It's 7 PM and the radiology technologist calls me to speak to Mrs. Jones about a chest x-ray of her 3-year-old daughter. I introduce myself and as the little girl looks up at me, she breaks into a racking cough and starts to cry. I touch her forehead as I brush her hair back; she is obviously febrile. I ask her how I can help.
The mother explains that she has been sent by her pediatrician to have a chest x-ray, but she isn't sure that she wants her daughter to have an x-ray. She has read newspaper articles about radiation from computed tomographic (CT) scans. Her husband told her that the government recently released information that fallout from atomic bomb tests may have caused more cancer than had previously been thought. All this has made her wonder whether she might hurt her daughter, rather than help her, by coming to the radiology department.
This concerned mother is certainly not alone. Radiation risk has resurfaced as an active area of research as well as an area of public interest. All health care providers who order or perform imaging studies that use ionizing radiation should be aware of the risks and benefits of ionizing radiation in imaging studies.
For the pediatrician these questions are particularly important. Our patients frequently cannot give consent for their care. Their parents and medical caretakers have the responsibility to speak for the child. Children are also at higher risk from ionizing radiation than are adults. The goal of mis article is to present information needed to assess the risk and benefit of diagnostic imaging using ionizing radiation. We present information regarding the radiation exposure from diagnostic imaging and what we know of the risks of that exposure, and discuss speaking to families and to radiologists about the use of diagnostic imaging in children.
Radiation is energy that is produced by a source and travels through space or some other material. Light and sound are examples of radiation. What is often simply referred to as radiation is better termed ionizing radiation. Ionizing radiation can produce charged particles, or ions, as it passes through matter. The x-rays used in many diagnostic imaging studies are a type of ionizing radiation.
Diagnostic imaging is not the only way that people are exposed to x-rays and other forms of ionizing radiation. Everyone is exposed to ionizing radiation at all times. This exposure is termed background radiation and is denned as the average ionizing radiation exposure from all sources. In the United States the average annual background radiation has been estimated at 3.6 millisieverts (mSv).1 More than half of the background radiation derives from exposure to the radioactive gas radon. Internal radiation, terrestrial radiation, and cosmic rays together account for slightly less than onethird of the exposure. Medical sources account for approximately 15% of the total exposure. Excluding the 15% from medical sources, the average annual exposure to ionizing radiation in the United States is approximately 3 mSv.
The amount of radiation from each of these sources varies in different locations. For this reason, background radiation varies widely in different locations within the United States. In Denver, for example, the combination of high terrestrial radiation from the surrounding mountains and increased cosmic ray exposure at a 5000 foot altitude results in an annual exposure of approximately 6 mSv.1 The variability of background radiation also means that we make choices that affect our exposure to ionizing radiation by such decisions as where we choose to live and whether our home has a basement. A useful way to provide a context for any ionizing radiation exposure is to compare it to background radiation.
We can alter our ionizing radiation exposure in many other ways. A common activity that increases radiation exposure is airline flight. For a recent article in the Wall Street Journal, the radiation doses for several different flights were measured. During a single flight from San Francisco to New York, the passengers were exposed to 0.25 mSv.2 This is the equivalent of 1 month of average background radiation.
Another comparison that may be useful in assessing ionizing radiation exposure is comparison to the regulatory exposure limits for workers who are exposed to ionizing radiation. The maximum allowable ionizing radiation exposure for medical personnel, for example, is 50 mSv annually, with a maximum cumulative dose of 10 mSV multiplied by the worker's age.
IONIZING RADIATION EXPOSURE FROM DIAGNOSTIC IMAGING
The majority of diagnostic imaging studies use x-rays. Examples include chest radiographs, upper gastrointestinal (GI) series, angiography, and CT scanning. The amount of radiation used in these imaging studies varies widely. In order to express the amount of radiation in a useful way we will use the term equivalent dose. The equivalent dose is the amount of ionizing radiation applied evenly to the entire body that would result in the same radiation risk as the radiation exposure from a certain imaging study. This term is useful because it allows a comparison between the radiation risk from very different studies, such as a head CT and a hip radiograph. It also allows comparison between imaging studies and background radiation.
