INTRODUCTION — Ionizing radiation from medical imaging now accounts for nearly one-half of the radiation exposure experienced by the population in the United States [1,2]. Medical imaging may contribute less to total radiation exposure in other countries, however; 2005 data from the United Kingdom, for example, suggest that an average annual dose of imaging accounted for approximately one-sixth of annual exposure to ionizing radiation . Based on extrapolation models from nuclear accident and atomic bomb survivors, one study estimates that use of computed tomography (CT) may account for 1.5 to 2 percent of all future cancers in the United States .
This topic will present an overview of the measurement of radiation associated with medical imaging, the effects of radiation exposure, the radiation associated with specific diagnostic imaging studies, and issues in clinical decision-making, including informed consent. Other related topics, including discussions of radiation injury, diagnostic imaging in pregnancy, and radiation exposure in cardiovascular imaging, are presented separately.
INCREASING USE OF IMAGING STUDIES — The average radiation dose per United States resident per year has been continuously increasing over the past 30 years . Among participants in managed care plans in the United States, for whom there would be no financial incentive to promote excess testing, increased use of computed tomography (CT) resulted in a doubling of per capita radiation exposure between 1996 and 2010 . Although the immediate and long-term benefits of medical imaging are widely discussed, their risks are often overlooked. The radiobiological detriment from a single examination is likely very small. However, the cumulative effect of small individual doses in a patient population undergoing frequent and often repeated imaging may present a public health concern for risks such as radiation-induced malignancies.
The majority of radiation from medical imaging arises from CT. Use of CT has grown exponentially over the last several decades. Almost three-quarters of all CT examinations in the United States were ordered in acute care settings; in one large urban hospital, the overall rate of CT examinations ordered per patient doubled between 2001 and 2007, with the rate of chest CT examinations increasing sixfold . CT rates continue to rise; among older adults, 428 CT examinations per 1000 person-years were ordered in 2016 versus 204 CT examinations per 1000 person-years in 2000 .
CT is increasingly used to diagnose or exclude many life-threatening conditions. In the acute care setting, the most common indications for CT are trauma, stroke, and suspected pulmonary embolism. In the nonacute setting, the majority of CT imaging is in patients with cancer. In such settings, the short-term risk of a patient dying from the underlying illness far exceeds their lifetime risk of radiation-induced cancer .
IONIZING RADIATION — Ionizing radiation involves the detachment of electrons from subatomic particles or electromagnetic waves (eg, photons) at the atomic or molecular level, thus forming ions. Ionizing radiation is found in the natural environment in the form of radon or cosmic rays. Ionizing radiation is used for interventional and diagnostic radiology procedures, and, in higher doses, for therapeutic and palliative procedures in the context of radiation oncology. With large catastrophic exposures, such as from an accident at a nuclear industrial facility, skin damage and acute radiation sickness may result. (See "Clinical manifestations, evaluation, and diagnosis of acute radiation exposure".)
At the lower doses used in diagnostic and interventional procedures, ionizing radiation may cause DNA damage that increases the risk for future cancer . It should be noted that ionizing radiation causes structural damage at the cellular or molecular level, while nonionizing radiation (which is not associated with medical imaging) causes damage through the transfer of heat (eg, microwave heating). Relative sources of exposure to ionizing radiation in the United States are shown in a figure (figure 1).
Biologic effects — The proposed mechanism for cellular damage by ionizing radiation involves the production of free radicals. These free radicals interfere with chemical bonds between molecules that regulate key cellular processes . Direct interaction between ionizing radiation and cellular macromolecules leads to DNA mutation or cell death, while indirect interaction causes free radical damage to essential cell enzymes .
Tissues are more or less susceptible to radiation injury, based on their rate of cellular proliferation and their degree of cellular differentiation. Thus, blood-forming (lymphoproliferative) organs with rapid cell turnover are the most sensitive, while nervous tissue with little or no cell turnover is the least sensitive. Relative radiosensitivities of different organs and tissues are listed in a table (table 1).
