectly placed or sized background region is one of the most common sources of LVEF error in practice.

Why the counts-proportional-to-volume assumption holds

The linearity between counts and volume is what makes MUGA geometry-free, but it depends on conditions the physicist and technologist must protect:

Stable labeling. Poor red-cell labeling shifts counts into background organs and biases the ratio.
Consistent geometry. The best-septal left anterior oblique (LAO) view — typically near 45° LAO, adjusted patient-by-patient — separates the left ventricle from the right ventricle and left atrium so the LV counts are not contaminated by adjacent chambers. ¹
Adequate count statistics. A planar study is typically acquired to roughly 6 million counts, over a bounded acquisition time, so that the count differences between frames are statistically meaningful. ¹
Good gating. The ECG R-wave triggers each cycle, and a beat-length acceptance (rejection) window — commonly on the order of ±15% around the mean R–R interval — discards ectopic or irregular beats that would blur the volume curve. ¹

Temporal sampling: how many frames?

The number of frames per cardiac cycle sets the temporal resolution of the volume curve. Planar MUGA typically uses 16 to 24 frames per cycle. Sixteen frames are adequate for a robust systolic LVEF, but 24 or more frames sharpen the sampling of the true end-systolic minimum and materially improve diastolic indices such as peak filling rate and time-to-peak filling — parameters that can detect trastuzumab-related dysfunction before systolic LVEF falls. ⁴ Higher heart rates compress the cycle, so more frames help most when the patient is tachycardic. ^{1, 4}

The same gated cine also supports Fourier (phase and amplitude) analysis: a first-harmonic fit to each pixel's time–activity curve yields phase images that map the timing of regional contraction, quantifying mechanical dyssynchrony. ⁴

The acquisition at a glance

Parameter	Typical planar MUGA value	Notes
Radiopharmaceutical	Tc-99m-labeled autologous red blood cells	In vitro labeling preferred for image quality ⁹
Administered activity	~740–1110 MBq (20–30 mCi)	Justify and optimize under ALARA ¹
Energy window	140 keV ± 10%	Tc-99m photopeak ¹
View	Best-septal LAO (~45°)	Separates LV from RV/LA ¹
Frames per cycle	16–24	≥24 for diastolic indices ^{1, 4}
Beat acceptance	ECG R-wave gating, ~±15% window	Rejects ectopy/arrhythmia ¹
Target counts	~6 million (planar)	Adequate count statistics ¹
Effective dose	~5–8 mSv	Method/coefficient dependent ^{12, 13}

Clinical Impact

The reproducibility argument

The clinical value of MUGA is dominated by one property: reproducibility. Because it is count-based and operator-independent, MUGA can detect small serial changes in LVEF that noisier methods miss. Reported figures are compelling — semi-automated ERNA in experienced hands shows intraobserver reproducibility on the order of a few LVEF percentage points, with interobserver variability wider but still tight relative to alternatives. ³ Classic test–retest work found rest ejection-fraction correlations around 0.95 between repeat studies. ^{10, 11} For a serial-monitoring test, that consistency is the whole point.

Cardio-oncology and the cardiotoxicity threshold

MUGA's home turf is monitoring cancer-therapy-related cardiac dysfunction. Anthracyclines and HER2-targeted agents such as trastuzumab can depress LVEF, and management decisions hinge on whether a real decline has occurred. The commonly used definition of cardiotoxicity is a fall in LVEF below 50%, or an absolute drop of at least 10 percentage points from the patient's baseline. ⁴ Detecting a 10-point change reliably requires a method whose measurement noise is well under 10 points — precisely where MUGA's reproducibility earns its place. Diastolic dysfunction detectable on well-framed MUGA may even precede the systolic drop in some patients on trastuzumab-based therapy. ⁴

MUGA versus the alternatives

MUGA does not exist in a vacuum, and understanding its relationship to other LVEF methods prevents dangerous apples-to-oranges comparisons across serial studies.

