Practical Strategies to Reduce Pediatric CT Radiation Dose Thomas R. Nelson, PhD
The objective of this article is to provide a brief review of CT scanning radiation sensitivity in children and explain CT scan parameters that affect radiation dose. We discuss key factors influencing radiation dose and study quality and how these factors can be used to optimize scan protocols with the goal of reducing pediatric CT radiation dose without compromising diagnostic quality. Finally, we provide some practical tips for reducing radiation doses to children. Key Words: Pediatric, CT, radiation, dose, dose reduction, automatic exposure control, protocol optimization J Am Coll Radiol 2014;11:292-299. Copyright © 2014 American College of Radiology
INTRODUCTION
New CT technology has facilitated more rapid and sophisticated patient scanning. The potential for increased patient radiation dose resulting from this technology makes it essential to ensure that each CT scan performed is both medically indicated and uses optimized scanning protocols. Children are not small adults and have quite different imaging needs. Their cells are dividing in different ways and are more radiation sensitive, and their body proportions are different. Radiation’s effects on children’s cells also are cumulative and radiation-related risk increases with each dose. Thus, it is important to optimize scan protocols to the individual to maximize diagnostic information and minimize radiation dose. There are a number of strategies that are available to help patients, technologists, and physicians understand and optimize pediatric CT scanning. Although the radiation dose from a single CT scan is relatively low, when adult-sized imaging techniques are used, children receive a higher radiation dose. Because children’s cells are more susceptible to the effects of radiation, there is a slightly increased risk for developing cancer from stochastic, or nondeterministic, effects. Also, there may be changes to children’s genes from radiation doses received over the course of their lifetimes with the possibility that genetic effects may be passed on to their offspring. Deterministic effects also can occur above certain dose thresholds and typically are the most readily visible effects that we identify, such as hair loss at high doses. Some studies have suggested that pediatric
Department of Radiology, University of California, San Diego, La Jolla, California. Corresponding author and reprints: Thomas R. Nelson, PhD, Department of Radiology, M-0610, University of California, San Diego, La Jolla, CA, 92093-0610; e-mail: [email protected].
292
CT scan doses are higher than necessary, so we need to be particularly vigilant in how we optimize scanning protocols to minimize dose. Defining CT scan radiation risk is complex for several reasons. First, risk assessment involves a population of patients; second, there is a risk versus benefit assessment for each CT study; and third, there is an intrinsic risk for developing disease that is independent of radiation. The risk of cancer induction for individuals increases dramatically with decreasing age, becoming an important concern in younger individuals and women below the age of 20, thus, the necessity of carefully evaluating the risk versus benefit of pediatric CT scanning. CT RADIATION DOSE
Radiation dose is measured in units of absorbed dose, which is the amount of energy absorbed per unit of mass and has units of Gray (1 J/kg). Because some tissues are more radiation sensitive than others, it is sometimes useful to determine the effective dose (E) that has units of the Sievert (Sv) to estimate the stochastic risk from radiation. E is estimated from the absorbed dose for the critical organ (mGy) multiplied by an organ-specific conversion factor (k) that depends on the radiosensitivity of the organ and the age of the individual being scanned. Although E is not a true measure of dose, it makes it possible to compare biologic effects between different types of diagnostic examinations or different acquisition parameters [1,2]. Additionally, comparison of a CT scan effective dose to natural background dose (w3 mSv/year) can help patients assess the relative risk of a CT scan. CT radiation dose is measured using a parameter called volume CT dose index (CTDIvol) [3] that is reported in milliGray (mGy). CTDIvol is measured in 1 of 2 acrylic phantoms that represent the head (16 cm) or the body (32 cm) under standard scan conditions. The ª 2014 American College of Radiology 1546-1440/14/$36.00 http://dx.doi.org/10.1016/j.jacr.2013.10.011
Nelson/Reducing Pediatric CT Radiation Dose 293
Fig 1. This example shows CTDIvol values for 3 different phantom sizes (32 cm, 16 cm and 10 cm diameter). The measured CTDIvol increases as the phantom gets smaller for the same technical parameters. The reported console CTDIvol depends on the scan parameters selected particularly the field-of-view (FOV). The CTDIvol for a 16 cm phantom measurement has good agreement with the console value for the 16 cm FOV but is off if the FOV does not match the actual size of the object being scanned (red box). A similar situation occurs with the 32 cm phantom where the CTDIvol measurement has good agreement with the console value for the 32 cm FOV (brown box). Note that the 10 cm phantom measured CTDIvol value is significantly larger than the console value, pointing out the necessity of using the correct FOV for the patient’s size and anatomy. CTDIvol ¼ volume CT dose index.
