Y-90 Bremsstrahlung SPECT/CT Imaging & Dosimetry

Introduction

Yttrium-90 barely emits gamma rays, so the only way to image where the dose actually went is to detect the faint bremsstrahlung X-rays created as its beta particles slow down in tissue. Post-therapy Y-90 bremsstrahlung SPECT/CT confirms that radioembolization microspheres landed in the target, screens for unintended lung or extrahepatic deposition, and — with careful quantitative correction — supports the absorbed-dose calculations that connect a treatment to its expected outcome.

Transarterial radioembolization (TARE), also called selective internal radiation therapy (SIRT), delivers millions of Y-90-loaded glass or resin microspheres into the hepatic arterial supply of liver tumors. Because the microspheres are permanently trapped, more than 95% of the administered activity stays in the liver, and the therapeutic dose is delivered locally by short-range beta particles. That same physics — a nearly pure beta emitter — is what makes Y-90 so difficult to image and quantify.

This guide explains the decay physics that makes bremsstrahlung imaging both possible and challenging, how to choose the energy window and collimator, how scatter, attenuation, and resolution-recovery corrections turn a blurry qualitative scan into a quantitative dose map, and how bremsstrahlung SPECT/CT compares with Y-90 PET/CT. DRPS supports this work through its PET/CT and nuclear medicine physics and medical physicist consulting services across Florida, Maryland, Virginia, Washington DC, California, and Nevada.

Topic Explanation

Why a beta emitter produces an image

Yttrium-90 decays to stable zirconium-90 by beta-minus emission with essentially no primary gamma rays, so its detectable photon signal comes almost entirely from bremsstrahlung — the continuous X-ray spectrum produced when energetic electrons decelerate in matter. ¹ The beta particle is energetic (a maximum energy of about 2.28 MeV and a mean of about 0.93 MeV), and its short range in tissue is what makes Y-90 an effective local therapy. As each beta slows down, a small fraction of its energy is radiated as bremsstrahlung photons spanning a broad energy range, and it is those photons that a gamma camera records. ¹

There is a second, far rarer pathway. A tiny internal pair-production branch — measured at (31.86 ± 0.47) × 10⁻⁶, roughly 32 per million decays — produces positron–electron pairs and therefore 511 keV annihilation photons. ² That vanishingly small branch is the entire physical basis for Y-90 PET, which is more quantitative than bremsstrahlung SPECT but requires a PET/CT scanner and long acquisitions to collect enough of those rare events.

The core imaging challenge

Conventional nuclear medicine imaging exploits a discrete photopeak — the 140 keV line of Tc-99m, for example — so an energy window can accept true photons and reject scatter. Bremsstrahlung has no photopeak. Its spectrum is continuous, and the high-energy tail penetrates collimator septa and scatters extensively, producing a diffuse background that blurs the image and reduces contrast. Monte Carlo analysis has shown that on a conventional gamma camera, even with a reduced-energy acquisition window, the true geometric (unscattered) X-rays can represent less than 15% of the recorded events — which is why dedicated detector and collimator designs have been explored for this task. ¹⁴ Every energy-window choice is therefore a compromise: a wide window collects more counts but admits more scatter and penetration, while a narrow window improves contrast at the cost of statistics. This is why Y-90 bremsstrahlung SPECT was historically treated as a qualitative "did it go where we aimed?" check rather than a quantitative dosimetry tool. For the broader quantitative-SPECT framework this builds on, see our quantitative SPECT/CT calibration guide.

Key Technical Principles

Y-90 decay data at a glance

Property	Value	Relevance to imaging and dosimetry
Decay mode	β⁻ to stable Zr-90	Nearly pure beta emitter; almost no primary gamma ¹
Physical half-life	~64.05 h (~2.67 d)	Sets cumulated-activity and dose calculations ¹
Beta maximum energy	~2.28 MeV	High energy drives bremsstrahlung production ¹
Beta mean energy	~0.9337 MeV	Used in the absorbed-dose constant ¹
Internal pair-production branch	(31.86 ± 0.47) × 10⁻⁶	Basis for Y-90 PET (511 keV) ²
Detectable SPECT signal	Continuous bremsstrahlung X-rays	No photopeak → window is a compromise