The probability of occurrence of the chronic effects of ionizing radiation is a function of the total radiation dose, although the severity of effects cannot solely be attributed to radiation and is influenced by other factors (eg, environmental, genetic). While induction of future cancer is the effect of greatest concern, other potential outcomes include shortening of lifespan and cataract formation in the eyes . Radiation-induced cancers may develop decades after exposure and include myeloma, leukemia, lung cancer, thyroid cancer, breast cancer, bone cancer, and skin cancer [12,13].
Dose-response model — Several theoretical dose-response models correlate radiation exposure with risk of developing cancer. There is controversy regarding the reliability of these models to predict cancer risk and there is a paucity of data to support or refute the use of one model over another. Of these, the linear no-threshold (LNT) model is the one most widely used. The LNT model provides a conservative risk estimate based on the assumption that any exposure to ionizing radiation, however small, can induce future cancer. It adheres to the as low as reasonably achievable (ALARA) principle of radiation safety designed to minimize radiation doses and release of radioactive materials [9,14-16].
The LNT model is based on extrapolation of data from survivors of the Hiroshima atomic bomb . These data are used to project the cancer risk from estimates of very small radiation doses (eg, equivalent to a single chest radiograph). Other studies suggest that a meaningful threshold is not reached by very low-dose radiologic studies, or that very low doses may even have a beneficial adaptive effect, fostering DNA repair and immune stimulation . The United States National Council on Radiation Protection and Measurements (NCRP) and the United States National Academy of Sciences Biological Effects of Ionizing Radiation (BEIR) committees suggest that LNT is the best model for estimating the possible risks of ionizing radiation, while recognizing that it may err on the side of caution [9,14,15].
MEASURING RADIATION DOSE
Terminology — Several terms are used to describe different aspects of radiation and radiation dose [19,20].
●Radiation exposure refers to the concentration of ionizing radiation in a specific volume of air, measured in roentgens (R).
●Absorbed dose signifies how much radiation is absorbed by a specific tissue, measured in grays (Gy) or rads. Dose area product (DAP) or kerma area product (KAP) is the absorbed dose multiplied by the area radiated, and it is measured in gray square centimeters (Gy cm2).
●Effective dose takes into account the particular tissue or organ that absorbs the radiation and reflects the equivalent whole-body dose that would result in the equivalent risk from a nonuniform source of irradiation such as computed tomography (CT). It is the weighted average of doses to the irradiated organs and is measured in sieverts (Sv) or rems (100 rem = 1 Sv = 1000 mSv).
Effective dose, when applied using the linear no-threshold (LNT) model, can provide an estimate of the potential harms caused by low doses of radiation involved in diagnostic imaging. Projections on carcinogenesis, decreased life expectancy, and hereditary effects from ionizing radiation are derived at the population level.
Effective dose — Effective dose is the method most commonly used in clinical practice and the medical literature to estimate a patient's risk of detriment from radiologic imaging. This detriment includes all cancers induced by exposure. Calculating the effective dose allows for comparison across different imaging modalities and spatial dose distributions. Additionally, the effective dose can be a useful parameter to weigh the potential benefits of the radiologic examination against the possible adverse effects of ionizing radiation.
The effective dose related to any radiologic examination or procedure can be considered in the context of the annual effective dose to individuals from natural background radiation sources, including radon, cosmic rays, terrestrial, and internal sources. This background effective dose is approximately 3 mSv. Exposure to an individual dose of 50 mSv or a lifetime dose of 100 mSv has not been associated with health risks; as such, these have been adopted as guidelines for upper thresholds for diagnostic imaging in clinical practice . The International Commission on Radiological Protection (ICRP) estimates a 4 to 5 percent increased relative risk of fatal cancer after an average person receives a whole-body radiation dose of 1 Sv (1 Sv = 1000 mSv) .
A retrospective study of claims data for 2005 to 2008 from an insurance database in the United States for approximately 950,000 enrollees aged 18 to 64 years found that nearly 70 percent of enrollees underwent at least one imaging examination . The median effective dose from imaging per year was 0.1 mSv; moderate, high, and very high effective doses (>3 to 20 mSv, >20 to 50 mSv, and >50 mSv, respectively) were incurred by 194, 19, and 2 enrollees per 1000 per year, respectively.
