Diagnostic Reference Levels in Nuclear Medicine

A diagnostic reference level in nuclear medicine is a benchmark administered activity — usually the 75th percentile of a survey of the activities facilities actually use — that flags procedures using unusually high activity for review. It is not a dose limit and it does not apply to an individual patient. It is one of the simplest, most defensible tools a nuclear medicine program has for confirming that its administered activities are reasonable and for finding the procedures worth optimizing.

Diagnostic reference levels (DRLs) began in diagnostic radiology, but the concept transfers naturally to nuclear medicine, where the quantity being optimized is the administered activity of the radiopharmaceutical rather than the output of an x-ray tube. A facility that knows how its administered activities compare to national benchmarks can tell, at a glance, whether its protocols sit in the mainstream or whether a procedure is an outlier worth a second look. This article explains what DRLs mean in nuclear medicine, how they are derived, how they differ from dose limits, and how a facility can use them. DRPS provides this analysis as part of its PET/CT and nuclear medicine physics services across Florida, Maryland, Virginia, Washington DC, California, and Nevada.

Introduction

Optimization is the forgotten pillar of radiation protection. Justification asks whether a procedure should be done; dose limitation protects workers and the public; but optimization — keeping doses as low as reasonably achievable while still answering the clinical question — is where most of the day-to-day improvement in patient dose actually happens. In nuclear medicine, optimization has a convenient handle: the administered activity, a single number recorded for every study. DRLs turn that number into a feedback loop.

The International Commission on Radiological Protection formalized the modern DRL framework, and the U.S. National Council on Radiation Protection and Measurements translated it into the American context with the companion idea of the achievable dose.^{1, 2} National and regional surveys across Japan and Korea have since published nuclear medicine DRLs derived from facility data, and in the United States the American College of Radiology has derived DRLs and achievable administered activities from its accreditation database.^{3, 4, 5, 6} Together they give a facility a concrete way to benchmark itself.

For the companion diagnostic-imaging concept, see our article on diagnostic reference levels in radiology, and for the single-procedure version of this optimization story, F-18 FDG PET dose optimization.

Topic Explanation

What a DRL is — and what it is not

A diagnostic reference level is a benchmark value of a dose-related quantity for a standard procedure performed on a typical patient. In nuclear medicine, that quantity is the administered activity (in MBq) for a named study — for example, a bone scan with Tc-99m MDP, or an oncologic FDG PET scan. The DRL is set by surveying many facilities, taking each facility's median administered activity for the procedure, and then taking the 75th percentile of those facility medians as the DRL.¹

The single most important thing to understand about a DRL is what it is not. It is not a dose limit. Dose limits are regulatory ceilings that protect radiation workers and members of the public and must not be exceeded. A DRL applies to a typical patient undergoing a typical protocol, not to any individual, and exceeding it is not a violation — it is a prompt to investigate.¹ A large or clinically complex patient may justifiably receive more than the DRL. The DRL earns its keep when a facility's typical practice sits above the benchmark, because that usually means activity is being used that could be reduced without harming the diagnosis.

DRL versus achievable administered activity

NCRP Report No. 172 added a useful second benchmark to the DRL: the achievable dose, or in nuclear medicine the achievable administered activity.² Where the DRL is the 75th percentile and flags the high outliers, the achievable administered activity is the median — the 50th percentile — of the survey distribution. The DRL says "if you are above here, investigate"; the achievable value says "half of your peers already operate at or below here, so this is a realistic target." The two together give a facility both a ceiling and a goal.

Local versus national DRLs

National DRLs describe the practice of a whole country or region. A local DRL is a facility's own benchmark, derived from its own data, and is often tighter than the national value. The optimization cycle is: survey your own administered activities, compare to national DRLs, investigate procedures above them, and then set local DRLs that lock in the improvement. National surveys from Japan, Korea, and elsewhere were built exactly this way, by collecting facility-median activities and reporting the 75th percentile.^{3, 4, 5}

Key Technical Principles

The 75th-percentile definition

Formally, if facility $i$ reports a median administered activity ${\bar{A}}_i$ for a given procedure, and $n$ facilities are surveyed, the national DRL for that procedure is the 75th percentile of the set of facility medians:

and the achievable administered activity is the median of the same set, $P_{50}$.^{1, 2} The two-stage structure — median within each facility, then percentile across facilities — is deliberate: it prevents one busy or unusual facility from dominating the benchmark and it describes the distribution of ordinary practice rather than the extremes.