Choosing the acquisition: window and collimator

For quantitative bremsstrahlung SPECT, the acquisition is optimized to suppress the penetration-and-scatter background rather than to maximize raw counts. A medium-energy general-purpose (MEGP) collimator is a common choice, balancing sensitivity against septal penetration; high-energy collimators further reduce penetration at the cost of sensitivity. One validated quantitative protocol uses a relatively narrow 90–125 keV primary window together with a higher-energy background-compensation window to estimate and subtract the penetration/scatter contribution. ⁸ Broad windows (for example 50–250 keV) collect more events but degrade quantification, so the "best" window is method- and camera-specific and must be validated on the local system rather than copied blindly. ⁸

From qualitative scan to quantitative dose map

The corrections that make bremsstrahlung SPECT quantitative are the same ones formalized for internal-emitter dosimetry in MIRD Pamphlet No. 23 — CT-based attenuation correction, scatter correction, and collimator–detector resolution recovery — adapted for the bremsstrahlung spectrum, often with Monte Carlo modeling of the energy-dependent response. ⁶ The payoff is large. With appropriate attenuation and background compensation, one study improved the recovery coefficient of a 37 mm sphere from 39% to 90% and reduced spurious cold-lung counts from 44% to 14%; a commercial quantitative-SPECT workflow has been shown to agree with Monte Carlo voxel dosimetry to within about 3%. ^{8, 9} Because essentially all of the administered activity is permanently trapped in the liver, the total intrahepatic Y-90 activity is known to better than 95%, which provides a strong physical constraint for self-calibration of the reconstructed images. ⁹

Worked example: the Y-90 absorbed-dose constant

For a nearly pure beta emitter whose short-range particles are absorbed essentially where they are deposited, the mean absorbed dose to a uniformly loaded compartment follows a simple constant. Start from the cumulated activity (total number of decays) for complete decay of an initial activity $A_{0}$ :

\tilde{A} = A_{0} \cdot τ = A_{0} \cdot \frac{T _{1/2}}{ln 2}

With $T_{1/2} = 64.05 h = 230, 580 s$ , the mean life is $τ = 230, 580/0.693 \approx 3.327 \times 1 0^{5} s$ . For $A_{0} = 1 GBq = 1 \times 1 0^{9} Bq$ , the total number of decays is:

\tilde{A} = 1 \times 1 0^{9} \times 3.327 \times 1 0^{5} \approx 3.327 \times 1 0^{14} decays

Each decay deposits, on average, the mean beta energy $\overset{ˉ}{E}_{β} = 0.9337 MeV = 1.496 \times 1 0^{- 13} J$ . The total energy released per GBq is therefore:

E = 3.327 \times 1 0^{14} \times 1.496 \times 1 0^{- 13} \approx 49.8 J per GBq

Dividing by mass gives the well-known Y-90 dose constant:

D (Gy) \approx 49.7 \times \frac{A ( GBq )}{M ( kg )}

often quoted as 49.67 Gy·kg·GBq⁻¹ (the familiar "about 50 Gy per GBq per kg"). ^{1, 3} As a worked case, delivering 2.0 GBq of Y-90 to a perfused liver lobe of 1.5 kg gives a mean absorbed dose of $49.67 \times 2.0/1.5 \approx 66 Gy$ for a single-compartment model — the starting point that partition- and voxel-based dosimetry then refine using the imaged activity distribution.

Worked example: lung shunt and the lung-dose limit

Some injected particles bypass the tumor capillary bed and lodge in the lungs. The lung shunt fraction (LSF) is estimated before treatment from a Tc-99m macroaggregated albumin (MAA) scan:

LSF = \frac{C _{lung}}{C _{lung} + C _{liver}}

where $C$ are geometric-mean counts over lung and liver regions. Applying the same dose constant with an assumed lung mass of about 1 kg, the projected lung absorbed dose is:

D_{lung} \approx 49.67 \times \frac{A ( GBq ) \times LSF}{M _{lung} ( kg )}

Treatment planning keeps the lung dose within accepted limits — on the order of 30 Gy per treatment and 50 Gy cumulative — to avoid radiation pneumonitis. ^{3, 4} After treatment, bremsstrahlung SPECT/CT provides a qualitative check on whether significant lung deposition actually occurred.