ESTIMATING RISK FROM RADIATION EXPOSURE — No epidemiologic studies document a direct link between exposure to imaging examinations or image-guided procedures and cancer. A deterministic effect between imaging, such as a diagnostic computed tomography (CT) examination, and cancer will undoubtedly be difficult to show.
Epidemiologic basis for risk estimates — Risk estimates of radiation-induced cancer from medical imaging are mainly based on extrapolation of existing data from Japanese survivors of atomic bombings that took place in 1945 [23,24]. Radiation doses to more than 100,000 individuals have been estimated based on the individuals' distance from atomic bomb epicenters; exposed individuals have been followed for cancer incidence and mortality for more than 60 years. Analysis of this patient registry demonstrates a linear relationship between the risk of developing any solid cancer and increasing radiation dose. However, this linear relationship is not observed when extrapolated below 100 mSv, in the range of doses conferred by diagnostic imaging . Patient radiosensitivity is heavily dependent on age at the time of exposure, with children at greater risk of developing future radiation-induced cancer than adults.
Some experts believe that direct comparisons of radiation dose from CT examinations and atomic bomb survivors can be made, without the need for extrapolation of existing data, since roughly one-third of the atomic bomb survivors studied were exposed to low radiation doses in the range of single or multiple CT examinations . This subpopulation of survivors had a small, but statistically significant, increased risk of developing future cancer after exposures in the range analogous to radiation associated with a single abdominal CT examination. However, for very low-dose radiologic examinations, such as conventional radiography and mammography, the linear no-threshold (LNT) model is still required to estimate possible cancer risks, and the actual risk associated with low-dose medical imaging remains controversial.
While there are no data showing a causal relationship between actual medical examinations such as CT and cancer, there is epidemiologic evidence of future cancer development from exposure to x-rays and gamma rays. In the 1930s through the 1950s, repeat fluoroscopic examinations were standard of care in tuberculosis patients, used to follow therapeutic pneumothoraces. The radiation doses for these fluoroscopic studies were slightly higher than current CT doses, and these patients were found to have an increased risk of developing breast cancer . In another study involving more than 400,000 nuclear industry workers in 15 nations, an excess relative risk for cancer deaths was identified, equivalent in magnitude to that found in Japanese atomic bomb survivors .
Future cancer risk — Estimating the risk of developing cancer from an individual imaging examination is controversial, given the stochastic nature of harm from ionizing radiation and the limited epidemiologic data. However, there is mounting consensus in the medical and scientific communities that the risk is real, however small to the individual.
To estimate the lifetime risk of developing cancer attributable to any single imaging examination for an individual patient, the American Society of Radiologic Technologists has an online radiation risk calculator. Cumulative radiation dose and cancer risk from several studies over time can also be calculated. The University of California at San Diego also has a radiation risk statement calculator that creates a standard statement about the effective dose a patient will receive as part of their imaging examination.
The 2006 Biological Effects of Ionizing Radiation (BEIR) VII lifetime attributable cancer risk model predicts that 1 in 1000 persons exposed to 10 mSv (which is in the range of a single diagnostic CT of the neck, chest, abdomen, or pelvis) will develop cancer due to that single exposure .
The International Commission on Radiological Protection (ICRP) has confirmed that doses with repeated diagnostic examinations, such as CT, could reach levels associated with an increase in lifetime cancer risk .
●One study estimated that roughly 29,000 future cancers could be attributed to CT in the United States in 2007 alone .
●Another study reports that 1 in 500 women and 1 in 660 men will develop cancer from their abdominal CT if the examination is performed at the age of 20 .
●In the case of elective screening, one study estimated that a 45-year-old adult undergoing one single full-body CT examination would accrue an additional lifetime attributable cancer mortality risk of 0.08 percent from a single examination; the lifetime attributable cancer mortality risk is 1.9 percent per lifetime for 30 years of annual examinations .