The imaging study with the lowest radiation exposure is the posteroanterior (PA) chest radiograph. The equivalent dose of a single PA chest radiograph is approximately 0.02 mSv. An upper GI produces an equivalent dose in the range of 3 mSv. A CT scan of the chest and abdomen has an equivalent dose of 5 to 10 mSv, and cardiac catheterization produces an equivalent dose of up to 20 mSv.
A major area of concern is the risk of radiation from CT scanning. A study at the University of New Mexico found that 11% of imaging studies performed were CT scans, and that those scans were responsible for 67% of the effective dose to the population from all diagnostic imaging studies. Eleven percent of these CT scans were performed in children.3 During the last decade the indications for CT scanning have increased markedly. Abdominal CT has become a commonly ordered study for a child with suspected appendicitis or renal stones, and CT scanning is the examination of choice for the evaluation of complications of pneumonia. In addition to an increase in the number of CT scans, new advances have made CT scanning one of the fastest changing modalities available. The use of multislice CT scanners that obtain four or more slices at one time allows CT scanning to be both faster and more accurate. These improvements frequently come at the cost of a higher ionizing radiation dose. Deciding the best balance among speed, detail, and radiation dose is an increasingly complex task.
Several recent articles have addressed this aspect of CT scanning in children. Patterson et al. reviewed the CT technique recorded on CT scans sent to the Duke Medical Center Department of Radiology for second opinions.4 The authors reported that the majority of studies did not use techniques appropriate for children. This included factors that resulted in increased radiation dose and factors that decreased diagnostic accuracy. Patterson and his coauthors found that the radiation dose for these examinations was as much as 6 times higher than the dose that would have been used by the radiologists at Duke. In an article by Huda et al., studies performed at a large academic center showed no correlation between patient size and CT dose,5 indicating that the same dose was used for all patients, rather than decreasing the dose for smaller patients. Donnelly et al. described a table that can be used to choose the technique for CT scanning that adjusts the radiation dose depending on the part of the body that is scanned and the patient's size.6 The use of this technique in young children was shown to provide a decrease in radiation dose of up to 75% without loss of diagnostic confidence.
These reports indicate that the radiologist performing pediatric CT scanning requires special knowledge and skill to maximize diagnostic accuracy while limiting ionizing radiation dose.
IONIZING RADIATION RISK
Much of the recent concern regarding radiation risk followed a publication in 2001 by David Brenner et al. from Columbia University.7 Here the authors used new cancer occurrence data from atomic bomb survivors exposed to low-dose ionizing radiation. The ionizing radiation exposure from a single CT scan and the risk of fatal cancer from atomic bomb exposure to similar amounts of ionizing radiation were calculated. These authors suggested that the risk of a fatal cancer from a single CT scan was in the range of 1 in 1000.
There are limitations to Brenner's publication. The ionizing radiation dose from a CT scan that was used in Brenner's calculations was several times higher than the dose currently used in most pediatric centers. More important, the atomic bomb data are unclear regarding the level of risk from an exposure at the level of a single CT scan in a child using appropriate technique. In the most recent study on low-dose atomic bomb exposure8 the authors stated that "the results provided useful risk estimates in the range of 0.05 to 0.1 Sv" (50 to 100 mSv). These authors did not find definite evidence of increased risk at the 5 to 10 mSv level of a single CT scan.
It is also important to understand that the risks reported by Brenner et al. were calculated rather than observed. No studies have evaluated the potential effect of CT scanning on the incidence of cancer. Such a study would be very difficult to perform because the lifetime risk of a fatal cancer is approximately 20%. If one accepts a 1 in 1000 additional risk from one CT scan, the study would need to be able to detect the difference between a cancer rate of .200 and .201. Assessing the increased risk with respect to the known incidence of cancer also provides a context for the risk estimates.