Method	LVEF basis	Geometric assumption	Relationship to MUGA
MUGA / planar ERNA	Counts ∝ volume	None	Reference for reproducibility ^{2, 3}
Gated SPECT (blood pool / MPI)	Counts, 3D	Minimal	Good correlation at rest (r ≈ 0.82); may underestimate post-stress LVEF ^{6, 7}
CZT-SPECT ERNA	Counts, 3D	None	LVEF correlates strongly with planar (r ≈ 0.93–0.99); lower dose ⁷
Echocardiography	Geometric (area-length / disks)	Yes	Wide limits of agreement with MUGA; not interchangeable ⁵
Cardiac MRI	Volumetric	None	Volume reference standard; ERNA RVEF agrees reasonably (r ≈ 0.81) ⁸

The most consequential point in that table is the mismatch with echocardiography. A cardio-oncology study directly comparing 3D echo and MUGA reported limits of agreement on the order of 30 percentage points between the two, and the two modalities flagged cardiotoxicity in very different fractions of patients. ⁵ The lesson is not that one is right and one is wrong, but that a program must follow a given patient with one consistent method and not switch methods mid-treatment.

Practical Optimization Tips

A dependable MUGA program comes down to protecting the count-to-volume linearity and the reproducibility that follows from it.

1. Get the labeling right

Use in vitro labeling when image quality and serial consistency matter most. Verify labeling efficiency indirectly by watching for free-pertechnetate signs (stomach, thyroid, and high background), and troubleshoot the stannous/pertechnetate timing if the target-to-background ratio is poor. ⁹

2. Standardize the view and background ROI

Acquire a true best-septal LAO with the ventricles cleanly separated, and standardize how the background ROI is drawn. Because background subtraction directly scales the denominator and numerator, an inconsistent background region injects variability that defeats the purpose of a reproducible test. ¹

3. Frame for the question

Use at least 16 frames for LVEF; move to 24 or more when diastolic function or dyssynchrony is being assessed, or when the heart rate is high. Do not under-frame a study whose purpose is subtle diastolic change. ⁴

4. Protect the gating

Set a sensible beat-acceptance window and monitor the rejected-beat fraction. A patient in atrial fibrillation or with frequent ectopy may need list-mode or arrhythmia-tolerant acquisition; a poorly gated study blurs end-systole and biases LVEF. ¹

5. Keep the method consistent across serial studies

For any patient followed over time, lock the radiopharmaceutical, activity, view, framing, and processing so that a change in reported LVEF reflects the heart, not the protocol. Reproducibility is a property of a disciplined workflow, not just of the physics.

Common pitfalls to avoid

Sloppy or inconsistent background ROIs. The single most common LVEF error. ¹
Poor red-cell labeling. Free pertechnetate wrecks the target-to-background ratio. ⁹
Under-framing a diastolic study. Sixteen frames can miss the information the exam was ordered to find. ⁴
Switching modalities mid-treatment. MUGA and echo are not interchangeable; comparing across them can manufacture a false "change." ⁵
Ignoring arrhythmia. Bad beats accepted into the average distort end-systole. ¹

Regulatory Considerations

MUGA is a medical use of byproduct material, so it is governed by NRC (or Agreement State) regulations for radioactive material, and its protocol should conform to the current consensus procedure standard. The physics program supports both the regulatory and the quality frameworks. ¹

Key references:

SNMMI/EANM Procedure Standard for Gated Equilibrium Radionuclide Angiography (2020) — the current consensus document for acquiring, processing, interpreting, and reporting ERNA, superseding the older SNM Version 3.0 guideline. ¹
10 CFR Part 35 — Medical Use of Byproduct Material — governs authorized users, written directives where applicable, dose calibrator and instrument requirements, and radiation safety for the Tc-99m used in the study.
ICRP Publication 128 — the reference compilation of radiopharmaceutical dose coefficients used to estimate the effective dose from a Tc-99m red-blood-cell study, which lands on the order of 5–8 mSv for a planar acquisition. ^{12, 13}

Tc-99m for MUGA is byproduct material regulated by the NRC in non-Agreement jurisdictions and by the state in Agreement States. Among the states DRPS serves, Florida, Maryland, Virginia, California, Nevada, New York, Pennsylvania, and New Jersey are NRC Agreement States, while Washington, DC and Delaware are regulated directly by the NRC for radioactive material. Activity should always be justified and optimized under ALARA, and the generator, dose calibrator, and camera must be under a documented QC program. For the upstream isotope and instrument QC that a MUGA program relies on, see our guides on Tc-99m generator quality control and cardiac SPECT MPI quality control.

Frequently Asked Questions (FAQs)

What is a MUGA scan?

