The cumulative radiation dose paradigm in pediatric imaging

Abstract

Medical imaging professionals have an accountability for both quality and safety in the care of patients that have unexpected or anticipated repeated imaging examinations that use ionizing radiation. One measure in the safety realm for repeated imaging is cumulative effective dose (CED). CED has been increasingly scrutinized in patient populations, including adults and children. Recognizing the challenges with effective dose, including the cumulative nature, effective dose is still the most prevalent exposure currency for recurrent imaging examinations. While the responsibility for dose monitoring incorporates an element of tracking an individual patient cumulative radiation record, a more complex aspect is what should be done with this information. This challenge also differs between the pediatric and adult population, including the fact that high cumulative doses (e.g.,>100 mSv) are reported to occur much less frequently in children than in the adult population. It is worthwhile, then, to review the general construct of CED, including the comparison between the relative percentage occurrence in adult and pediatric populations, the relevant pediatric medical settings in which high CED occurs, the advances in medical care that may affect CED determinations in the future, and offer proposals for the application of the CED paradigm, considering the unique aspects of pediatric care.

Introduction

The value of medical imaging, including those modalities that depend on ionizing radiation, in the care of patients is not questioned. However, these overwhelming benefits of medical imaging do not permit a dismissive attitude towards risks, real or potential, in our covenant of welfare with those that we help care for. Specifically related to ionizing radiation, members of the medical imaging community have an individual accountability and collective responsibility for the radiation protection tenets of justification and optimization. These are no less important to acknowledge for an individual examination than in the sphere of multiple imaging examinations. This scenario of multiple exposures is both compelling and provocative as dose accumulation currently is invoking additional dimensions of radiation responsibility beyond those we are more familiar with for a single event. Factors in this additional scope of responsibility include how often do multiple examinations occur, in what settings, for what populations (e.g., age), the influence of advances in technology and clinical care, the impact that measures of multiple exposure events have in influencing individual imaging decisions and in longer term imaging strategies such as for chronic conditions, and the consequences on attendant attitudes of patients and caregivers. There are profound differences in the imaging care of children compared with adults, especially with respect to the use of radiation^1,2 and these are also considerations in the discourse of repeated radiation exposures. Understanding the relative significance of repeated exposures in children leads to the subsequent discussion of what can or more importantly what should be done to manage this.

For the purposes of the following discussion, some terms in the construct of CED warrant clarification. While “repeated” or “recurrent” or “multiple exposures” for imaging examinations are terms that have been used, cumulative effective dose (CED) is currently prevalent and will be used subsequently. Threshold doses for CED, also labeled as “high” vary but most are >100 mSv.³ An “examination” is one that depends on ionizing radiation. “Cumulative” applies to separate examinations, not repeated exposures within an examination such as additional series with a CT examination. The pediatric age group discussed here most often represent individuals < 18 years of age although in some publications on CED this group has been included in a broader and arbitrary definition of “younger“, such as <40 years.^4–6 “Parent” represents any caregiver responsible for the welfare of the child. Finally, “risk” relates to the stochastic risk with the inherent uncertainties typical of diagnostic levels of radiation exposure.⁷

To the best of my knowledge, and recognizing that expert opinions on CED are available that are age-generic,^8,9 there is no contemporary publication that addresses both the scale of CED in children compared with adults and discusses the value of this information in fashioning medical imaging decisions for recurrent imaging for children. The citations provided are intended to provide an audit of information on the settings and frequency of CED, emphasizing children. The adult studies cited are also not intended to be comprehensive, but are relatively current and serve as a sample range for the frequency of what is a considered high CED in that population. The adult CEDs, including the clinical context these doses are encountered, most importantly can serve as a measure for comparison with frequency and settings in the pediatric population.

There are several reasons for reporting CED. First it is currently a topic that is receiving increased attention in the literature. In addition, the International Atomic Energy Agency (IAEA) recently completed technical meetings on cumulative dose in March 2019 and October 2020.^10,11 A resultant statement from the IAEA was recently released.¹² The International Society of Radiology is also preparing a document on cumulative dose (personal communication; Guy Frija, MD). The increasing prevalence of radiation dose monitoring demonstrates that it is important to account for an individual’s single as well as cumulative exposure history. The accountability for radiation dose management for individual exams has an understood consequence for repeated exposures. There is, then, a professional duty to know the frequency of and settings for high CED for any age as there are important differences between adult and pediatric populations.