Brenner's is not the only recent article that has suggested that ionizing radiation from diagnostic imaging carries a risk of later cancer. A recent case control study found that the risk of leukemia increased by 50% to 100% in children who had received two or more radiologic procedures of any kind.9 High altitude flight, previously mentioned, is an example of a non-medical activity that carries an ionizing radiation risk. It has been estimated that 10 000 miles of long distance airline travel result in an increase in the cancer risk rate for children of lin 5OW.10 This is very similar to the single CT risk that can be calculated using Brenner's calculations and the lower dose techniques now in use.
The article by Brenner remains an important contribution to the literature. Perhaps most important is that these authors specifically indicate that there is a risk associated with CT scanning, something that has not been addressed directly by most health care practitioners. The authors also point out several additional important facts. In the United States 2.7 million CT scans are performed annually on children younger than 15 years. The frequent use of CT scanning makes CT dose reduction an effective way to decrease the ionizing radiation exposure to the pediatric population. The authors also note that children are at greater risk from ionizing radiation than adults. First, children have a higher lifetime risk of fatal cancer from ionizing radiation than adults. This is because of a longer life during which cancer can develop, and likely also because of increased radiation sensitivity of developing tissues. Second, when the same CT technique is used on children and adults, children receive a higher radiation dose. Thus the risk is further increased. It is reasonable to estimate that if the same CT technique is used on a 4-year-old and a 40-year-old, the 4-year-old will have 10 times the lifetime risk of a fatal cancer.
This information on radiation risk should also be understood in terms of the many risks that are a part of daily life. Nearly every activity carries with it some risk. Occupational risk of death can range from 1 in 100 for a Formula 1 racing driver or deep sea fisherman, to 1 in 1 000 000 for an office worker. The annual risk of dying in an automobile accident is approximately 1 in 5000.
THE ROLE OF COMPUTED TOMOGRAPHIC SCANNING IN PEDIATRIC CARE
The information provided above helps to understand the risk of CT scanning, but it is important to acknowledge the benefits that CT scanning can provide. Computed tomographic scanning is probably the most flexible and helpful diagnostic imaging modality available. If forced to choose only one imaging modality to evaluate all patients, most pediatric imagers would choose CT scanning. In a large radiology department, lifesaving information is obtained from CT scanning nearly every day. Computed tomographic scanning is widely available and easily tolerated. Deciding to use CT scanning as part of pediatric care is a question of risk and benefit. The task for health care providers is to decrease the risk without decreasing the benefits of CT scanning and then to use CT scanning wisely.
Several researchers involved in the study of radiation risk have pointed out the difference between population risk and individual risk. Reducing the risk associated with CT scanning could have an important benefit to the population because of the large number of CT scans performed each year. For an individual child, however, the risk of a CT scan is so much less than the potential benefit that there is no case in which a medically indicated CT scan should not be performed because of concern about the ionizing radiation risk.
THE PEDIATRICIAN'S ROLE
Three specific areas can be identified: ordering studies, ensuring that imaging is performed with attention to minimizing dose, and helping patients and families to understand this aspect of their care.
First, although the current information is not clear regarding the exact level of risk from ionizing radiation, the data suggest that there is a small risk from CT scanning. Computed tomographic scanning should be ordered with attention to both the risk and the benefit. In most cases, the benefit greatly outweighs the risk. In some cases, however, it may not. Computed tomographic scanning for long-term abdominal pain, for example, is very unlikely to provide clinically useful information. A child with suspected appendicitis may be sent directly for CT to speed up diagnostic evaluation prior to surgical referral. Clinical evaluation by a surgeon prior to imaging will lengthen the evaluation process if the surgeon decides to order a CT scan following clinical evaluation. In the appropriate clinical situation, however, the child may be taken to surgery without imaging studies. In this case the CT scan would be unnecessary. The greatest possible decrease in risk occurs when an unnecessary study is not performed. The pediatrician can also decrease dose by identifying the specific area to be examined and the clinical concern. For example, an abnormal area at the apex of the lung can be evaluated by a CT limited to the apices rather than a complete chest CT.