Estimated effective doses for the most common imaging examinations and procedures are provided below. (See 'Radiation dose for common imaging examinations and procedures' below.)
IMAGING MODALITIES WITH IONIZING RADIATION — Most of the ionizing radiation from medical imaging originates either from x-ray (computed tomography [CT], fluoroscopy, and radiography or plain film) or from nuclear medicine examinations involving radioactive tracers. Fluoroscopy is used in interventional procedures (eg, cardiac catheterization, neuroangiography) and diagnostic examinations (eg, gastrointestinal tract examination, retrograde urethrography). Nuclear medicine examinations (eg, technetium-99m methylene diphosphonate bone scan, fluorodeoxyglucose positron emission tomography [PET]) involve introducing a radioactive tracer into the body, which then emits gamma rays as it undergoes nuclear decay.
CT examinations result in radiation doses that are 100 to 250 times larger than those from conventional plain radiographs and represent the fastest growing component of the United States population's collective dose. The radiation dose attributable to fluoroscopic procedures, including interventional cardiology procedures, is discussed elsewhere. (See "Radiation dose and risk of malignancy from cardiovascular imaging".)
Radiation dose for common imaging examinations and procedures — The average effective doses of the most common diagnostic examinations and interventional procedures in the United States for the adult population have been calculated . The following tables are adapted from this summary report and other published summary reports [31,32,34,35]. It should be noted that these are average effective doses for adults and that actual dose for any one patient may differ from the average effective dose reported by up to 10-fold (table 2 and table 3 and table 4 and table 5 and table 6).
CT examinations — The use of CT has increased more than threefold in the United States since 1993 [6,7,30] and accounts for the majority of radiation from medical imaging. It is difficult to accurately estimate individual organ effective doses from a CT examination due to the unique physical characteristics of each patient, including their weight, age, and sex. Larger patients may require increased dose to penetrate additional body mass to generate adequate images, while younger patients will have more years of life left in which they may develop radiation-induced cancer. Moreover, since radiosensitivity differs by tissue type, the dose to a female's breasts or ovaries will differ from a male's dose to the testes even with the same magnitude of exposure.
A number of technical factors also influence radiation dose associated with CT, including the x-ray beam energy, tube current, tube voltage, exposure time, patient thickness, and scanning pitch (overlap of adjacent CT slices). Multiphasic CT examinations, in which the same organ is scanned in different phases of contrast enhancement to improve lesion detection and characterization, can increase the radiation dose up to fourfold .
To help develop a general estimate of effective dose for common CT examinations, researchers developed standard patient models using Monte Carlo simulations that allow estimated organ doses to be combined with appropriate weighting factors. These standard patient models result in estimated effective doses for each type of examination for the average adult patient . Software programs are available for gauging CT radiation dose after inputting certain technical parameters. Two estimations commonly used to quantify radiation dose, the volume CT dose index and the dose length product (DLP), are available on all modern CT scanners. These parameters cannot be directly converted into an estimate of effective dose but can gauge the difference in dose a patient will receive when altering specific technical scanning parameters (eg, pitch) . While frequently referred to in the context of modifying protocols to decrease dose to the patient, these two figures do not describe the actual radiation dose to the patient.
One analysis used risk models from the Biological Effects of Ionizing Radiation (BEIR) report combined with age- and sex-specific CT examination frequencies in the United States in 2007 and used Monte Carlo simulations to project incidence of CT radiation-related cancers with assumptions on dose of individual examinations . Their analysis estimated that approximately 29,000 (95% CI 15,000-45,000) future cancers could be attributable to CT in the United States in 2007, with distribution by imaging type as follows:
●Abdominal and pelvic CT – 14,000 (95% CI 6900-25,000)
●Chest CT – 4100 (95% CI 1900-8100)
●Head CT – 4000 (95% CI 1100-8700)
●Chest CT angiography – 2700 (95% CI 1300-5000)
In 2011, the National Academy of Medicine (then the Institute of Medicine) identified CT as one of the two major environmental causes of breast cancer (the other being postmenopausal hormone therapy) and recommends avoidance of medically unnecessary CT imaging as one of the most important steps women can take to reduce their risk of breast cancer . Some experts believe that 30 percent or more of advanced imaging studies ordered in the United States may be unnecessary .