A MUGA (multigated acquisition) scan, also called equilibrium radionuclide angiography (ERNA) or gated blood pool imaging, is a nuclear medicine test that measures cardiac function — most importantly left ventricular ejection fraction (LVEF). The patient's red blood cells are labeled with technetium-99m, and a gamma camera records the changing counts in the heart over many cardiac cycles synchronized to the ECG.

How does MUGA measure ejection fraction?

MUGA measures LVEF from counts, not geometry. Because the detected counts from labeled blood are proportional to the blood volume in the chamber, ejection fraction is computed as end-diastolic counts minus end-systolic counts, divided by end-diastolic counts, after subtracting a background region. No assumption about the shape of the ventricle is required, which is the key theoretical advantage over geometric methods.

Why is MUGA considered so reproducible?

MUGA is highly reproducible because it is count-based and largely operator-independent: it does not depend on tracing endocardial borders or assuming a ventricular geometry, and averaging over hundreds of heartbeats reduces noise. Studies report low interobserver variability and excellent test-retest correlation, which is why MUGA is favored for detecting small serial LVEF changes.

When is MUGA preferred over echocardiography?

MUGA is often preferred when small, serial changes in LVEF must be detected reliably, such as monitoring for chemotherapy-induced cardiotoxicity. MUGA and echocardiography are not interchangeable — studies report wide limits of agreement between the two — so a program should generally follow a patient with one consistent method rather than switching between them.

How much radiation does a MUGA scan involve?

A planar Tc-99m red-blood-cell MUGA study delivers an effective dose on the order of 5 to 8 millisieverts, depending on administered activity and the dose coefficient used. Newer CZT-based SPECT gated blood pool techniques with individualized dosing can reduce this substantially. As with any nuclear medicine study, activity should be justified and optimized under the ALARA principle.

What is a normal LVEF, and what defines cardiotoxicity?

Normal resting LVEF is generally about 50% or higher. In cardio-oncology, cancer-therapy-related cardiac dysfunction is commonly defined as a fall in LVEF below 50% or an absolute drop of at least 10 percentage points from the patient's baseline. Because MUGA is reproducible, it is well suited to flagging these threshold crossings.

How many frames per cardiac cycle does MUGA use?

Planar MUGA typically uses 16 to 24 frames per cardiac cycle. Sixteen frames are adequate for a reliable LVEF, but 24 or more frames improve the temporal sampling of the volume curve and are preferred when diastolic function parameters such as peak filling rate are needed, especially at higher heart rates.

Key Takeaways

MUGA measures LVEF from counts, not geometry. Detected counts are proportional to blood volume, so ejection fraction needs no assumption about ventricular shape.
Background subtraction is central. The LVEF formula subtracts a background ROI from the ventricular counts; a poorly placed background region is a leading source of error.
Reproducibility is the whole point. Being count-based and operator-independent, MUGA reliably detects small serial LVEF changes — ideal for cardiotoxicity monitoring.
Cardiotoxicity threshold: LVEF < 50% or a ≥ 10-point drop from baseline. Detecting that reliably demands a low-noise method.
Frame for the question. 16 frames suffice for systolic LVEF; use ≥ 24 for diastolic indices and dyssynchrony.
Do not mix methods across serial studies. MUGA and echocardiography have wide limits of agreement and are not interchangeable.

Conclusion

MUGA endures not because it is new or fast, but because it is trustworthy in exactly the situation that matters most: telling whether a patient's heart function has really changed. Its count-based physics sidesteps the geometric assumptions and observer subjectivity that add noise to other LVEF methods, and that noise reduction is what lets a clinician act on a modest serial decline with confidence.

For the medical physicist and technologist, the job is to protect the chain that makes those counts meaningful — good red-cell labeling, a clean best-septal view, a consistent background ROI, adequate framing, and disciplined gating. Do that, and MUGA delivers the reproducibility that keeps a cancer patient's cardiac monitoring honest.

How DRPS Can Help

Diagnostic Radiation Physics Services supports nuclear cardiology programs with gamma camera acceptance testing and annual physics surveys, dose calibrator and instrument QC, protocol review for gated blood pool and cardiac SPECT studies, ALARA and radiation safety program support, and staff training. For a MUGA program specifically, DRPS can help standardize acquisition and processing so that serial LVEF measurements stay reproducible and defensible.

