Abstract
Millions of patients benefit from medical imaging every single day. However, large-scale multicenter studies published in the last 6 years have brought new result that allow extrapolation to nationwide or global levels. They have opened a new era wherein millions of patients are receiving cumulative doses in three digits of mGy of organ doses or three digits of mSv of cumulative effective dose (CED) every year. One out of 125 patients can be exposed to an effective dose ≥50 mSv from a single computed tomography (CT) exam, and 3 out of 10,000 patients undergoing CT exams could potentially receive cumulative effective doses ≥100 mSv in a single day. Recurrent imaging with CT, fluoroscopically guided interventions (FGI), and hybrid imaging modalities such as positron emission tomography/computed tomography (PET/CT) are more prevalent today than ever before. Although a major fraction of patients with such high doses is ≥60 years of age, the number of patients <60 years and with diseases that do not substantially shorten life expectancy is estimated to be nearly half a million added every year. Moreover, data from a centre that employs the best form of justification and the best form of optimisation indicate that we have reached a limiting point in providing safer imaging to many patients using currently available imaging machines and justification and optimisation tools. This leads us to an unprecedented era where collective thinking and actions are needed, and it provides a fitting opportunity for the call given by the International Commission on Radiological Protection (ICRP) to develop solutions to achieve safer imaging, a mission that necessitated the creation of ICRP in 1928.
Keywords
INTRODUCTION
The International Commission on Radiological Protection (ICRP) was established in 1928. A number of factors necessitated the creation of ICRP, but primarily, it was driven by growing concerns about the effects of ionising radiation (tissue injuries) among workers in medical facilities. In subsequent years, it became evident that stochastic effects are possible and may have long latent periods. There may be controversy on stochastic effects at low doses, but there is general agreement among almost all international and national organisations involved in radiation effects, radiation protection, and radiation regulations that stochastic effects at organ doses of ≥100 mGy and possible in many organs (ICRP, 2005, 2007; NCRP, 2018). In recent years, there has been increasing evidence of radiation effects at doses of a few tens of mGy, whether acute or protracted (Rühm et al., 2022).
Now, in 2026, we are in a similar or rather more severe situation than in 1928, with millions of patients receiving relatively high doses of several tens of mSv of cumulative effective dose (CED) and rather ≥100 mSv (Brower and Rehani, 2021; Rehani and Brady, 2021). It must be emphasised that choosing 100 mSv as a parameter to focus upon is not based on any threshold for radiation effects, but at this level of effective dose, many organs in the body receive >100 mGy of dose at which there is a high degree of consensus on stochastic effects among official organisations dealing with radiation effects (Rehani et al., 2020a, 2020b; Zewde et al., 2022). Since many organs are exposed in each computed tomography (CT) scan or in most medical imaging examinations, effective dose and cumulative effective dose are the only way out for communication, whereas organ doses can be used when estimating risk in each organ (Martin et al., 2020). The largest part of such doses comes from CT exams, whereas fluoroscopic-guided interventions, hybrid imaging, and several nuclear medical exams also contribute to high patient doses. Recent studies on the magnitude of patients involved (Ferretti et al., 2025a, 2025b) are creating an alarming situation and need actions by stakeholders to make patients safer. It is clear that we are in an unprecedented era never before witnessed since the discovery of x-rays and radioactivity over 125 years ago.
This paper touches upon various aspects not only to elaborate on the extent of the high doses involved but also to conduct cause analysis, identification of patients where such high doses may be of little concern, and what actions are needed by ICRP and professional societies.