From administered activity to effective dose

A DRL is expressed in administered activity because that is what a facility controls and records. To translate it into patient dose, the administered activity $A$ is multiplied by a radiopharmaceutical-specific effective dose coefficient $e$ (in mSv/MBq) tabulated by the ICRP:^{7, 8}

For adult F-18 FDG, the ICRP-tabulated effective dose coefficient is approximately $e=0.019mSv/MBq$.⁷ A representative adult oncology FDG administration — surveys place the 75th-percentile activity near 235 to 252 MBq³ — therefore corresponds to:

from the radiopharmaceutical alone. In a PET/CT or SPECT/CT study the CT component adds to this, which is why modern DRLs track the CT dose index and dose-length product separately from the administered activity.³

Representative national administered-activity DRLs

The table below lists 75th-percentile administered activities for common adult procedures from a large national survey, to illustrate the magnitude and spread of nuclear medicine DRLs. These are survey-derived benchmarks, not prescriptions; your own national or local values should govern practice.

The survey that produced these values also found that more than 90% of facilities fell within 50% of the median for these procedures, and that administered activities clustered near package-insert and guideline recommendations — evidence that a DRL program drives convergence toward a sensible central value.³

Clinical Impact

Finding the procedures worth optimizing

The practical power of a DRL program is triage. A facility cannot review every protocol at once, but it can rank them: the procedures where its typical administered activity most exceeds the national DRL are the ones with the most headroom for optimization. In the United States, the ACR derived DRLs and achievable administered activities for adult myocardial perfusion imaging directly from accreditation data, giving facilities a domestic benchmark to compare against and, in the authors' assessment, identifying opportunities to further reduce patient exposure with modern cameras and protocols.⁶

Pediatric administered activities

Children are the population where administered-activity optimization matters most, because of their longer life expectancy and higher radiosensitivity. Pediatric nuclear medicine scales activity to body weight (MBq/kg) with a minimum activity floor to preserve image quality in the smallest patients. For years, the North American consensus guidelines and the European EANM pediatric dosage card recommended different activities for the same study, producing avoidable variation in children's doses.^{9, 10} A joint working group harmonized the two systems so they yield similar weight-based activities, and comparisons of the resulting effective doses showed where each system had been higher.^{11, 12} A pediatric whole-body FDG PET/CT survey illustrated the effect: mean administered activities by ICRP age band of roughly 80, 119, and 176 MBq were materially lower than the 99, 166, and 286 MBq of the older North American guideline and the 120, 189, and 302 MBq of the earlier EANM card.¹³ The direction of travel is consistently downward, toward the lowest activity that still answers the clinical question.

The hybrid-imaging CT component

A DRL program that only tracks administered activity misses half the dose in a modern hybrid study. In PET/CT and SPECT/CT, the CT can contribute a large share of the total effective dose, and national DRL surveys now report CT dose indices and dose-length products for the hybrid CT alongside the administered activity.³ Optimizing a hybrid protocol means asking two questions: is the administered activity reasonable, and is the CT dose matched to its purpose — whether attenuation correction and localization only, or a diagnostic-quality CT. For the imaging-biomarker side of this, see our article on PET SUV quantification.

Practical Tips

• Record and audit administered activity by procedure. The dose calibrator measurement that precedes every administration is your raw DRL data. Capturing it systematically turns routine QC into an optimization dataset. See dose calibrator quality control.
• Use the facility median, then compare to the national DRL. Take your median administered activity for standard-size adults for each procedure and compare each median to the national DRL. The procedures above it are your optimization priorities.
• Aim for the achievable value, not just under the DRL. Staying below the 75th percentile is the floor; the median (achievable) value is the better target where image quality allows.
• Standardize before you minimize. Convergence of administered activity toward a sensible central value is itself an improvement; wide intra-facility variation usually signals protocols that are not being followed consistently.
• Weight pediatric activity to body weight. Adopt a harmonized weight-based pediatric scheme with a minimum activity, rather than scaling an adult activity by a rule of thumb.
• Optimize the CT in hybrid studies. Match the CT dose to its role; an attenuation-correction CT does not need diagnostic-CT output.
• Set local DRLs and revisit them. DRLs are not static; equipment and reconstruction improvements lower what is achievable, so re-survey periodically.

Regulatory Considerations

Diagnostic reference levels are an optimization tool recommended by international and national radiation-protection bodies, not a regulatory dose limit, but they intersect with the rules that govern medical use of radioactive material. In the United States, the medical use of byproduct material is regulated by the Nuclear Regulatory Commission under 10 CFR Part 20 and Part 35, or by an Agreement State under equivalent rules. Of the states DRPS serves, Florida, Maryland, Virginia, California, and Nevada are NRC Agreement States that license medical use under their own radiation-control programs, while Washington, DC is regulated directly by the NRC.