EFFORTS TO REDUCE THE RISK OF RADIATION FROM IMAGING — Three guiding principles of radiation protection are justification, optimization, and limitation .
●Justification refers to an appropriate indication for an imaging examination or procedure.
●Optimization, or the as low as reasonably achievable (the ALARA) principle, recommends that protocols be designed to obtain diagnostic images at the lowest possible dose to the patient and is referred to widely .
●Limitation refers to setting maximum aggregate doses over a given period of time. Traditionally, this has been focused on occupationally exposed individuals.
Several initiatives have been undertaken to implement these guiding principles.
●The American College of Radiology (ACR) has developed appropriateness criteria to provide guidance to health care providers in choosing radiologic examinations for specific clinical scenarios. The intent is to minimize the number of inappropriate examinations . Appropriateness guidelines have also been developed by the Royal College of Radiologists  and are in development by the European Commission .
The ACR guidelines are being incorporated into computerized physician order entry (CPOE) and decision support systems as part of the Protecting Access to Medicare Act of 2014, with decision support required for Medicare reimbursement at a future date . While early studies demonstrated a substantial decrease in the utilization of low-yield computed tomography (CT) studies with incorporation of CPOE and decision support systems [44-46], a subsequent report from the Medicare Imaging Demonstration project noted that most orders for advanced imaging were unable to be matched to specific appropriateness criteria . Consequently, entities that reimburse for imaging, both private and public (eg, Centers for Medicare and Medicaid Services), are implementing policies that require or financially incentivize use of CPOE or other decision support mechanisms for image ordering.
●Many researchers have focused on optimizing CT protocols to ensure that the ALARA principle is followed, with a particular focus on pediatric patients , especially those undergoing frequent diagnostic and surveillance imaging [49-51], and on new screening techniques that involve asymptomatic adult patients . Dose auditing and collaborative efforts to share best practices on CT radiation doses across institutions have led to further reductions in radiation doses for common CT procedures .
●There is renewed interest in documenting individual radiation doses and setting maximum aggregate exposures systematically in the United States, with legislative initiatives called for . California has passed a law that requires radiation dose information to be documented in the patient's medical record . Education campaigns about the radiation risks from medical imaging geared towards patients and clinicians are in development by large organizations, including the US Food and Drug Administration .
●In the United Kingdom, the Committee on Medical Aspects of Radiation in the Environment (COMARE) issued a report on personally initiated CT examinations for asymptomatic individuals, recommending regulation of CT services and that screening take place as part of a coordinated care program .
●"Image Wisely," a 2011 campaign to promote safe imaging, and "Choosing Wisely," a campaign to promote wise use of imaging resources and create lists of medical practices that are overused, were launched by multiple partner organizations, including the ACR, the Radiological Society of North America, the American Association of Physicists in Medicine, and the American Society of Radiologic Technologists [57,58]. United States medical organizations, working with Choosing Wisely, have identified a list of imaging examinations that were felt to be frequently overused. These are listed on the Choosing Wisely website .
●The Radiological Society of North America and the ACR have jointly developed a website, RadiologyInfo.org, to provide patients with information on radiation doses and health effects from imaging involving ionizing radiation .
Shielding — A small number of studies have examined the benefits of shielding to reduce radiation to the female breast, male gonads, and thyroid. Depending on the body part to be imaged, the benefits of shielding are offset by beam-hardening artifacts that degrade image quality and diagnostic accuracy. Shielding of these radiosensitive organs may substantially decrease associated radiation-related risks, especially in children and young adults. For CT, in-plane bismuth shields selectively lower dose of the underlying superficial tissue while allowing sufficient beam energy to pass through to yield a diagnostic image [61,62]. Commercially available breast shields reduce female breast dose by 40 to 61 percent [61-64], testicle capsules can decrease dose to male gonads by 87 to 96 percent [65,66], and thyroid shields have been found to reduce radiation dose to the thyroid by 31 to 57 percent [61,64].