Comparison of adult and pediatric high CED investigations

Several recent investigations have addressed CED in adults or unspecified ages,^13–15 or within or inclusive of the pediatric age range^4,16–30 (Table 1). These investigations illustrate some of the challenges with interpreting and comparing CED. These include various periods of time studied and sample sizes, varying threshold classifications of CED, a mix of generic inclusion criteria for imaging versus focus on specific settings such as oncology, and modalities with variable inclusion of radiography, fluoroscopy, and nuclear medicine imaging in addition to CT. Age is an essential component of approaches to CED given life expectancy and age-related radiation sensitivity. A CED of 100 mSv in a 90 year-old engenders a different deliberation than the same CED in a 9 year-old.

Table 1.

Investigations for Cumulative Effective Dose in Diagnostic Imaging Emphasizing Pediatric Populations

Investigator	Population Agea	Period of Study	CED Threshold or Other Value	Measure Meeting CED Threshold (Denominator)	Comments
Rehani et al13	NAb	5 years	>100 mSv	0.5–2.9% (estimated 1.2 billion)	Survey of 35 OECDb countries
Rehani et al4	All	5 years	Only those with ≤100 mSv were included	0.08% (of 8,952 for all ages meeting CED threshold); 0.08% of the total <20 years	CT exams; 90.4% of all patients with malignant diagnoses
Rehani et al17	All	5 years	>100 mSv (other thresholds discussed)	0.03% in a single day; ≤20 year olds accounted for 0.0009% of total population >100 mSv (3,880,524 CT patient days)	CT exams; 279 hospitals; assessed CED for patients in a single day
Jeukins et al19	All	4.5 years	>100 mSv	1.5% male 1.9% female (probability of CED > threshold); 0% < 18 years	CT exams. Malignancy accounted for 79.7% of high-dose cases
Frija et al20	All	4 years	>100 mSv	0.5%; (23 European countries); range 0–2.72%. 0% < 18 years	CT exams
Chen et al14	18–64 years	3 years	>20 mSv/year	3.3/1000 (0.33%) enrollees in 5 US healthcare markets	Cardiovascular disease
Stopsack et al15	>18 years	10 years	>100 mSv	1.9% (of 26,337 of all patients having at least 1 CT)	CT exams
Rehani et al18	All	5 years	>100 mSv	1.33% for all ages. For <20 year group, 0.3% of 33,407 of those meeting threshold; 0.004% of all patients (2,504,585); from publication’s table 5	CT exams
Gil et al16	All	8 years	>100 mSv	0.7% (984,481 enrollees in final cohort); 0–22 years 0.006% > 100 mSv	Multiple modalities
Johnson et al21	<6 years	Up to 5.2 years	Median CED overall 2.7 mSv	For >100 mSv, two diagnoses (5%, median, 95%) Transplant: 9.9,64,190 mSv Norwood: 0.7,29,114 mSv)	Total of 7 cardiac diagnoses; multiple modalities. Highest median CED 6.8 mSv
Ward et al22	<18 years	4 years CT and fluoroscopy; 1 year radiography	18 year estimated CED 3.5 mSv	Estimates based on 65 children	Cystic fibrosis; multiple modalities
Chawla et al23	<18 years	Nearly 5 years	>100 mSv	Estimated from publication’s figure 6: 14/78 or 18%; (78 children)	PET/CT: Average CED 78.9 mSv
Kim et al24	“Pediatric Patients”	12 years	Mean CED 40.2 mSv	(63 patients)	Neuroblastoma; multiple modalities. 44% also had radiation therapy with mean dose of 32 Gy
Johnsen et al25	<30 years	8 years	Mean CED 34 mSv	No patient >70 mSv (20 patients)	Ewing sarcoma; multiple modalities
Lee et al26	<18 years	5 years	“High” considered >30 mSv. Median CED 37.1 mSv	No child had CED >50 mSv (from publication’s figure 6)	58% of “high” dose patients in malignancy population
Özyörük et al27	<18 years	5 years	62.9 median CED for all diagnoses	No diagnosis upper quartile exceeded 100 mSv (from publication’s figure). (88 children)	seven cancer diagnoses. Multiple modalities
Lumbreras et al28	All	12 years	>100 mSv	1.5% of all ages and 0.3% < 15 year group of 2298 of those meeting CED threshold; <15 year group was 0.003% of entire population that was imaged (n = 154,520). 0% < 10 years	Multiple modalities
Brambilla et al29	Review of multiple pediatric investigations	varied	NA	Annual CED was <3 mSv per year all but inflammatory bowel disease group where annual CED 2–20 mSv in Crohn disease population	12 studies: cystic fibrosis, inflammatory bowel disease, congenital heart disease, shunted hydrocephalus, hemophilia, spinal dysraphism
Marcu et al30	<18 years	10 years prior to publication	NA	45 studies met criteria. No report or estimated pediatric lifetime CED of >100 mSv	For diagnostic imaging covered head/dental, chest, cardiac, spine/bone, abdomen/pelvis, and “general” categories of exposure