Clinical Impact

Bremsstrahlung SPECT versus Y-90 PET

Both modalities image the same Y-90, but through different physics, and they trade off accuracy against availability.

Feature	Y-90 bremsstrahlung SPECT/CT	Y-90 PET/CT
Signal imaged	Continuous bremsstrahlung X-rays	511 keV photons from rare pair production ²
Availability	Any modern SPECT/CT	Requires PET/CT and long acquisitions
Quantitative accuracy	Lower; tends to underestimate tumor dose	Higher; better absolute quantification
Reported tumor-dose bias	Underestimates by roughly 50% versus PET in one comparison	Reference standard in that comparison ¹¹
Best role	Distribution confirmation, lung/extrahepatic check, dosimetry where PET is unavailable	Quantitative post-therapy voxel dosimetry

A direct comparison of image-based dosimetry methods found that bremsstrahlung SPECT underestimated tumor absorbed dose by about 50% (−50 ± 13%) relative to Y-90 PET/CT, while a scaled scheme that combined the pre-therapy Tc-99m distribution with the Y-90 SPECT counts largely removed that bias. ¹¹ The practical message is that bremsstrahlung SPECT/CT is an excellent, universally available tool for confirming distribution and screening for extrahepatic deposition, and it can support dosimetry when quantitative corrections are applied and validated — but where absolute tumor-dose accuracy is paramount, Y-90 PET/CT is the stronger quantitative modality. For the pre-therapy planning side of this workflow, see our Y-90 radioembolization dosimetry guide.

Dose–response and why quantification matters

The reason to push bremsstrahlung SPECT toward quantitative accuracy is that liver radioembolization shows a genuine dose–response relationship. For hepatocellular carcinoma treated with glass microspheres, mean tumor absorbed doses around 160 Gy on Y-90 SPECT/CT-based voxel dosimetry have been associated with a higher probability of mRECIST response, and other series report glass-microsphere response thresholds of roughly 205 Gy and resin-microsphere thresholds near 100–120 Gy. ^{10, 12} On the safety side, the same voxel-dosimetry work reported no complications when the normal-liver mean dose stayed below about 44 Gy. ¹⁰ Complementary Y-90 PET/CT dose–response modeling has similarly linked higher tumor absorbed dose to a greater probability of tumor control, reinforcing that the relationship is real across imaging modalities. ¹³ These thresholds vary with microsphere type, dosimetry method, and imaging modality, so they are best treated as population-level guides interpreted alongside the specific method used — not fixed cutoffs. The link between the imaged activity map and these absorbed-dose numbers is exactly the MIRD schema for internal dosimetry applied at the voxel level.

Practical Optimization Tips

Getting usable quantitative bremsstrahlung SPECT

Use a medium- or high-energy collimator. A low-energy collimator will be overwhelmed by septal penetration from the high-energy bremsstrahlung tail. ⁸
Optimize and validate the energy window locally. Adopt a validated narrow primary window with background compensation rather than a wide catch-all window, and confirm its performance on your own camera with a phantom. ⁸
Always acquire SPECT/CT, not planar. CT provides the attenuation map for correction and the anatomic reference for identifying lung and extrahepatic activity.
Apply the full correction chain. Attenuation, scatter/penetration background compensation, and resolution recovery are what move the study from qualitative to quantitative. ^{6, 8}
Exploit the >95% trapping constraint. Knowing that nearly all activity is retained in the liver enables self-calibration and a physical cross-check on the reconstructed activity. ⁹
Verify camera QC first. Quantitative accuracy assumes the underlying SPECT system meets its performance and uniformity specifications; see SPECT/CT quality control and the NEMA NU 1-2023 performance framework. ⁷

Common pitfalls to avoid

Treating a bremsstrahlung window like a photopeak. There is no photopeak; the window is always a scatter-versus-counts compromise.
Reading absolute tumor doses off an uncorrected scan. Without the correction chain and validation, bremsstrahlung SPECT can substantially misstate the dose. ¹¹
Ignoring partial-volume effects. Small tumors are under-recovered; recovery coefficients from a phantom of similar sizes should inform interpretation. ⁸
Copying another site's protocol without validation. The optimal window and correction settings are camera- and software-specific. ⁸
Confusing pre-therapy MAA distribution with the actual Y-90 distribution. The MAA surrogate can differ from where the microspheres finally land, which is precisely why post-therapy imaging exists.