CLINICAL DECISION-MAKING AND INFORMING PATIENTS — Over the past several decades, a gradual lowering of the threshold for obtaining diagnostic radiologic imaging in the United States has been observed and attributed to a number of reasons, including improvements in diagnostic power of the technology, medical-legal concerns, patient demand, financial incentives, knowledge gap of the ordering clinician, and inaccessibility or lack of knowledge about previously performed examinations [67,68]. Patients and clinicians are largely unaware of the potential risks associated with medical imaging and that there is little consensus regarding who should provide informed consent to patients and in what clinical scenarios [69,70]. In most instances, the benefits of information gained from diagnostic imaging clearly outweigh the risks. However, for patients undergoing studies for inappropriate indications, frequent repeat studies for the same indication, or elective screening examinations, the future potential risks may outweigh the benefits.
Choice of imaging examination — The preferred imaging examination for a given clinical scenario varies with each practice setting. Important considerations in modality choice are diagnostic performance, availability of the technology and radiologist expertise, and safety considerations including radiation dose. The choice of when to image and with what modality is often discussed within the specific UpToDate topics. In addition, the American College of Radiology (ACR) appropriateness criteria provide general guidance for many common clinical scenarios including screening . When the decision is not obvious, consultation with the radiologist is helpful to facilitate patient referral.
Estimating examination dose — While estimating the radiation dose for any specific patient and the theoretical risks of future cancer associated with their radiologic examination is difficult, communicating with patients about these risks may be even more difficult. Estimated effective dose for a patient undergoing a particular radiologic examination or procedure can be calculated, using the dose length product (DLP), which is recorded as part of each computed tomography (CT) examination, combined with details of the imaged area and published conversion factors . For the purposes of patient-clinician communication, however, relative magnitudes as outlined in the tables are probably sufficient (table 2 and table 3 and table 5 and table 6).
Patient counseling — When referring a patient for imaging, health care providers should counsel the patient regarding its potential risks and benefits. The need for informed consent for the radiation-related risks of imaging studies remains controversial. Some experts believe that a formal risk-benefit discussion regarding radiation risks is not possible given the current state of scientific evidence and the uncertainty surrounding the actual magnitude of radiation-induced cancer risks . Many patients who undergo imaging studies are not informed of the radiation risks. A report in 2004 found that over 90 percent of patients undergoing CT were not informed of radiation risks prior to the examination . As of 2005, at least two-thirds of all United States academic medical centers had institutional guidelines addressing who should undergo informed consent for CT . While only 15 percent of centers explicitly informed patients about radiation risks in 2005, this number has likely increased due to increased patient awareness and widely publicized incidences of CT radiation overexposure for CT brain perfusion studies to rule out stroke . Moreover, as public concern increases regarding radiation associated with medical imaging, clinicians will likely be called upon to discuss the possible radiation-induced cancer risks associated with routine diagnostic imaging studies with their patients.
We present one approach for discussing potential radiation-induced cancer risks with patients:
●It should be made clear to the patient that their actual risk of developing cancer from any one or multiple radiologic examinations cannot be calculated exactly and that any resulting adverse effect is theoretical at diagnostic radiation doses.
●It should also be specified that representative effective doses for each examination, as detailed in the tables (table 2 and table 3 and table 5 and table 6), may be off by an order of magnitude (factor of 10) for any given patient and vary widely by institutional protocols, equipment specifications, age, sex, and weight .
●The patient should be given a general idea about the magnitude of the proposed radiation dose in terms that they can understand. It may be helpful to compare the estimated effective dose with a single chest radiograph (0.02 mSv), a transcontinental airplane flight (0.02 mSv), or annual individual radiation dose from the natural background (3.0 mSv). For instance, the dose from a single abdominal CT examination is comparable to the collective dose of natural background radiation for three years.