DRPS provides PET/CT and nuclear medicine physics and medical physics consulting across our service locations, including Florida, Maryland, Virginia, Washington DC, California, Nevada, New York, Pennsylvania, New Jersey, and Delaware.

A reproducible MUGA program is a clinical safeguard: it is what lets an oncologist trust that a 10-point LVEF drop is the patient, not the measurement.

Related Resources

References

Farrell MB, Galt JR, Georgoulias P, Malhotra S, Pagnanelli R, Rischpler C, Savir-Baruch B. SNMMI Procedure Standard/EANM Guideline for Gated Equilibrium Radionuclide Angiography. J Nucl Med Technol. 2020;48(2):126-135. doi:10.2967/jnmt.120.246405. PubMed
Sachpekidis C, Sachpekidis V, Moralidis E, Arsos G. Equilibrium radionuclide ventriculography: still a clinically useful method for the assessment of cardiac function? Hell J Nucl Med. 2018;21(3):213-220. PubMed
Sachpekidis C, Sachpekidis V, Kopp-Schneider A, Arsos G, Moralidis E. Equilibrium radionuclide angiography: intra- and inter-observer repeatability and reproducibility in the assessment of cardiac systolic and diastolic function. J Nucl Cardiol. 2021;28(4):1304-1314. doi:10.1007/s12350-019-01830-9. PubMed
Klein R, Nadouri D, Osler E, Johnson C, Dent S, Dwivedi G. Diastolic dysfunction can precede systolic dysfunction on MUGA in cancer patients receiving trastuzumab-based therapy. Nucl Med Commun. 2019;40(1):22-29. doi:10.1097/MNM.0000000000000941. PubMed
Nolan MT, Pathan F, Nott L, Black A, Pointon O, Marwick TH. Comparison of echocardiography and multi-planar gated acquisition scans for predicting cancer-treatment-related cardiovascular dysfunction. Heart Lung Circ. 2024;33(5):693-703. doi:10.1016/j.hlc.2024.03.010. PubMed
Acampa W, Caprio MG, Nicolai E, et al. Assessment of poststress left ventricular ejection fraction by gated SPECT: comparison with equilibrium radionuclide angiocardiography. Eur J Nucl Med Mol Imaging. 2010;37(2):349-356. doi:10.1007/s00259-009-1308-5. PubMed
Massardo T, Jaimovich R, Lavados H, et al. Comparison of radionuclide ventriculography using SPECT and planar techniques in different cardiac conditions. Eur J Nucl Med Mol Imaging. 2007;34(11):1735-1746. doi:10.1007/s00259-007-0472-8. PubMed
Apert A, Canu M, Jankowski A, et al. Comparison of cadmium zinc telluride ECG-gated SPECT equilibrium radionuclide angiocardiography to magnetic resonance imaging to measure right ventricular volumes and ejection fraction. J Nucl Cardiol. 2022;29(4):1647-1656. doi:10.1007/s12350-021-02653-3. PubMed
Ballinger JR, Gerson B, Gulenchyn KY, Ruddy TD, Davies RA. Technetium-99m red blood cell labeling in patients treated with doxorubicin. Clin Nucl Med. 1988;13(3):169-170. doi:10.1097/00003072-198803000-00005. PubMed
Pfisterer ME, Battler A, Swanson SM, Slutsky R, Froelicher V, Ashburn WL. Reproducibility of ejection-fraction determinations by equilibrium radionuclide angiography. J Nucl Med. 1979;20(6):491-495. PubMed
Slutsky R, Karliner J, Battler A, Pfisterer M, Swanson S, Ashburn W. Reproducibility of ejection fraction and ventricular volume by gated radionuclide angiography after myocardial infarction. Radiology. 1979;132(1):155-159. doi:10.1148/132.1.155. PubMed
Delhaye N, Louvard Y, Cussac-Buchdahl C, et al. Radiation dose to patients undergoing myocardial perfusion imaging and radionuclide angiography. AJR Am J Roentgenol. 2006;187(6):1615-1620. doi:10.2214/AJR.05.1193. PubMed
Mettler FA Jr, Huda W, Yoshizumi TT, Mahesh M. Effective doses in radiology and diagnostic nuclear medicine: a catalog. Radiology. 2008;248(1):254-263. doi:10.1148/radiol.2481071451. PubMed