^aAssumed at entry into study

^bAssumed; only number of CT examinations per country specified in methods

^cOECD: Organization for Economic Co-operation and Development

The different prevalence of CEDs between adults and children is important in informing the scope and scale of actions. Reviewing investigations that reported CED of high exposures (>100 mSv) in the pediatric age range from population that included adult data (ages < 20 years; one investigation age range was <30 years and another <22 years), the relevance to children is extremely low; the occurrence ranged from 0 to 0.3%.^4,16–20,28 Of those studies discriminating pediatric data, 5/13 (38%) reported no CED >100 mSv. Note that when results were available for all subjects as the denominator (no threshold considered) the percentage range of children meeting the >100 mSv threshold reported decreased to 0–0.08%^16–20,28 The adult or all-age reported percentage occurrence in populations imaged ranged up to 1.9%^{4,13–16,19,20,28} while Vano cites studies reporting 0.6–3.4%;⁸ the most recent survey from Europe reports an occurrence in 0.5% of those having CT scans,²⁰ agreeing with reports from Arellano and Gil.^5,16 Simply stated, any drive to address CED in children should be resonant with the fact that the range of the percent occurrence is approximately 1/20^th- 1/100^th that of the adult population. In addition, most of the reported high cumulative exposures are in the oncologic population in children as well as with adults;^4,19,20,26 these oncologic settings often have protocol driven and fixed imaging algorithms for surveillance. Further adjustments in radiation exposure for this surveillance are always valuable but resources, such as development of consensus imaging pathways, are especially time consuming and imaging in this population often is performed for acute or unexpected events rather than for the surveillance pathway. Furthermore, periods of time for chronic care, especially in oncologic care, used for CED estimations may not reflect the intensity of imaging surveillance for example early in diagnosis than in follow up years later. A five-year survey at the time of cancer diagnosis likely will not reflect the same survey period five years after diagnosis. Subtracting the oncologic population from those having a high CED in the pediatric population further diminishes the very low prevalence in the pediatric population relative to the adult population.