Regulatory Considerations

Y-90 radioembolization is the medical use of a byproduct material and sits under both NRC/Agreement State radioactive-material regulation and the professional-society standards that define how the therapy and its imaging are performed and documented. The physicist's post-therapy imaging and dosimetry work should align with the current guidance rather than legacy practice.

Key frameworks:

AAPM Medical Physics Practice Guideline 14.a (2023–2024) — the AAPM practice guideline for Y-90 microsphere radioembolization, covering the medical physicist's role in treatment, imaging, and dosimetry. ³
ACR multi-society Practice Parameter for Radioembolization of Liver Malignancies (revised 2025) — the joint ACR/ABS/ACNM/ARS/SIR/SNMMI parameter defining the clinical standard of care. ⁴
EANM Dosimetry Committee SOP for radioembolization (2021) — a unified methodology for Tc-99m MAA pre-therapy and Y-90 peri-therapy dosimetry, noting that quantitative bremsstrahlung SPECT is usable when dedicated correction methods are available. ⁵
MIRD Pamphlet No. 23 — the quantitative-SPECT methodology underlying voxel dosimetry. ⁶
NEMA NU 1-2023 — the current gamma-camera/SPECT performance standard, including tomographic contrast and absolute quantification accuracy. ⁷

Administration of Y-90 microspheres requires an authorized user and a written directive under 10 CFR Part 35, and the program must meet 10 CFR Part 20 radiation-protection requirements. Across the states DRPS serves, Florida, Maryland, Virginia, California, Nevada, Pennsylvania, New York, and New Jersey are NRC Agreement States that license medical use under their own radiation-control rules, while Washington, DC and Delaware are regulated directly by the NRC for radioactive material. Facilities should confirm which authority issues their license and what post-administration survey, dosimetry, and record-keeping expectations apply.

Frequently Asked Questions (FAQs)

Why does Y-90 produce an image at all if it is a beta emitter?

Yttrium-90 emits almost no gamma rays, but as its high-energy beta particles decelerate in tissue they produce a continuous spectrum of bremsstrahlung X-rays. A gamma camera can detect those X-rays, which is why post-therapy Y-90 imaging is called bremsstrahlung SPECT. There is also an extremely rare internal pair-production branch, about 32 per million decays, that produces the 511 keV photons used for Y-90 PET.

Why is Y-90 bremsstrahlung imaging harder than conventional SPECT?

Bremsstrahlung is a broad, continuous energy spectrum with no photopeak, so any energy window is a compromise between counts and scatter. The high-energy component causes collimator septal penetration and scatter that blur the image and degrade contrast. Achieving quantitative results requires careful window selection, a medium- or high-energy collimator, and scatter, attenuation, and resolution-recovery corrections.

What energy window and collimator should be used for quantitative Y-90 bremsstrahlung SPECT?

A medium-energy general-purpose collimator with a relatively narrow acquisition window is commonly used for quantitative work; one validated protocol uses a 90 to 125 keV primary window with a background-compensation window. Broader windows such as 50 to 250 keV collect more counts but include more scatter and penetration and degrade quantification. The optimal choice is method- and camera-specific and should be validated on the local system.

Is Y-90 PET more accurate than bremsstrahlung SPECT?

Yes, for absolute quantification. Y-90 PET/CT images the rare internal pair-production positrons and generally gives more accurate absorbed-dose estimates, while bremsstrahlung SPECT tends to underestimate tumor absorbed dose relative to PET. Bremsstrahlung SPECT/CT remains valuable because it is widely available on any modern SPECT/CT and confirms microsphere distribution and lung or extrahepatic deposition.