ASSESSING THE MAGNITUDE
Computed tomography (CT)
Publications on high cumulative doses in patients undergoing recurrent CT exams have appeared in the literature since 2009 (Griffey and Sodickson, 2009; Rehani, 2009; Sodickson et al., 2009; Desmond et al., 2012; Seuri et al., 2013). Most of the publications in previous years pertained to specific diseases or situations, e.g. cardiac diseases, renal disease or dialysis, gastrointestinal disorders, traffic and occupational accidents, and COVID-19 (Li et al., 2023). The data thus involved restricted cohorts that gave the impression that such patients with high doses (typically ≥100 mSv) may not be widely prevalent. Furthermore, it did not allow extrapolation of data to get a national or global picture. A number of recent studies adopted a different approach. They estimated the number of patients with defined levels of dose, such as those with CED ≥100 mSv as a percentage of all patients undergoing exams, such as CT exams, in that hospital. Additionally, they estimated the average number of CT exams per patient (Rehani et al., 2020a, 2020b, 2020c). That helped in transitioning to the number of patients as that information was not available, with most national and global data availability being on the number of CT exams, not the number of patients undergoing CT examination. The data from a large study covering 2.5 million patients who underwent 4.8 million CT exams found that patients underwent a median of 6 CT exams in a year and some patients received up to 109 exams over 5 years (Rehani et al., 2020a, 2020b, 2020c). Based on information from large studies covering several hundred hospitals and many countries, extrapolation using the most conservative numbers indicated that about 0.9 million patients may be getting added every year with CED of ≥100 mSv globally, only from CT exams (Brambilla et al., 2020; Rehani et al., 2020a, 2020b, 2020c). Another study covering 35 countries of the Organisation for Economic Cooperation and Development (OECD) region indicated that about 2.5 million patients reach a CED level ≥100 mSv in 5 years in these countries (Rehani and Hauptmann, 2020). Using average figures, it appears that nearly 2 million patients may be getting added each year with 100+-mSv doses globally as against 0.9 million mentioned above that was based on conservative estimation. Although not specifically assessed, it is apparent that patient doses for the same information have gone down over the years, but actual doses to a small number of patients (<1% to 6%) have increased (Rehani et al., 2020a, 2020b; Brower and Rehani, 2021). Many tend to convey that 1% or less is not worth giving importance. A recent paper has compared the perceptive value of 1% and concluded that despite there being a much higher chance of developing cancer from 100-mSv radiation than being involved in a commercial plane accident, there is much less emphasis on patient radiation safety than aviation safety (Mataac and Rehani, 2024).
Fluoroscopic-guided interventions and hybrid imaging
While the largest number of publications pertain to cumulative doses in CT, similar observations have been made in the use of fluoroscopically guided intervention (FGI) (Li et al., 2020; Rehani et al., 2020a, 2020b) and from positron emission tomography (PET), with hybrid PET/CT studies (Chawla et al., 2010; Indrakanti et al., 2022; Abuqbeitah et al., 2023). The percentage of patients with CED ≥ 100 mSv is much larger with PET/CT as most patients are for follow-up of malignant disease. However, the relative number of patients undergoing PET/CT versus CT alone is much smaller, typically 1/5th, and this fraction can vary by a large margin depending upon the hospital profile. A recent paper gives the distribution of patients in different exams at the author's hospital, which is a high-level tertiary care hospital in the USA (Li et al., 2023). The mean age at first procedure was 59.3 ± 14.2 years, and the median was 60.0 years.
Readers are referred to the review article for further details and more references (Brower and Rehani, 2021).
Total dose from exams involving high doses, age distribution, and body build
Till recently, the data on total cumulative dose to individual patients were mostly available in specific diseases or situations, e.g. cardiac diseases (Einstein et al., 2010; Kaul et al., 2010; Noor et al., 2011), renal disease or dialysis (De Mauri et al., 2011, 2012; Coyle et al., 2012), gastrointestinal disorders (Yang et al., 2022), traffic and occupational accidents (De Roo et al., 2022), and COVID-19 (Hadid-Beurrier et al., 2021). It may be kept in mind that several other publications may have appeared in the literature after this manuscript was written in Jan 2024.
A recent study at a major tertiary care hospital in the USA analysed the frequency of CED ≥ 100 mSv over a period of 4 years, with respect to patient gender, age, and body habitus classification derived from body mass index (BMI) (Li et al., 2023). This has been done for each imaging modality separately (CT, FGI, nuclear medicine) and all of them combined. Therapeutic treatments with radiopharmaceuticals were excluded.