Two systematic analyses among those listed in the Table provide additional insights into the challenges with CED reporting. In a thorough review to that point in time in the pediatric population with non-oncologic chronic conditions, Brambilla et al looked at 137 matches for search engine descriptors, excluding all but 12 investigations.²⁹ Investigators identified groups with low CED consisting of the number of studies with patients with cystic fibrosis (n = 3), inflammatory bowel disease (n = 3), congenital heart disease (n = 2), shunted hydrocephalus (n = 2), hemophilia (n = 1) and spinal dysraphism (n = 1). In this group, the annual CED was <3 mSv per year in all but the inflammatory bowel disease investigations. The annual CED was considered moderate (>3–20 mSv) in the studies in children with Crohn disease. Investigations reviewed by the investigators constituted a time period between 2007 and 2013; however, data represented time periods that were often more than a decade old and noted in one investigation in the Crohn population to date back to 1991. There was an excellent discussion of limitations with investigations of CED including effective dose estimations such as from CT parameters, lack of ability to account for the number of phases in a CT examination, lack of inclusion of all examinations using ionizing radiation modalities, variability in performance between institutions/practices, and relatively small sample sizes. Additional challenges in comparing studies and applying conclusions can be added to this list from Brambilla and include different lengths of time in which data were included, challenges in converting all ionizing radiation modalities to effective dose, especially in fluoroscopy and interventional procedures,³¹ regional practice patterns which may account for variations in reporting of high CED^20,28,29,32 and variable thresholds for and significance of “high” CED.³³ While this is a very informative review of the status discussing pediatric CED, there was no substantive comparison with adult CED, more developed discussion of how pediatric CED information might impact acute and longer term care, and material was dated (acknowledged by the investigators); together, these limitations support a contemporary and more comprehensive discussion of CED in the pediatric population. In a more recent publication, Marcu et al reviewed in detail a combination of radiation doses (a mixture of organ and effective doses), risk estimations and cumulative exposures for children over the prior 10 years from 45 publications culled from a large database search.³⁰ In this material, there was no report of cumulative doses over 100 mSv, nor projections based on annual estimations of reaching that threshold during the 18 years of childhood, although there were CED reports of 50–100 mSv.

The complex paradigm of cumulative dose in children

In anticipation of looking at the more controversial application of cumulative dose monitoring (i.e., in medical decision-making for an individual child, or in designing imaging algorithms for events or longer-term conditions that warrant repeat imaging such as some chronic conditions), the value and detriment both need to be considered. The most familiar - although not the only detriment³⁴ - is radiation exposure and that of a cumulative risk of cancer from this radiation. One could argue that there are more simple or accurate representations of cumulative risk. For example, simply knowing the number and type of prior examinations might in many settings be reasonable currency for the referrer to make decisions on the overall benefit and detriment of additional examination. More detailed dose estimation information could subsequently be derived on an as-needed basis from this examination information. A more detailed and generic measure of risk assessment would be accumulated organ dose although the large scale availability and application of this does not exist or at least is not evident in the general discussion regarding CED. Finally, a more representative element could be the effective risk or a risk index.^35,36 This metric can be estimated (with attendant uncertainties) and while might be steps closer to representing the detriment of ionizing radiation risk than CED, an adherence to cumulative risk in a paradigm of repeated examinations would almost certainly promote elevated concern, potentially resulting in either an exam not being requested by a referrer or even a patient/parent refusing an examination. Given the awareness and concern with the current representation of radiation as a component of a single imaging examination, this cumulative risk would be very difficult to defend, and might be best thought of as contributing to a construct of cumulative concern, anxiety, or fear. The value of effective risk should be carefully curated and, when and if use is warranted, through a targeted approach.

A factor that belongs in this discussion of the measures of cumulative doses is the acknowledgment, even if difficult to define, of the cumulative benefit of medical imaging. This promotion of benefit seems, based on current CED publications, to be less front and center than cumulative risk and also less extant in the discourse in publications with individual radiation events. CED must have a like accounting of the cumulative benefit. I would offer that the statement “Imaging shows us that your daughter is still clear from cancer” months or longer after therapy carries a much more profound and supportive impact than “Your child now has had seven CT examinations with a cumulative effective dose of 50 mSv”. We are responsible for cumulative reassurance as one of the benefits of medical imaging. Finally, we emphasize dose monitoring as a surrogate for exam performance and elements are evolving for the addition of a quality monitoring component,^37–40 some of which are automated. Performance is best represented by both dose and resultant quality with the cumulative benefit a result of both of these. For example, a series of CT examinations with CT dose indices (and resultant effective doses) that are relatively high in one patient may have reasonable quality that justifies that dose, providing for a more balanced, informed and supportive discussion with respect to the magnitude and importance of the CED.