All CT scans should be performed using the lowest dose that allows the radiologist to provide the necessary diagnostic information. As several articles cited above have documented, this is not always done, especially when CT scanning children. Using a "one size fits all" technique for CT scanning can result in a radiation dose 5 to 10 times higher than necessary. Ensuring that an appropriate dose is used is the responsibility of the radiologist, but the pediatrician should ensure that the imaging care for his or her patients is performed with these concerns in mind. The pediatrician should address these questions directly with the radiologists to whom he or she refers patients.
The pediatrician should inquire if pediatric-specific protocols are used when children are scanned. Computed tomographic scans of the chest should be performed with a lower dose technique than CT scans of the abdomen. Nearly all CT scans should be performed either with or without the use of intravenous contrast, but not both. Elmrinating one of these sets of images decreases radiation dose by 50%. Pediatricians should make it clear that attention to radiation dose is an important consideration when referring patients to a radiology group. An additional benefit of these discussions can be the development of a greater consultant role for the radiologist, with opportunities to suggest other clinical or imaging options when appropriate.
Discussing ionizing radiation risk with patients and families can be an important part of the imaging care of pediatric patients. These discussions require an understanding of the available information and a balanced approach to the risks and benefits of performing the diagnostic imaging study.
Some of the information provided above may be helpful to families. Many people are unaware that ionizing radiation exposure occurs throughout life and that man-made sources provide a small part of that exposure. Expressing radiation dose in terms of background radiation or airplane trips may be helpful. In our experience, most families do not seek detailed information. Rather, they need to be assured mat their physicians are aware of these concerns and have made every effort to limit ionizing radiation exposure.
Mrs. Jones asked good questions. How much risk is there to this chest x-ray? Why do we need to do a chest x-ray at all? Is there another test that would give us the same information with less risk?
After explaining that her daughter's chest xray would give about the same amount of radiation that she gets every 3 days of her life, and that depending on the results of the x-ray, different antibiotics might be required, Mrs. Jones agreed to proceed with the radiograph. The radiograph revealed a left pleural effusion and consolidation of the left lower lobe and lingula. Her daughter was admitted for intravenous antibiotics and rapidly improved.
1. Imbergamo PJ, Janower MS, and Murray L. Frequently asked questions. In: Radiation Risk: A Primer. American College of Radiology. 1996: vi.
2. Drucker J. Radiation in the skies. In: Wall Street Journal. March 3, 2002; 239:10.
3. Mettler FA Jr, Wiest PW, Locken JA, Kelsey CA. CT scanning: patterns of use and dose. / Radiol Prot. 2000; 20:353-359.
4. Paterson A, Frush DP, and Donnelly LF. Helical CT of the body: are settings adjusted for pediatric patients? AJR Am J Roentgenol, 2001;176: 297-301.
5. Huda W, Scalzetti EM, and Roskopf M. Effective doses to patients undergoing thoracic computed tomography examinations. Med Phys. 2000; 27:838-844.
6. Donnelly LF, Emery KH, Brody AS, et al. Minimizing radiation dose for pediatric body applications of single-detector helical CT: strategies at a large children's hospital. AJR Am J Roentgenol. 2001; 176:303-306.
7. Brenner DJ, Elliston C, Hall E, Berdon W. Estimated risks of radiation-induced fatal cancer from pediatric CT. AJR Am J Roentgenol. 2001; 176:289-296.
8. Pierce DA, Preston DL. Radiation-related cancer risks at low doses among atomic bomb survivors. Radiât Res. 2000; 154:178-186.
9. Infante-Rivard C, Mathonnet G, and Sinnett D. Risk of childhood leukemia associated with diagnostic irradiation and polymorphisms in DNA repair genes. Environ Health Perspect. 2000; 108:495-498.
10. National Council on Radiation Protection and Measurement. Radiation exposure and high-altitude flight. Bethesda, Maryland; 1995:5-12.