The results showed that among a total of 205,425 patients, 5.7% received CED ≥ 100 mSv (mean 184 mSv, maximum 1165 mSv) and their ages were mostly 50–64 years (34.1%), followed by 65–74 years (29.8%), ≥75 years (19.5%), 20–49 years (16.3%), and ≤ 19 years (0.29%). Body habitus in decreasing occurrence was obese (38.6%), overweight (31.9%), healthy weight (27.5%), and underweight (2.1%). It was noticed that 172 patients received ≥500 mSv and only 3 received ≥1000 mSv. In comparison, 5.3% of 189,030 CT patients, 1.6% of 18,963 FGI patients, and 0.19% of 41,401 nuclear medicine patients received CED ≥ 100 mSv from a single modality.
It was concluded that there was 89% contribution of CT to the cohort of patients with CED ≥ 100 mSv, with 70% of the cohort being obese and overweight and 64% of the cohort aged 50–74 years. It showed that there is a 15 to 18 times contribution from CT than that from fluoroscopically guided intervention or nuclear medicine to the cohort. Furthermore, 70% of those who received ≥100 mSv were either overweight or obese.
Using the factors from the above study, it is estimated that 1.5 to 3.1 million patients are getting added annually with 100+ mSv from CT, FGI, and nuclear medicine combined. Readers are referred to a more recent paper of 2025 that gives estimates in 27 countries of OECD region (Ferretti et al 2025b).
HIGH EXAM DOSE OR HIGH CUMULATIVE DOSE?
It is not always the cumulative dose, but there are some CT exams that impart ≥50 mSv and some even 100 or 200 mSv per exam. Details about these are available from a large study covering 279 CT facilities in the USA (Rehani et al., 2021). The study showed that 0.8% patient-days had ≥50 mSv and 0.03% had ≥100 mSv. Additionally, 9.41% patients had more than one CT exam in a single day. The top 20 CT imaging protocols that led to ≥50 mSv in a single day belonged to the body region (chest or abdomen and pelvis), and nearly one-third were angiographic studies. The study concluded that patients with 50+ mSv in a single day or a single exam are not rare. A more recent publication (Rehani et al., 2025) provides information of temporal increase in CT exams with more than 50 mSv dose.
Is cumulative dose important?
Unlike tissue reactions where one can apply gap corrections for the time gap between two consecutive exposures, there are no gap correction factors available for stochastic risks, and thus the only way is to sum up radiation doses from recurrent exams and get the cumulative dose. Future research should develop gap correction factors for repairs occurring between consecutive exposures, and this may be one of the points where research scientists and organisations need to direct research efforts.
ARE PATIENTS GETTING HIGH DOSES SICK WITH SHORT LIFE EXPECTANCY?
With advances in medicine, a large number of clinical conditions are becoming curable, and in many others life expectancy is increasing. For example, Crohn's disease, which entails chronic inflammatory bowel disease and requires almost lifelong CT scans or other imaging resulting in high doses, has a life expectancy of 80 years, which is close to the life expectancy of the general population. Similarly, heart diseases and many cancers such as prostate, testicular, thyroid, breast, and melanoma are mostly curable or lead to decades of life expectancy. The results from a study have shown that nearly half of the patients with ≥50 mSv and nearly one-third of 100+ mSv are living after 10 years (Mataac et al., 2024). Even if one excludes those who may have much reduced possibility to manifest radiation effects during their lifetime, such as those over 60 years of age and those having disease with a shorter life expectancy with a minimal or low probability of manifesting radiation effects, our calculations show that still one is left with about half a million patients getting added globally every year to the 100+ mSv cohort who may have enough life years to manifest radiation effects.
ARE HIGH DOSES OCCURRING DESPITE ADEQUATE IMPLEMENTATION OF JUSTIFICATION AND OPTIMISATION?
This may come as a surprise, but it is true despite the use of the ‘best’ form of justification and optimisation currently available. The author's institution was the first to implement a clinical decision support (CDS) system in radiology around 2004 (Sistrom, 2009). It involves incorporating appropriateness criteria provided by various professional societies into the CDS computer system. The referring physician enters clinical details about the patient in the computer. The CDS system provides both a quantitative and colour classification of appropriateness:
Score 1–3 = red (low utility, ‘usually not appropriate’) Score 4–6 = yellow (marginal, ‘may or may not be appropriate’) Score 7–9 = green (indicated, ‘usually appropriate’)
The CDS database comprises over 3000 clinical scenarios and 15,000 criteria (ACR, 2024a). The system not only incorporates ACR Appropriateness Criteria® but also criteria from the American College of Cardiology (ACC), the National Comprehensive Cancer Network (NCCN), and the Society of Nuclear Medicine and Molecular Imaging (SNMMI). There have been several publications demonstrating the utility of the CDS system in reducing overuse of imaging and enhancing appropriateness. Attempts by many organisations in the past to provide appropriateness or referral criteria/guidelines to doctors and to educate them to use them did not result in the desired results. Thus, the CDS system implementation is the best system possible currently.