Effective dose, Accumulation and Relevance in Children

Martin et al⁴¹ recently reviewed the topic of effective dose for all ages, providing a basis for this metric, including limitations on potential applications although stating, “uncertainty should be acknowledged but should not detract from the value of the E [effective dose]“⁴¹ (Page 91). Moreover, the International Commission on Radiation Protection has drafted a forthcoming comprehensive position on this metric in medical imaging. These are additional challenges that are modality based with the very ability to determine effective dose, such as for individual fluoroscopic exams or fluoroscopically-guided exams especially for children.³¹ What is more relevant is the cumulative aspect of effective dose. As Durand et al discuss, cumulative dose supplies cumulative uncertainties which are only amplified when considering risk.⁴² A threshold for cumulative dose alerting should not be uniform; this would necessarily vary by age and medical condition even without considering what one should do with such a threshold. Even without these considerations, the threshold for what is labeled as “high-dose” varies.^26,28,32,33 Interpretation of the significance of this threshold is more complicated in children than in adults. One issue relevant to children is what does any threshold mean and how might this need to be considered dynamically in terms of risk based on the age of the patient, given changing susceptibility to radiation throughout the first 18 years of life. The period of time for the total cumulative effective dose is also an issue relevant to pediatric care as children age out of the specialty and accounting in those early years will almost certainly not reflect the entire contribution of a lifelong chronic illness.³³ There is no certainty for what might be imaging evaluation in adulthood. Simply stated the lifetime radiation record is far from understood, let alone predictable. Inherent in this aging process is some consideration to when the discussion of CED might change from the caregiver to the patient who becomes an adult, and how this discussion should be configured. This is a unique aspect of the transition of care into adulthood. These types of transitions particularly with chronic conditions can be a purgatory between adult and pediatric providers.^33,43,44

The landscape for use of ionizing radiation in children is also shifting, as recently discussed with CT patterns in younger children.⁴⁵ As Brambilla also notes for the non-oncologic conditions he identified with CED in children,²⁹ there is a migration to nonionizing modality evaluation in many of the conditions he cited. For example, for inflammatory bowel disease, MR enterography is increasingly valuable and replacing upper gastrointestinal and small bowel follow-through exams and has increasingly recognized benefits compared with ionizing radiation modalities.^33,46,47 Imaging evaluation of cerebral ventricular size, especially in a setting of ventriculoperitoneal shunts, has evolved in pediatrics over the past few years to low dose CT and increasingly fast MR examinations.⁴⁸ With advent of PET/MR, including a lower dosage of radiopharmaceutical, radiation dose can be substantially diminished in the pediatric population especially given the contributions of PET imaging to CED dose profiles.^49–51 In interventional radiology, ultrasound as opposed to CT guidance is more common in many like applications in children compared with adults.⁵²

Other elements in weighing reports on CED in children and adults are advancements in both clinical care and need for imaging as well as technological advancements in imaging equipment over the past 10 to 20 years.^8,22 Improvements in understanding of disorders including chronic conditions may obviate current imaging practice, with anticipation that diagnostic imaging may be less frequent. It should also be recognized although that improvements in outcome and survival in childhood may be consequential in increasing CED populations into adulthood, recognizing that this trade off with survival and additional imaging needs highlights our duty to emphasize the cumulative benefit in what we do.

Actions based on Cumulative Effective Dose in pediatric imaging

Given what are globally seen as the benefits and challenges with the measure of CED relevance to children, and understanding why this construct is timely especially through comparisons with adult and pediatric investigations, what can we say about what to act on with CED information in pediatric population? Contributing to the substantially lower percentage of high CED in children is that children undergo much less frequent imaging than adults. This is obvious as they are generally healthier. Looking at ionizing radiation exposure is in United States in the pediatric population, in 2016, 91% of annual per capita ionizing radiation collective effective dose was from background sources and 9% for medical imaging.⁵³ Compare that with the nearly 50% per capita collective effective dose from medical sources in adults. The pediatric contribution to the total collective effective dose is 3%.⁵³ Moreover, in children CT accounts for about 85% of all of the medical ionizing radiation contribution to the collective effective dose in children, although the percentage of examinations is approximately 9.5% of the total from all ionizing radiation modalities.⁵³ This offers validation for the fact that the majority of CED publications report CT exams alone or in combination with other modalities.