What is bremsstrahlung SPECT used for after radioembolization?

Post-treatment bremsstrahlung SPECT/CT confirms that the Y-90 microspheres deposited in the targeted tumor and liver territory rather than in the lungs, bowel, or other extrahepatic sites, and it supports post-therapy voxel or partition dosimetry to relate the delivered tumor and normal-liver absorbed doses to the expected clinical response.

What tumor absorbed dose is associated with response in liver radioembolization?

Reported thresholds vary by microsphere type, dosimetry method, and imaging modality. For hepatocellular carcinoma treated with glass microspheres, mean tumor absorbed doses on the order of 160 Gy or higher on Y-90 SPECT/CT-based voxel dosimetry have been associated with a higher probability of response, while reported resin-microsphere thresholds are often lower. These values are population averages, not individual guarantees, and should be interpreted with the specific dosimetry method used.

How is the lung shunt fraction related to Y-90 imaging?

Before treatment, a Tc-99m macroaggregated albumin scan estimates the lung shunt fraction, the proportion of injected particles that pass through the liver to the lungs, so the projected lung absorbed dose can be kept within accepted limits of about 30 Gy per treatment and 50 Gy cumulative. After treatment, bremsstrahlung SPECT/CT can qualitatively confirm whether significant lung deposition occurred.

Key Takeaways

Y-90 imaging exists because of bremsstrahlung. A nearly pure beta emitter has no photopeak, so the detectable signal is the continuous X-ray spectrum from decelerating betas. ¹
The window is always a compromise. With no photopeak, a narrow validated window plus background compensation and a medium/high-energy collimator gives the best quantitative result. ⁸
Corrections make it quantitative. Attenuation, scatter/penetration, and resolution recovery — the MIRD 23 framework — turn a qualitative scan into a dose map, improving sphere recovery dramatically. ^{6, 8}
PET is more accurate; SPECT is more available. Bremsstrahlung SPECT tends to underestimate tumor dose versus Y-90 PET but is universally available for distribution confirmation and dosimetry. ¹¹
Dose–response is real. Reported glass-microsphere response thresholds near 160–205 Gy and a normal-liver tolerance around 44 Gy make accurate post-therapy dosimetry clinically meaningful. ^{10, 12}
Align with current guidance. AAPM MPPG 14.a, the 2025 ACR multi-society parameter, and the EANM SOP define how this imaging and dosimetry should be performed. ^{3, 4, 5}

Conclusion

Y-90 bremsstrahlung SPECT/CT is a case study in extracting a real, clinically useful measurement from an inconvenient physical signal. The isotope that makes radioembolization therapeutically elegant — a short-range, nearly pure beta emitter — is the same one that makes imaging hard, because it offers only a diffuse, photopeak-free bremsstrahlung glow to work with. Modern SPECT/CT, a carefully chosen window and collimator, and a validated correction chain nonetheless turn that glow into a distribution map good enough to confirm treatment delivery and support post-therapy dosimetry.

The medical physicist's role is to know where the accuracy comes from and where its limits are: to optimize and validate the acquisition, apply and check the corrections, exploit the strong physical constraint that nearly all activity is trapped in the liver, and be honest about when Y-90 PET is the better quantitative tool. Done well, post-therapy imaging closes the loop between what was prescribed and what was delivered — which is the whole point of dosimetry.

How DRPS Can Help

Diagnostic Radiation Physics Services supports nuclear medicine and radioembolization programs with quantitative SPECT/CT and PET/CT setup and validation, energy-window and reconstruction optimization for Y-90 bremsstrahlung imaging, post-therapy dosimetry review, camera performance testing to NEMA standards, and radiation-safety and authorized-user program support through our PET/CT and nuclear medicine physics and medical physicist consulting services.

DRPS supports facilities across our service locations, including Florida, Maryland, Virginia, Washington DC, California, Nevada, New York, Pennsylvania, New Jersey, and Delaware.

Good post-therapy imaging is not an afterthought — it is the verification step that makes a dosimetry-driven radioembolization program defensible.