ACR had established a dose index registry (ACR-DIR) (ACR, 2024b) that collects data online from a couple of thousand facilities and provides aggregated values of reference dose quantities like CTDIvol, dose–length product (DLP), and size-specific dose estimates (SSDE). It thus regularly provides a national benchmark for various exams and allows facilities to compare their values with national benchmarks. The author's institution has maintained lower dose levels for the most common CT imaging exams. With continuous monitoring of doses, regular meetings of the dose committee to review doses, we are able to achieve optimisation and provide best practices.
Despite the use of the above best approaches towards justification and optimisation, we have thousands of patients with 100+ mSv every year, and the percentage of patients with high CED is higher than many others. Of course, our hospital has the most complicated cases, a large number of clinical trials, and the most active research programme. From a patient radiation safety perspective, it does raise the question of ‘Now what?’ How do we achieve the safety of these thousands of patients every year who are getting high doses despite the use of the best available system in the world for justification and optimisation?
WHERE DO WE GO FROM HERE?
It is clear that there was never a time in history when patients were getting such high doses either as cumulative doses from recurrent exams or in specialised individual exams. As indicated by Brower and Rehani (2021), unsatisfactory image quality was a driving factor for repeat imaging in the past. For the present situation, it is not poor-quality images but higher-than-needed quality or follow-up of disease or new clinical conditions that drive recurrent imaging. Of course, there are possibilities of overuse of imaging and unoptimised imaging, but in a large proportion, it appears that patients do need these exams (Rehani et al., 2020b).
IAEA has been active in pursuing actions on tracking of radiation doses of patients since 2009, starting with its smart card project to joint position statements endorsed by WHO, several international and national organisations, and professional societies (Rehani, 2009, 2017; Rehani and Frush, 2010; ESR et al., 2011; IAEA, 2021; Vassileva and Holmberg, 2021). There have been misinterpretations resulting in a statement steered by AAPM based on fear of misuse of cumulative dose (AAPM et al., 2021).
The suggested way forward could be the following:
Continue to keep the emphasis on the implementation of the principle of justification of ICRP for each imaging exam utilising ionising radiation, weigh the possibility of using non-ionising alternatives without hampering clinical needs, and ensure optimisation by performing the needed exam with minimal dose without affecting the diagnostic value of the intended clinical purpose. Imaging and clinical professional societies should develop appropriateness use criteria (AUC) for serial imaging as criteria till recently have been largely based on individual exams rather than situations where a series of exams is foreseen. ICRP should deliberate on solutions to enhance the radiation safety of patients who are receiving or may receive high doses despite the full use of justification and optimisation principles. It must be kept in mind that we are not dealing with a lack of justification and lack of optimisation here as there are many patients who do need many exams for their care, and we need to provide safer imaging for these patients. The role of industry must be emphasised as it can play the most crucial and most contributory role at this juncture, as has been pointed out explicitly in recent publications (Rehani, 2013; Kachelrieß and Rehani, 2020; Rehani et al., 2020a, 2020b; Brower and Rehani, 2021; Rehani and Brady, 2021). There is a need to utilise feedback available from referring physicians which must be utilised (Rehani and Berris, 2012; Winford and Bharija, 2021; Rehani et al., 2022). There have been myths underplaying radiation risks and overplaying benefits even when they may not be supported by data. Overemphasis on benefits is as harmful as underemphasis. One cannot say that since flying by air has immense benefit one needs to mainly emphasise benefit and not risk. Surgeries are immensely useful, and no one only emphasises benefits of surgery without weighing the risk. The ICRP principle does not imply benefit-only consideration as has tended to be implied by some (AAPM et al., 2021).