CT scanner console reports the CTDIvol based on manufacturer factory measurements for each scan protocol configuration selected by the technologist. The medical physicist verifies the scanner console values by measuring the dose in the acrylic phantoms for each CT scanner as a part of routine quality control performance assessment. The CTDIvol only provides an estimate of patient dose whose accuracy depends on how close the patient size is to the phantom size (Fig. 1). The article by Seibert et al in this journal highlights how important the selection of phantom size is in dose estimation [4]. CTDIvol is a slice-specific dose measurement. To obtain a better overall estimate of the dose to the organ or patient, the dose-length product (DLP) is used. The DLP is the product of the length of the radiated scan volume and the average CTDIvol over that distance. It has units in milliGray-centimeter (mGy-cm). The CTDIvol and DLP depend on the patient’s anatomy and the corresponding scan’s technical factors over the entire scan area. As can be seen in Figure 2, CTDIvol varies with the anatomy being scanned. The DLP is an integration of individual CTDIvol values throughout the anatomy. E is the DLP multiplied by the conversion factor (k) that takes into account organ size and its radiosensitivity [2,5]. As can be appreciated in Table 1, k depends on the region of the body being scanned and the age of the individual being scanned. CT SCAN PARAMETERS AFFECTING RADIATION DOSE
Although new CT technology offers the capability to use subsecond rotational speeds plus multislice detectors to
image patients rapidly, an often underappreciated side effect is that faster scanning invites more frequent scanning of patients and more sophisticated studies with potentially higher radiation doses. There are several parameters that contribute to the radiation dose from modern CT scanners, including beam energy, beam current, detector configuration, pitch, dose length, and the number of imaging phases over the same anatomy. Beam Energy
The x-ray tube potential (kVp) is selected when setting up the scan protocol and determines the x-ray photon
Fig 2. CTDIvol values for a chest and abdominal CT scan that uses mA modulation to optimize the scan parameters based on the anatomy being imaged. The dose-length product is the sum of all the individual CTDIvol values. CTDIvol ¼ volume CT dose index.
294 Journal of the American College of Radiology/Vol. 11 No. 3 March 2014
Table 1. Normalized effective dose (mSv) per doselength product (mGy-cm) for adults and pediatric patients (k [mSv]/[mGy-cm]) Region 0 1 5 10 of Body Years Year Years Years Adult Head and neck Head Neck Chest Abdomen and pelvis Trunk
.0013 .011 .017 .039 .049
.0085 .0067 .012 .026 .030
.0057 .0040 .011 .018 .020
.0042 .0032 .0079 .013 .015
.0031 .0021 .0059 .014 .015
.044
.028
.019
.014
.015
Adult head and neck and pediatric patient k factors assume 16-cm head CT dose phantom data. All other k factors assume 32-cm diameter CT body phantom [18,19], based on ICRP report 60 [1].
energy. The radiation dose changes when changing between 2 different kVps and is approximately proportional to the square of the percentage change in tube potential, all other parameters being fixed, and so can have a large impact on radiation dose. Decreasing kVp in children can reduce the radiation dose and may improve soft tissue contrast (Fig. 3). However, there are several factors that should be taken into account when lower kVp techniques are considered. First, the milliAmpere second (mAs) will likely have to be increased to keep noise levels constant; second, a weight- or sizebased kVp/mAs/dose technique chart should be used to determine when a lower kVp is appropriate; third, a lower kVp may require longer scan times because of mAs limits that can increase motion artifacts; and fourth, a lower kVp may increase iodine conspicuity but not necessarily improve other soft tissue contrast. Tube Current
The tube current determines the number of electrons accelerated across the x-ray tube and, thus, the number of x-rays produced (x-ray fluence). Tube current is given in units of mA. Dose is directly proportional to the product of the mA and the slice scan time (s) or mAs. Taking into account the patient’s body size makes it possible to achieve significant reductions in dose by reducing mAs (Fig. 4).
Fig 3. Reducing kVp from 120 kVp to 100 kVp can reduce dose by w23%, which is especially important in children and also may improve soft tissue contrast as shown in this example (adapted from [17]). CTDIvol ¼ volume CT dose index.
