Ron Lattanze, MBA, provided his commentary regarding why H.R.2541 or the Nuclear Medicine Clarification Act of 2025 should be passed.
Ron Lattanze, MBA, provided his commentary regarding why H.R.2541 or the Nuclear Medicine Clarification Act of 2025 should be passed.

In response to a recent CancerNetwork® feature, titled H.R. 2541: Examining Its Legislative Impact on Radiopharmaceutical Practice. Ron Lattanze, MBA, reached out to provide commentary on this piece. The feature which focuses on the Nuclear Medicine Clarification Act of 2025 provided 2 perspectives regarding the pros and cons of this potential legislative bill.
Lattanze, chief executive officer (CEO) of Lucerno Dynamics, focuses on the how this bill could bring more awareness to the nuclear medicine aspect of oncology. Additionally, he further sheds light on this area of care, providing insights and perspective that may not have been previously discussed.
I was extremely pleased to see recent CancerNetwork coverage on the Nuclear Medicine Clarification Act. I feel that Roman Fabbricatore did a respectable job with the article. However, I would like to comment on or clarify several points made by Fabbricatore Jackson W. Kiser, MD, Mary Ajango, and Thomas Hope, MD, who were quoted in the article. I am extremely familiar with this topic since I petitioned the Nuclear Regulatory Commission (NRC) to remove their incorrect reporting exemption regarding extravasations over 5 years ago.1 For full disclosure, I am the CEO of Lucerno Dynamics. Lucerno has developed technology that can help clinicians identify, determine the severity of, and reduce the frequency of extravasations. In the case of administrations of high-dose therapeutics, our technology can help clinicians detect and stop an extravasation as it happens. Please note that our technology is NOT the only way a nuclear medicine center can identify or reduce extravasations. These goals can be accomplished in so many other ways, and we encourage nuclear medicine centers to use whatever method works for them.
Fabbricatore’s article suggests that the Nuclear Medicine Clarification Act (H.R. 2541) changes how radiopharmaceutical extravasations are reported. That is not entirely correct, since extravasations are currently not being reported at all.H.R. 2541 would simply have the NRC treat these accidental exposures of radiation to a patient no differently than the NRC treats other accidental exposures. As pointed out in Fabbricatore’s article, Ajango is perfectly correct in her analysis. The long-standing NRC exemption for reporting ANY extravasations makes zero sense. As Ajango said, if a radiopharmaceutical is accidentally spilled ON a patient and resulting radiation dose to the patient’s skin or underlying tissue exceeds the reporting threshold (0.5 Sv) then it needs to be reported to the NRC.2 If the same radiopharmaceutical is accidentally spilled INTO the patient’s tissue, instead of the patient’s vein as intended and prescribed, and the extravasation resulted in a dose 100 or 1000 times the 0.5 Sv reporting threshold, the NRC does not require that accident to be reported. This exemption does not make sense, and large extravasations need to be reported.
Fabbricatore suggested extravasations do not require reporting because the NRC cited reporting would result in a “significant licensee and regulatory burden with no additional benefit to patient safety.”That is a recent NRC response related to my petition but is not the reason for the reporting exemption. The reason extravasations are not reported is because the NRC mistakenly created an internal policy that exempted reporting of extravasations.3 The1980 exemption was based on the incorrect premise that extravasations were an accident that could not be avoided and therefore there was no reason to make unavoidable accidents reportable. We know that premise is not correct. Extravasations, like most accidental exposures, are almost entirely preventable.
In his article, Fabbricatore also introduced a clinical comment and Hope mistakenly intimated patient harm with the NRC regulation. Patient harm is not relevant to the NRC reporting requirement. The purpose of NRC’s medical event reporting is to identify potential mishandlings of isotopes.4 If an accident exposes a patient to an absorbed dose beyond the reporting threshold – a threshold that the NRC revised upward in 2001 and that was endorsed by the medical community – then this event requires reporting.5 The goal of the regulation is to identify the root cause of the accident so that these learnings can be shared and future patients will not experience the same issue. While it is true that some medical events may have resulted in clear patient harm, a mishandling of a medical isotope does not necessarily require patient harm to be reportable.6 To be clear, extravasations that result in large accidental radiation exposures qualify as medical isotope mishandling. No isotope that is supposed to be administered intravenously is ever intended to end up in the tissue at the injection site.