For cumulative effective dose, how and also to what end should this information be used to advance imaging care in children? Some scrutiny of CED application is warranted and supported by the evolving presence of dose monitoring across all ages. The importance and evolution of dose monitoring is widely supported including the Bonn Call for Action⁵⁴ and the Basic Safety Standards.⁵⁵ Dose monitoring programs afford individual exam and multiple exam dose information. Our professional emphasis on dose management for each exam in individual patients applies and has benefits when a cumulative dose profile occurs from multiple examinations. The cumulative nature of the dose here is the expected cumulative justification and optimization for each exam. We should fully account for all medical radiation delivered and as imaging professionals be active participants in discussions for what should be done with this information. Much of the work in the past few years with cumulative effective dose has been very valuable in clarifying the frequency of and in many cases clinical settings in which high cumulative doses occur or are likely to occur, including projections over lifetime. This information fulfills an important part of our stewardship as a community of imaging professionals and should be embraced, including a call for broader groups, including industry, to facilitate this accounting.

However, the application of CED inherent in the accountability element is a different dimension than simply the very defensible act of accounting; more simply, the responsible use of CED is more challenging, especially for children. This is underscored with a brief review of facts discussed in more detail earlier. The percentage occurrence in adults and children is quite different and can be about two orders of magnitude lower in children. In addition, the oncologic population is often cited in children and much of the surveillance care is prescribed. Non-oncologic populations are much less impacted. The current movement even in the pediatric population cited such as ventriculoperitoneal shunt evaluation and inflammatory bowel disease is more toward magnetic resonance imaging. Also, CED reported are often more than 10 years old and can be more than two decades old, and unrepresentative of increasing dose reduction strategies for examinations, technical advances, and migration away from ionizing radiation modalities. Medical diagnosis and management, including for chronic conditions from clinical standpoint are constantly being advanced, and it is not unreasonable to expect less imaging, including ionizing radiation imaging, need. Suffice it to say that the investment in minimizing repeated exposures beyond what we do for any individual examination has a relatively small signature.

Beyond the necessary accounting for CED (or surrogate), a more challenging issue is what would this information be helpful for? Arguments have been advanced for the importance of and the lack thereof of CED at the point of care.^9,20,56–61 An application of CED is in decision making, both in the acute as well as imaging planning for chronic conditions which likely warrant recurrent imaging examinations. An assumption about CED in point-of-care decision making is that this information is available, and that CED measures and the relationship to risk are understood by the provider who would (or together with the patient or parents) make imaging decisions. The former is embedded in the appropriate use criteria, as one example of decision support, but educational materials linked to this decision process could also be of value.⁸ However, the level of detail provided must be understandable, and also potentially contribute to informed discussions about decision between the provider and the patient or caregiver. The latter has been challenging.^62–68 The presence of cumulative dose in terms of designing appropriate imaging strategies beyond what might exist for individual point-of-care decision implies a shared effort by imaging experts and clinical professionals.

Proposals for the application of Cumulative Effective Dose from medical imaging in the pediatric population

• Clarification of and promotion by the imaging community on the cumulative benefit of recurrent imaging should be included in efforts and be emphasized. This can include quantification of cumulative quality into the performance of imaging examinations as an enhancement of imaging performance beyond dose alone.

• There needs to be continued adherence to justification and optimization as this is relevant for an individual and recurrent imaging examinations. Resources directed to addressing improvements in individual examination (e.g., protocol) dose reduction strategies have a potentially much larger return than those targeting multiple exposures especially due to the infrequent occurrence of high CED in children. The following is a hypothetical example. Consider an overall population of 100,000 children of which 1% (1,000 children) has one abdomen pelvis CT in a five-year period. In this overall population, .001% (or one patient, based on previously summarized reviews of adult and pediatric percentage of occurrence) has a high cumulative effective dose (>100 mSv) in this time period from several CT examinations. Restructuring CT protocols to reduce multiple phase examination occurrence to a single phase from 10% of the 1,000 children (100 children) to 5% (50 children) for an 8 mSv estimated effective dose per phase reduces the overall population effective dose from CT of the abdomen and pelvis by 400 mSv (8 mSv × 50 children). Efforts invested in reducing the CED by 50% for this single patient is a savings of 50 mSv, or 1/8^th the return.