Fig 4. A plot showing the relative beam current across a range of body sizes for both head and abdominal studies as a function of patient age. Adjusting mAs for patient body size can offer a potentially significant reduction in radiation dose. CTDIvol ¼ volume CT dose index.
Multiple Detector CT Systems
Modern CT scanners use multiple detector systems that first became available in the mid 1990s. Multiple detector CT scanners significantly reduce scan times while providing equal resolution in X, Y, and Z directions. The number of data channels times the detector width for each data channel determines the beam width, or aperture, collimation. As the aperture increases, the relative CTDIvol decreases, with the result that it is possible to achieve significant dose savings by using the widest possible aperture subject to considerations related to pitch as is discussed in the next section. Pitch
During scanning, the table moves through the gantry either in a continuous mode for helical scanning or in a step-and-shoot mode for axial scanning. The scan pitch is the ratio of the table feed per table gantry rotation divided by the beam width (Fig. 5). When the table moves the same distance as the beam width, the pitch is equal to one. The dose is inversely proportional to the pitch, so a pitch less than one means that the beam
Fig 5. The effect of changing the pitch of the scan. (Left) With a pitch less than 1, the beam width has some overlap at each view angle from rotation to rotation and a commensurate increase in dose. (Middle) With a pitch equal to 1, there is no overlap of beam width at each view angle and no view angles that are not covered at certain table positions. (Right) With a pitch greater than 1, some view angles are not covered by the beam width at certain table positions leading to possible gaps in scan data although there is a decrease in dose (adapted from [11]).
Nelson/Reducing Pediatric CT Radiation Dose 295
Effective Tube Current Time Product
The effective tube current time product is mAs divided by the pitch, also reported in mAs. Dose is directly proportional to the effective tube current time product. Lowering the mAs is the most direct way to reduce the radiation dose. However, lowering the mAs too much can produce a noisy, lower quality image resulting in a misdiagnosis or requiring a repeated scan. Fig 6. The consequences of helical over-ranging for scan coverage and dose. The use of a smaller aperture can reduce the dose for short helical acquisitions. A large aperture has significant excess dose at the ends of the acquisition and is not recommended unless a long acquisition is used, in which case the scan will complete more rapidly with a larger aperture. A short scan range will obtain dose-reduction benefits if a smaller aperture is used, because it will reduce overscan excess radiation dose.
overlaps previously radiated tissue with each rotation and the dose goes up. A pitch greater than one means that the dose will go down, but some tissue will not be imaged completely, potentially reducing image quality or introducing gaps that may miss important anatomy. Helical Over-Ranging
Helical over-ranging can be an important issue for pediatric patients who have shorter acquisition distances than adults. Over-ranging refers to the part of the scan that does not include a complete rotation at a given slice location [6] (Fig. 6). The dose from helical over-ranging is far greater for children than for adults and typically occurs at the end of the acquisition. Generally, it is better to perform a single helical scan acquisition rather than multiple scans. Fig 7. This figure shows the multiple scan phases for a brain perfusion scan. The perfusion scan has a significantly increased dose (CTDIvol 16.4 mGy) (DLP 3,500 mGy-cm) compared with a single scan (CTDIvol 45.9 mGy) (DLP 0.6 mGy-cm), even when dose-reduction methods are being used. Contrast intensity time distributions are useful; however, a significantly higher dose can result and, thus, scan protocol optimization is essential. CTDIvol ¼ volume CT dose index. DLP ¼ dose length product.
Exposure Length
The exposure length is the distance the patient anatomy in the Z direction is exposed to the x-ray beam. Exposure length always should be optimized by limiting the scan to the organ of interest. Scan Phases
The number of scan phases refers to the number of times the same patient anatomy is radiated, such as might occur in a multiphase liver study or a brain perfusion scan (Fig. 7). Multiphase scans significantly increase the dose to an organ, and most reported CT overdose events have been the result of multiphase scans. The CT perfusion of the head study (Fig. 7) shows multiple acquisitions of the same anatomy. This can be useful to produce intensity time distributions for a contrast material. However, that portion of anatomy scanned multiple times receives a significantly higher dose, and, therefore, scan protocol optimization is essential. The use of multiphase CT scanning in children should be limited to absolute necessity. ENHANCED DOSE-REDUCTION STRATEGIES
There are several enhanced dose-reduction strategies that can be used to help optimize scan protocols and reduce patient dose, including using localizer images to
296 Journal of the American College of Radiology/Vol. 11 No. 3 March 2014
Fig 8. This figure shows longitudinal tube current modulation through the chest and abdomen (upper left). Incorporation of angular tube current modulation can further reduce mA depending on patient dimension (upper right). Finally, combining these both results in optimum mAs reduction using both rotation and also longitudinal position (bottom) (adapted from [11]).