●The possibility of future development of cancer must be weighed against the possible benefits, given the clinical scenario and suspected diagnosis. Alternatives to the imaging examination involving ionizing radiation, if any, must be reviewed during the discussion.
The question of who should obtain consent is not readily resolved . The clinician requesting the test has the best information about the patient's history, need for the examination, and prior radiation history, while the radiologist has the most knowledge about specific radiation risks. However, discussing consent and alternative management strategies is likely most appropriate prior to scheduling the examination, especially when special preparations (eg, bowel preparations for CT colonography) are required.
Screening studies — Screening examinations have been developed, including coronary CT screening, CT lung cancer screening, and CT colonography [76-78]. The small, increased risks of radiation-induced cancer should be weighed against the potential benefits of screening examinations on an individual basis. Therefore, a full discussion of risks, benefits, and alternatives of imaging-based screening between clinicians and patients is important [78,79].
SPECIAL POPULATIONS — There are specific patient populations for whom ionizing radiation should be avoided if at all possible.
Children and adolescents — The risks of exposure to ionizing radiation are higher in pediatric patients than adults; such exposure should be avoided if possible, and alternative diagnostic imaging modalities (eg, ultrasound and magnetic resonance imaging [MRI]) considered if appropriate. Radiosensitivity is increased in pediatric patients, in particular involving tissues such as the thyroid, gonads, and bone marrow. In addition, children's smaller size accounts for a greater radiation dose compared with that imparted to adults . One study estimated that radiation dose from a single computed tomography (CT) examination in a one-year-old child is an order of magnitude higher than for adults .
Radiation risk in children is further compounded with their longer lifespan following exposure so that there is a longer time over which radiation-induced cancers can occur . Due to this increased sensitivity to radiation, lifetime radiation risk from just one CT examination in a pediatric patient is an important factor in the individual risk-benefit assessment. A 2021 meta-analysis found that CT exposure in childhood is associated with increased cancer risk, but no associated risk was found for radiographs .
Estimates of the risk of future malignancy in pediatric patients undergoing CT vary. One study assessed the risks of developing a fatal cancer from CT and estimated the lifetime cancer mortality risk attributable to a single radiation exposure in a one-year-old child to be 1 in 550 following an abdominal CT and 1 in 1500 following a brain CT . These findings equate to a 0.35 percent increase in cancer deaths over the present background rate. A 2014 systematic review estimated that an absorbed dose of 10 mGy (eg, one CT angiogram of the head and neck) at less than 10 years of age leads to one additional brain tumor per 10,000 patients .
The risks of ionizing radiation in children are related to the cumulative dose received, as well as the radiosensitivity of the tissues exposed. As examples:
●In a cohort study of approximately 175,000 individuals in the United Kingdom who underwent CT when they were younger than 22 years of age, the risk of developing leukemia was tripled for those who received a cumulative radiation dose ≥30 mGy compared with those who received a cumulative radiation dose <5 mGy; the risk of developing a brain tumor nearly tripled for those who received a cumulative radiation dose ≥50 mGy .
●A cohort study including approximately 825,000 South Korean patients undergoing appendectomy for appendicitis compared those who had at least one perioperative CT scan with those without a CT scan, and then followed all patients for 8.2 years for the development of subsequent malignancy . Exposure to abdominopelvic CT scan (estimated red bone marrow radiation dose 14.7 mGy) was associated with an increased risk of subsequent hematologic malignancy, primarily leukemia (incidence rate ratio [IRR] 1.26; 95% CI, 1.09-1.45); this increased risk was most evident among children age ≤15 years (IRR 2.14; 95% CI, 1.35-3.40).
●In the European EPI-CT study, which evaluated 650 000 patients without a history of cancer or benign brain tumors who received a head or neck CT before age 22 years, a dose-response relationship between CT-related radiation exposure and brain cancer was observed . The mean cumulative brain dose was 47.4 mGy (SD 60.9) among all patients, and 76.0 mGy (100.1) among those with brain cancer. The authors calculated that this risk would translate to one radiation-induced brain cancer for every 10,000 people receiving a single head CT examination (using an average brain dose of 38 mGy) in the 5 to 15 years after examination.