Ajango also stated that patients want greater transparency. Pam Kohl, an extravasated patient, agrees in her article, “Transparency – a patient-centric view of radiopharmaceutical extravasations”7. She asks 23 questions at the end of her article that combined with the fact that large extravasations are not required to be reported, confirm without a doubt that there is little transparency regarding how radiopharmaceutical extravasations are handled today. What is clear is that there are over 100 peer-reviewed clinical articles that clearly show how a large extravasation can negatively affect patient care.8-10 While only a few are referenced, I am happy to discuss the multitude that support this claim upon further request. Additionally, there is scientific evidence that shows how large doses of radiation are not good for healthy tissue.11,12 Therefore, it follows that a patient who is on the receiving end of a large extravasation would want to know. But no one is looking for, tracking or reporting extravasations today, so large extravasations are often hidden. Based on how extravasations affect care and healthy tissue, transparency is warranted.
One of the reasons extravasations are hidden is because Hope is correct; radiopharmaceuticals are typically low volume injections. What that means for the patient and the clinician performing the administration, is that the extravasation is often very hard to visually detect. It does not cause a noticeable swelling at the injection site. It is disingenuous to suggest that just because a radiopharmaceutical administration involves a small volume of fluid means that an extravasation is not important. In addition to negatively affecting the images used to guide patient care, an extravasation can negatively affect the patient’s healthy tissue. While a small volume may not cause the harmful condition known as compartment syndrome, a volume of 1 mL to 1.5 mL of a radiopharmaceutical can contain a high level of radioactivity – for example, 10 mCi of 18F-FDG may be administered in a volume of 1.5 mL and result in 1.1 trillion disintegrations in the first 60 minutes. This can result in a very high absorbed dose (9.1 Gy) to 5 cc of healthy tissue.13 If this radiation is concentrated in a small area of the body and not distributed systemically as prescribed, then it can compromise the procedure and expose healthy tissue to a high radiation dose.14-16
Extravasations are also difficult to detect because the patient does not typically feel pain or stinging. Most radiopharmaceuticals are not vesicants like chemotherapy pharmaceuticals, so the patient is unlikely to sense that something has gone wrong and thus cannot provide feedback. Furthermore, extravasations are difficult to identify after the administration because most nuclear medicine imaging does not include the injection site. Kiser a global expert on extravasations, also featured in Fabbricatore’s article, clearly summarizes the difficulty in detecting extravasations in one of his many publications on this topic. He states, “due to the nature of radiopharmaceuticals and methods of administration, most radiopharmaceutical extravasations are unnoticed by clinicians and patients; and that even when visible, they are rarely included in the radiology report.”17
Hope reported that he has not seen any adverse effects from any radiopharmaceutical extravasation. I am sure he is correct, but for several reasons this personal experience regarding clinical outcomes is not a valid reason for community or academic centers to skip reporting large extravasations if and when they do happen.
The first reason centers should not skip reporting is because extravasations do happen. In Fabbricatore’s article Hope pointed out that nuclear medicine extravasations are not uncommon. Hope is correct again; the clinical evidence is clear they do happen. Frequently. When centers imaged the injection site, or used technology they have developed, or used our technology to proactively assess the injection site, the results were clear. On average, across all the studies involving thousands of patients from centers that have chosen to proactively monitor the injection site, the extravasation rate is over 17%.18-28 Shouldn’t patients know if their administrations resulted in an extravasation? If these extravasations were large, shouldn’t they be measured and documented in the patient’s record? The Society of Nuclear Medicine and Molecular Imaging (SNMMI) agreed. In their letter to the NRC, they state, “In the event an extravasation or partial extravasation is suspected, imaging of the injection site should be performed using a gamma camera.”29
Another reason centers should not skip reporting is because the effects of ionizing radiation can take weeks, months, or years to manifest.30 Shouldn’t large extravasations be documented so these patients (especially younger patients) could be followed?