Related Resources

References

International Commission on Radiological Protection. ICRP Publication 107: Nuclear Decay Data for Dosimetric Calculations. Annals of the ICRP. 2008;38(3). icrp.org
Selwyn RG, Nickles RJ, Thomadsen BR, DeWerd LA, Micka JA. A new internal pair production branching ratio of 90Y: The development of a non-destructive assay for 90Y and 90Sr. Applied Radiation and Isotopes. 2007;65(3):318-327. doi:10.1016/j.apradiso.2006.08.009. PubMed
Busse NC, Al-Ghazi MSAL, Abi-Jaoudeh N, et al. AAPM Medical Physics Practice Guideline 14.a: Yttrium-90 microsphere radioembolization. Journal of Applied Clinical Medical Physics. 2024;25(2):e14157. doi:10.1002/acm2.14157. PubMed
Johnson T, Spieler BO, Toskich BB, et al. ACR-ABS-ACNM-ARS-SIR-SNMMI Practice Parameter for Radioembolization of Liver Malignancies. American Journal of Clinical Oncology. 2025;49(1):10-24. doi:10.1097/COC.0000000000001234. PubMed
Chiesa C, Sjogreen-Gleisner K, Walrand S, et al. EANM dosimetry committee series on standard operational procedures: a unified methodology for 99mTc-MAA pre- and 90Y peri-therapy dosimetry in liver radioembolization with 90Y microspheres. EJNMMI Physics. 2021;8(1):77. doi:10.1186/s40658-021-00394-3. PubMed
Dewaraja YK, Frey EC, Sgouros G, et al. MIRD Pamphlet No. 23: Quantitative SPECT for patient-specific 3-dimensional dosimetry in internal radionuclide therapy. Journal of Nuclear Medicine. 2012;53(8):1310-1325. doi:10.2967/jnumed.111.100123. PubMed
National Electrical Manufacturers Association. NEMA NU 1-2023: Performance Measurements of Gamma Cameras. 2023. nema.org
Siman W, Mikell JK, Kappadath SC. Practical reconstruction protocol for quantitative 90Y bremsstrahlung SPECT/CT. Medical Physics. 2016;43(9):5093. doi:10.1118/1.4960629. PubMed
Balagopal A, Kappadath SC. Characterization of 90Y-SPECT/CT self-calibration approaches on the quantification of voxel-level absorbed doses following 90Y-microsphere selective internal radiation therapy. Medical Physics. 2018;45(2):875-883. doi:10.1002/mp.12695. PubMed
Kappadath SC, Mikell J, Balagopal A, et al. Hepatocellular Carcinoma Tumor Dose Response After 90Y-Radioembolization With Glass Microspheres Using 90Y-SPECT/CT-Based Voxel Dosimetry. International Journal of Radiation Oncology, Biology, Physics. 2018;102(2):451-461. doi:10.1016/j.ijrobp.2018.05.062. PubMed
Brosch J, Gosewisch A, Kaiser L, et al. 3D image-based dosimetry for Yttrium-90 radioembolization of hepatocellular carcinoma: Impact of imaging method on absorbed dose estimates. Physica Medica. 2020;80:317-326. doi:10.1016/j.ejmp.2020.11.016. PubMed
Garin E, Rolland Y, Edeline J. 90Y-Loaded Microsphere SIRT of HCC Patients With Portal Vein Thrombosis: High Clinical Impact of 99mTc-MAA SPECT/CT-Based Dosimetry. Seminars in Nuclear Medicine. 2019;49(3):218-226. doi:10.1053/j.semnuclmed.2019.01.006. PubMed
Dewaraja YK, Devasia T, Kaza RK, et al. Prediction of Tumor Control in 90Y Radioembolization by Logit Models with PET/CT-Based Dose Metrics. Journal of Nuclear Medicine. 2020;61(1):104-111. doi:10.2967/jnumed.119.226472. PubMed
Walrand S, Hesse M, Wojcik R, Lhommel R, Jamar F. Optimal design of Anger camera for bremsstrahlung imaging: Monte Carlo evaluation. Frontiers in Oncology. 2014;4:149. doi:10.3389/fonc.2014.00149. PubMed