Pregnant women — The maternal and fetal health effects of ionizing radiation in pregnancy are discussed in detail separately. (See "Diagnostic imaging in pregnant and lactating patients", section on 'Radiography and computed tomography' and "Diagnostic imaging in pregnant and lactating patients", section on 'Nuclear medicine'.)
AIRPORT PASSENGER SCREENING — There has been increasing concern regarding radiation exposure from whole-body transmission or backscatter radiograph scanners at airport security checkpoints that have been deployed in airports across the United States. Reports calculate the dose emitted from backscatter radiograph scanners as being equivalent to approximately three to nine minutes of radiation received from the environment through normal daily living [88,89]. Put another way, among 100 million annual passengers, only six cancers may be attributed to airport radiograph screening over the lifetime of these individuals. Thus, evidence suggests that there is negligible individual risk from airport passenger screening . Millimeter-wave scanning is a similar technology without the use of ionizing radiation, which may become a comparable technology and preferable alternative to radiograph airport passenger screeners [90,91].
SUMMARY AND RECOMMENDATIONS
●Ionizing radiation from medical imaging now accounts for nearly one-half of the radiation exposure experienced by the population in the United States. The cumulative effect of small individual doses in a patient population undergoing frequent and often repeat imaging may present a new public health concern for risks such as radiation-induced malignancies. (See 'Introduction' above.)
●At doses used in diagnostic and interventional procedures, ionizing radiation releases free radicals that may cause DNA damage. Tissues are variably susceptible to radiation injury, based on their rate of cellular proliferation and their degree of cellular differentiation (table 1). Radiation-induced cancers may develop decades after exposure and include myeloma, leukemia, lung cancer, thyroid cancer, breast cancer, bone cancer, and skin cancer. (See 'Ionizing radiation' above.)
●The linear no-threshold (LNT) model assumes that any exposure to ionizing radiation, however small, can induce future cancer. (See 'Dose-response model' above.)
●"Effective dose" takes into account the particular tissue or organ that absorbs the radiation and reflects the equivalent whole-body dose that would result in the equivalent risk from a nonuniform source of radiation such as computed tomography (CT). Effective dose is measured in sieverts (Sv) or millisieverts (mSv) and allows comparison of exposure across different imaging modalities.
Annual effective dose is approximately 3 mSv from the natural background. Exposure to an individual dose of 50 mSv or a lifetime dose of 100 mSv has not been associated with increased health risks; as such, these have been adopted as guideline upper thresholds for diagnostic imaging in clinical practice. (See 'Effective dose' above.)
●Few epidemiologic studies document cancer arising directly from radiologic medical examinations or procedures; most risk estimates are based on extrapolations from individuals exposed to radiation following the 1945 atomic bombing in Japan. Estimating the risk of developing cancer from an individual radiologic examination is controversial, given the stochastic nature of harm from ionizing radiation and the limited epidemiologic data. However, it is possible that doses achieved with repeated diagnostic CT examinations could reach levels associated with an increase in lifetime cancer risk. (See 'Future cancer risk' above.)
●The average effective doses for the most common diagnostic examinations and interventional procedures in the United States for the adult population have been calculated (table 2 and table 3 and table 5 and table 6). CT examinations result in radiation doses that are 100 to 250 times larger than those from conventional plain radiographs and represent the fastest-growing component of the United States population's collective dose. (See 'Radiation dose for common imaging examinations and procedures' above.)
●Children and adolescents are at higher long-term risk of malignancy from exposure to ionizing radiation. When appropriate, alternative diagnostic imaging modalities, such as ultrasound or magnetic resonance imaging (MRI) should be considered in this population. (See 'Children and adolescents' above.)
●Three guiding principles of radiation protection are justification, optimization, and limitation. When referring a patient for imaging, health care providers should counsel the patient regarding its potential risks and benefits. Discussion about the possible risks of ionizing radiation should be part of the discussion with modalities that involve x-ray or nuclear medicine tracers. (See 'Clinical decision-making and informing patients' above.)
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