• CED efforts should include age-based distinctions especially between pediatric populations and adult populations as the combination of age groups diminishes the significance of conclusions for the frequency of occurrence of high CED, application in different medical conditions, and risk considerations (i.e., implications for a high CED for an 9 year-old versus a 90 year-old).

• Collaboration should be cultivated between imaging and clinical specialists for longer-term imaging strategies that adhere to justification for individual examinations but also are inclusive of measures for and communication of cumulative exposure, if any, that are relevant and structured to be accessible, clear and helpful to the individual specialty in making acute and longer term imaging decisions.⁶⁸ It may be that in an emergency setting that any prior radiation history in clinical decision support is irrelevant thus unnecessary to the immediate decision for imaging care of another episode of acute abdominal pain while in setting of chronic kidney stones that this information on the type and frequency of imaging and attendant dose information is important for shared decisions between the urologist and patient or parent. Models for hierarchical cumulative monitoring programs that afford various levels of information could include the number and type of examinations, effective dose, and organ doses and effective risk, as appropriate for the medical setting and specialty.

• Appropriate societies and organizations should consider recommendations for measures of CED, including what constitutes a high CED that should have age and medical setting considerations, as well as promoting clarity about the significance and complexity of snapshots in time (e.g., single complex event such as hospitalization for major trauma versus a 3- or 5- year survey) relative to lifetime projections and estimations of exposure. These should include emphasis on continuity of mechanisms to address CED during the transition of pediatric to adult care and be mindful that the concept of a threshold in medical care can serve as a guide but not a prohibition.

• Cooperation with industry is essential in the development and implementation of technology through efficient and effective alerts that are configurable and can identify outliers in CED during dose monitoring based on relevant population (i.e., age and medical setting) that may warrant investigation or action. Pediatric expertise should provide input for relevant thresholds for such work in the pediatric age group.

• There should be continued investment in understanding of low-level radiation risk relative to dose metrics, with modifications in threshold CED based on greater understanding of risk.

• Improvements in and implementation of dose monitoring technology and programs should be facilitated. Information from dose monitoring programs, including those for CED, should be integrated, in addition to general radiation educational and communication material, in clinical decision support.^8,20,68,69 Communication strategies should be created by imaging societies including communication expertise with content and delivery methods that are mindful of cumulative anxiety in these dialogues.^70,71

These proposals overlap to a large extent with those recently released from the IAEA,¹² in turn supported by two recent detailed reviews,^72,73 although some points and the emphasis on overlapping aspects of the above statements and IAEA statements about CED vary. The above statements are intended to relate specifically to CED without revisiting existing approaches to radiation protection in medicine. The above statements also emphasize actions that account for children in the CED paradigm, and encourage more detailed engagement with clinical services and standardization of the CED paradigm. More unique items from the IAEA action items include the additional importance of collaboration with industry for continued improvements in dose reduction developments and also promote customization of imaging protocols.

Conclusions

The responsibility for single examination dose and quality by the imaging community applies for repeated examinations. As we gain a necessary understanding about the frequency across ages and in select medical settings, existing guidelines and recommendations centered on justification and optimization should be emphasized for their impact in radiation management. When recurrent imaging, with potentially relatively high cumulative doses, occurs, close alliances with respective clinical specialties would serve to address if, when and what cumulative exposure metrics are valuable in designing or revising surveillance imaging strategies. Efforts directed at the paradigm of cumulative effective dose in children should take into account specific populations involved (age and disorder), the infrequent occurrence of relatively high CED, the migration of imaging care away from ionizing radiation, and considered communication that is engaging, informed and sensitive in shared decision making among colleagues, patients and families that minimizes the anxiety and potential resistance to appropriate medical care.