Hope questioned the purpose of the legislation. The purpose is clear. After several patients who received nuclear medicine died in the late 1970s in a Columbus, OH, hospital, Congress asked the NRC to create a system to minimize the chance that future patients would be harmed by a medical isotope. In 1980 and through 2001, NRC created a simple process that makes sense: if a patient receives an accidental radiation dose over a defined threshold that clearly represents a potential mishandling of an isotope (0.5 Sv), as the result of a lack of training, quality control, or human error, this event should be reported. The goal is to share lessons learned and minimize these same accidents in the future. Yet, the NRC, at the request of the medical community in 1980 and today, continues to avoid making clinicians report extravasations. Despite accepting my petition in 2022 that the exemption is incorrect, NRC continues to drag their feet to ensure reporting.31 Nearly 3 years later, large extravasations that exceed the existing threshold are still not reportable. In the summer of 2025, the NRC Commissioners advanced a proposed rule that would require reporting of only those extravasations deemed likely to injure the patient as determined by the clinicians responsible for the extravasation.32 The proposed rule is simply another way to exempt extravasations from reporting by using subjective criterion to allow the clinicians who are against reporting to decide whether to report or not. This proposed rule would create the only medical event with subjective reporting criteria. Congress, by passing the Nuclear Medicine Clarification Act, would do what the NRC has failed to do—protect patients.
What would be the impact of H.R. 2541? For centers that do not extravasate, there is nothing to report and it should not be burdensome. Kiser, with nearly 20,000 administrations monitored, claims there is no burden and the SNMMI in the same letter to the NRC stated:
“Good nuclear medicine practices already include imaging injection sites in certain cases using gamma cameras.”29
For centers that routinely extravasate and cause high accidental radiation doses to healthy tissue, they should report. Until these centers improve their practices, patients have the right to know which centers routinely make mistakes in the administration of radiopharmaceuticals. This is a basic patient right.
Methods and technology exist today that can identify extravasations that SNMMI says can compromise the quality and quantification of procedures and that also result in high levels of radiation to healthy tissue.33 It is unfathomable that in the year 2025, we would simply accept them with no effective monitoring. Interestingly, clinicians and physicists spend a substantial amount of time and money to ensure the highest quality procedure and image. There are daily efforts to ensure cameras are calibrated properly to produce the best images. They work diligently to ensure the radiopharmaceutical handling is flawless so that the precise amount of radiation prescribed is delivered to the patient, including measuring minute amounts of residual radiopharmaceutical remaining stuck to the syringe plastic after the injection. Why then are they reluctant to monitor what happens during the actual radiopharmaceutical injection, the 1 step that has the most chance to negatively affect the quality and quantification of the procedure? Some therapeutic radiopharmaceuticals can cost patients and payers $50,000 per administration. Some diagnostic radiopharmaceuticals are used to help determine or guide neurological, cardiology, and oncology treatments that cost hundreds of thousands of dollars. It is hard to believe that the medical and payer communities are not pushing for effective monitoring themselves.
Hope suggests peptide receptor and PSMA-targeted therapy may not lead to complications; I believe that is misleading. If extravasated, the beta energy emissions from peptide receptor and PSMA therapies only travel 1 mm to 2 mm in the tissue. The resulting damage may not be visible on the patient’s skin, but make no mistake, this is basic physics. This energy is being accidentally deposited in healthy tissue below the skin and if it is a high dose, it will destroy the tissue over time. No patient wants or would be comfortable with these therapies being extravasated and some future therapies will employ more dangerous isotopes. Furthermore, none of those opposed to reporting these extravasations would ever want to experience a large extravasation for themselves or a family member and not be told that it happened and not have it measured.
Hope also suggested that an extravasation is reabsorbed; it does not remain in the arm and therefore reduces the risk of prolonged tissue damage. It’s true that the patient’s lymphatic system can clear a radiopharmaceutical from the tissue, but that doesn’t mean that the tissue near the extravasation didn’t receive an unnecessary high dose of radiation before the lymphatic clearance was completed. Additionally, the reabsorption of the extravasation has exposed the patient’s lymphatic system to an unintentional radiation dose. Reabsorption is also misleading in other ways. Imaging typically happens approximately 1 to several hours after diagnostic radiopharmaceuticals are administered and sometimes 24 hours after therapeutic radiopharmaceuticals are administered. In cases where the injection site is captured in the image, the amount of radiation detected in these delayed images will be less than the radiation that was present at initial extravasation based on the patient’s individual biological clearance.34Without understanding the clearance rate, clinicians can be deceived regarding the actual severity of the extravasation.