The DRL framework itself comes from guidance rather than regulation:

• ICRP Publication 135 is the current international statement on diagnostic reference levels in medical imaging, including how DRLs apply to administered activity in nuclear medicine and the preference for weight-based values where appropriate.¹
• NCRP Report No. 172 provides the U.S. framework and adds the achievable dose (the median) as an optimization target alongside the DRL.²
• ICRP Publications 128 and 103 supply the radiopharmaceutical effective dose coefficients and the effective-dose methodology used to translate administered activity into patient dose.^{7, 8}
• IAEA GSR Part 3 incorporates DRLs into the international Basic Safety Standards and applies them to diagnostic nuclear medicine.¹⁴

Accreditation programs reinforce the practice: facility participation in accreditation requires documented, optimized protocols, and the ACR's own accreditation data have been used to derive U.S. nuclear medicine DRLs.⁶ A DRL program is best documented as part of the facility's optimization and quality program, reviewed by a qualified medical physicist. DRPS supports this through PET/CT and nuclear medicine physics, medical physicist consulting, and accreditation support.

Frequently Asked Questions (FAQs)

What is a diagnostic reference level in nuclear medicine?

A diagnostic reference level (DRL) in nuclear medicine is a benchmark value of administered activity for a given procedure, typically set at the 75th percentile of a survey of the activities facilities actually use for a standard-size adult patient. If a facility's typical administered activity is above the DRL, it is a signal to investigate whether the activity can be reduced without losing diagnostic quality.

Is a DRL a dose limit?

No. A DRL is not a regulatory dose limit and it does not apply to an individual patient. Dose limits protect workers and the public; DRLs are optimization tools that apply to typical patients and typical protocols. A single patient may justifiably receive more activity than the DRL for a valid clinical reason, and that is not a violation.

How is a DRL different from an achievable administered activity?

A DRL is usually the 75th percentile of the survey distribution and is meant to flag the higher-activity outliers. An achievable administered activity, as described by NCRP, is the median — the 50th percentile — and represents a more aspirational optimization target that about half of facilities already meet. Using both gives a facility a ceiling to stay under and a goal to aim for.

How do I set up a DRL survey at my facility?

Collect the administered activity your facility uses for standard-size adults for each common procedure, take the facility median for each, and compare those medians against the national DRLs. Procedures where your typical activity exceeds the national DRL are the ones to review first. The process mirrors how national surveys derive DRLs from facility-median data.

How are pediatric administered activities handled?

Pediatric nuclear medicine scales administered activity to body weight, usually in MBq per kilogram, with a minimum activity to preserve image quality in the smallest patients. The North American consensus guidelines and the EANM pediatric dosage card were harmonized so that the two systems give similar weight-based activities, which reduces unnecessary variation in children's doses.

How does administered activity relate to effective dose?

Effective dose is estimated by multiplying the administered activity by a radiopharmaceutical-specific effective dose coefficient (in mSv per MBq) tabulated by the ICRP. For F-18 FDG the adult coefficient is about 0.019 mSv/MBq, so a 250 MBq administration corresponds to roughly 4.8 mSv from the radiopharmaceutical, before any CT component in a PET/CT or SPECT/CT study.

Do DRLs include the CT dose in SPECT/CT and PET/CT?

Modern DRL frameworks address both components separately: the administered activity for the radiopharmaceutical and the CT dose index and dose-length product for the CT portion. A hybrid study should be optimized on both axes, because the CT can contribute a substantial share of the total effective dose.

Key Takeaways

• A DRL is a benchmark, not a limit. It is typically the 75th percentile of surveyed facility-median administered activities, used to flag procedures for review — not a ceiling that must never be crossed for an individual patient.
• Pair the DRL with the achievable value. The median (achievable administered activity) is the optimization target; the 75th percentile is the investigation trigger.
• Benchmark, then optimize. Compare your facility-median activities to national DRLs, and start with the procedures where you most exceed them.
• Pediatric activity is weight-based and converging. Harmonized North American and EANM schemes scale activity to body weight with a minimum, and have trended downward.
• Hybrid studies need two-axis optimization. Track and optimize both the administered activity and the CT dose in PET/CT and SPECT/CT.
• Document it as a physicist-reviewed program. A DRL survey is most useful when it is recorded, trended, and revisited as technology improves.

Conclusion

Diagnostic reference levels give nuclear medicine a rare thing: an optimization tool that is simple, quantitative, and built from data a facility already collects. The administered activity is recorded for every patient; a DRL turns that routine number into a benchmark that reveals which protocols sit in the mainstream and which deserve a closer look. Used correctly — as an investigation trigger rather than a limit, paired with the achievable value as a target, and applied with judgment to individual patients — a DRL program steadily pulls a facility's doses toward the lowest activity that still answers the clinical question.

The work is not glamorous, but it is exactly what good optimization looks like: measure, compare, investigate the outliers, and lock in the gains with local reference levels. A facility that runs this loop can show, with data, that its nuclear medicine doses are reasonable — which is the whole point.

How DRPS Can Help

Diagnostic Radiation Physics Services helps nuclear medicine and PET facilities build and document administered-activity optimization programs. This includes surveying your administered activities, comparing them against national DRLs and achievable values, identifying the procedures with the most optimization headroom, setting local DRLs, optimizing the CT component of hybrid studies, and documenting the program for accreditation and inspection. Our work spans PET/CT and nuclear medicine physics, medical physicist consulting, and accreditation support.

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

Optimization is a habit, not a one-time project. A DRL program is how a nuclear medicine facility makes that habit visible and defensible.

Related Resources

• Diagnostic reference levels in radiology
• F-18 FDG PET dose optimization
• Pediatric nuclear medicine dosing
• PET SUV quantification
• Radiopharmaceutical dosimetry and ICRP 128
• Dose calibrator quality control
• PET/CT and nuclear medicine physics
• Accreditation support

References

1. International Commission on Radiological Protection. Diagnostic Reference Levels in Medical Imaging (ICRP Publication 135). Ann ICRP. 2017;46(1). icrp.org
2. National Council on Radiation Protection and Measurements. Reference Levels and Achievable Doses in Medical and Dental Imaging: Recommendations for the United States (NCRP Report No. 172). Bethesda, MD: NCRP; 2012. ncrponline.org
3. Watanabe H, Ishii K, Hosono M, et al. Report of a nationwide survey on actual administered radioactivities of radiopharmaceuticals for diagnostic reference levels in Japan. Ann Nucl Med. 2016;30(6):435-444. doi:10.1007/s12149-016-1079-6. PubMed
4. Abe K, Hosono M, Igarashi T, et al. The 2020 national diagnostic reference levels for nuclear medicine in Japan. Ann Nucl Med. 2020;34(11):799-806. doi:10.1007/s12149-020-01512-4. PubMed
5. Song HC, Na MH, Kim J, et al. Diagnostic reference levels for adult nuclear medicine imaging established from the national survey in Korea. Nucl Med Mol Imaging. 2019;53(1):64-70. doi:10.1007/s13139-019-00585-y. PubMed
6. Becker MD, Butler PF, Bhargavan-Chatfield M, et al. Adult gamma camera myocardial perfusion imaging: diagnostic reference levels and achievable administered activities derived from ACR accreditation data. J Am Coll Radiol. 2016;13(6):688-695. doi:10.1016/j.jacr.2016.02.004. PubMed
7. International Commission on Radiological Protection. Radiation Dose to Patients from Radiopharmaceuticals: A Compendium of Current Information Related to Frequently Used Substances (ICRP Publication 128). Ann ICRP. 2015;44(2S). icrp.org
8. International Commission on Radiological Protection. The 2007 Recommendations of the International Commission on Radiological Protection (ICRP Publication 103). Ann ICRP. 2007;37(2-4). icrp.org
9. Gelfand MJ, Parisi MT, Treves ST. Pediatric radiopharmaceutical administered doses: 2010 North American consensus guidelines. J Nucl Med. 2011;52(2):318-322. doi:10.2967/jnumed.110.084327. PubMed
10. Treves ST, Gelfand MJ, Fahey FH, Parisi MT. 2016 update of the North American consensus guidelines for pediatric administered radiopharmaceutical activities. J Nucl Med. 2016;57(12):15N-18N. PubMed
11. Lassmann M, Treves ST. Paediatric radiopharmaceutical administration: harmonization of the 2007 EANM paediatric dosage card (version 1.5.2008) and the 2010 North American consensus guidelines. Eur J Nucl Med Mol Imaging. 2014;41(5):1036-1041. doi:10.1007/s00259-014-2731-9. PubMed
12. Grant FD, Gelfand MJ, Drubach LA, Treves ST, Fahey FH. Radiation doses for pediatric nuclear medicine studies: comparing the North American consensus guidelines and the pediatric dosage card of the European Association of Nuclear Medicine. Pediatr Radiol. 2015;45(5):706-713. doi:10.1007/s00247-014-3211-x. PubMed
13. Alkhybari EM, McEntee MF, Willowson KP, Brennan PC, Kitsos T, Kench PL. An Australian local diagnostic reference level for paediatric whole-body F-18 FDG PET/CT. Br J Radiol. 2019;92(1096):20180879. doi:10.1259/bjr.20180879. PubMed
14. International Atomic Energy Agency. Radiation Protection and Safety of Radiation Sources: International Basic Safety Standards (GSR Part 3). Vienna: IAEA; 2014. iaea.org