I understand that nearly every medical procedure has an associated risk. I also understand the incredible benefits that nuclear medicine provides. The risk-benefit ratio of a nuclear medicine procedure changes when a large radiopharmaceutical extravasation happens. These risks are almost entirely preventable. If centers are required to report these large extravasations, they will have to tell patients. Centers will do what they should have done 45 years ago: they will ensure that clinicians who administer radiation have the proper tools and training to minimize extravasations. Because CT extravasations (large swelling) and chemotherapy extravasations (intense burning) are obvious to patients and compromise their procedure, physicians have acted to reduce these extravasations. Large national registries, involving hundreds of thousands of patients, document the improvement over time.35,36 These CT and chemotherapy administrations experience an extravasation rate of 1 to 2 patients in a thousand. As previously noted, the literature suggests 172 nuclear medicine patients in 1000 are extravasated.
To be fair, I am not suggesting that all of these extravasations are large enough to be reportable or will be material to the patient and their care, but our experience monitoring over 40,000 administrations suggests that many are. The science of quality improvement is also clear; small mistakes or “near misses” are important to capture. Doing so leads to improvement. The evidence from nuclear medicine quality improvement projects also shows that when centers start monitoring for extravasations, improvement happens within months.24 The occurrence of all radiopharmaceutical extravasations would plummet if NRC made large extravasations reportable like any other medical event.
Some physicians will agree with Hope that reporting and documenting is burdensome. Patient safety experts will tell you arguments like these are common from those not wanting to be monitored. I would add these arguments are flat out wrong. Effective monitoring of radiopharmaceutical administration takes very little time. Measuring large extravasations takes 5 minutes or less and can be done using free software made available by the University of Tennessee at Knoxville (https://gsm.utmck.edu/research/MITRP/RIDE.cfm). Once a center understands what causes their extravasations, the frequency of these mistakes plummet. Kiser and his team, who have monitored almost 20,000 radiopharmaceutical administrations, know that being focused on reducing extravasations and being focused on the patient is not burdensome.17
As I noted above, I petitioned the NRC to correct their reporting exemption. While the NRC staff may not have known extravasations could be prevented in 1980, they certainly knew their policy was incorrect since 2008 – this is documented in their own meeting notes with their advisers who told them extravasations could be reduced with better tools and training.37 These advisers told the NRC that they simply didn’t want to report extravasations. They told the NRC some extravasations result in high radiation doses but that they didn’t want to deal with all the “blah, blah, blah” involved in reporting.38
I have diligently worked with the NRC since 2018. I have met with the staff and the commissioners. I have presented to their advisers. Time and time again, global experts and I have shared scientific evidence supporting the need to report extravasations no differently than any other medical event, and time and time again, NRC has failed to take the appropriate steps to correct their mistake. The NRC’s own Inspector General has found that paid NRC advisers who were leaders in the medical communities that are opposed to reporting provided their input to the staff.39 Input that the staff ended up using to make an improper recommendation to the Commissioners (https://lucerno.com/wp-content/uploads/2023/02/2023-02-15-ICR-letter_signed.pdf). As the inventor of the PET/CT scanner, David W. Townsend, PhD, stated about this issue, “the regulated are regulating the regulators”.
As a result of my experience with the NRC, I completely endorse the Nuclear Medicine Clarification Act. Congress needs to act to push NRC to do what they should have done in 1980 and then again in 2008. Just as Congress had to intervene in the 1990s to introduce the Mammography Quality Standards Act, Congress needs to ensure that the NRC and the nuclear medicine community are transparent with the quality of nuclear medicine administrations. We are all going to be nuclear medicine patients at one time or many times in our lives. It is time to fix the extravasation issue.
Elevating the Quality of Cancer Care via Cross-Department Collaboration
Experts from Sibley Memorial Hospital discuss how multidisciplinary work has enhanced outcomes such as survival and resource use at their institution.