optimize the kVp and mAs to customize the scan to an individual patient’s anatomy. Faster computers facilitate more sophisticated image reconstruction algorithms, such as iterative reconstruction, and also can improve image quality while permitting reduced mAs and, therefore, reduced-dose imaging. Automatic kVp Selection
Localizer images can be used to estimate the patient attenuation and body size to select an optimum kVp to reduce radiation dose and improve soft tissue contrast. Larger patients generally require higher kVp, whereas smaller patients can be imaged using reduced kVp. Studies involving contrast also can select specific kVp values to enhance contrast material visibility. Tube Current Modulation
Scanner dose control with automatic tube current modulation uses localizer images to adjust the mAs based on patient anatomy (Fig. 8) [7,8], but requires the
patient to be precisely centered in the gantry to work properly. A consequence of mAs modulation is that although the CTDIvol dose will decrease for smaller patients, it will increase for larger patients. As mentioned, it is very important to position the patient at the center of rotation. When the patient is located below the center rotation, the anterior-posterior (AP) projection localizer image will appear to show a smaller patient than is actually present, leading to a reduction in mAs and possibly poor image quality. Conversely, if the patient is located above the center of rotation, the AP projection localizer image will appear to show a larger patient than is actually present, potentially leading to an increase in mAs and dose. Therefore, it is essential to confirm that the patient is centered in the gantry. ECG-based tube current modulation also can be used for cardiac studies to turn on radiation only during a specific time in the cardiac cycle for which imaging is desired [9].
Nelson/Reducing Pediatric CT Radiation Dose 297
Fig 9. This figure shows a typical multidetector CT renal perfusion study image using 160 mAs with filtered back projection (FBP) that is diagnostic (A), 16mAs with filtered back projection that is very noisy and difficult to interpret (B), and 16 mAs with iterative reconstruction that produced a diagnostic quality image (C) [18].
Organ-based tube current modulation is used to decrease tube current over radiation-sensitive areas, such as the breast, and increase it in other areas that are less radiation sensitive. This strategy must use sufficient average mAs to produce a diagnostic quality image so areas other than the “reduced dose” target area may receive a higher than normal dose. Iterative Reconstruction Algorithms
Using more sophisticated reconstruction algorithms also can reduce CT radiation dose. Traditional CT reconstruction is based on filtered back projection. Some improvements in image quality and signal-to-noise ratio can be obtained by adjusting the filtered back projection algorithms. There are significant limitations in how far these methods can be used to improve image quality to assist in dose reduction, as the increased statistical noise from lower mAs requires more severe filtering that can decrease spatial resolution. Iterative reconstruction algorithms use information acquired during the scan that is incorporated into repeated reconstruction steps to ultimately converge on an image with less noise and better image quality compared with filtered back projection. As a result, iterative reconstruction allows acquisition reconstruction of diagnostic quality images at far lower radiation doses (Fig. 9) [10]. In-Plane Shielding
Patient dose can be limited in some cases by using inplane shielding. In-plane shielding is controversial, but when properly used, it potentially reduces the dose to sensitive organs [11]. Shielding can only be applied after the localizer images are acquired; otherwise, it could increase the dose by giving incorrect values for automatic dose-modulation algorithms.
Patient Immobilization
Obtaining high-quality pediatric CT scans requires immobilizing infants and children during the acquisition to reduce motion artifacts and avoid repeat scans. One approach is to use different acquisition protocols with faster scan acquisition parameters. Alternatively, immobilization devices may be used to limit movement. OVERALL STRATEGIES TO MINIMIZE CT RADIATION DOSE IN PEDIATRIC PATIENTS
New scanner technology offers a variety of opportunities to reduce radiation dose while obtaining improved diagnostic information [12]. Automated tube current modulation, automated tube potential selection, and iterative reconstruction all can reduce patient dose while providing high-quality diagnostic studies. It is important to remember that new scan technology also offers new challenges by inviting overuse and more complex studies, including dynamic studies with repeated scanning that can produce very high doses [13]. The scan protocol needs to be optimized to the patient by making adjustments for patient age and size using dosereduction scan techniques whenever possible while still obtaining diagnostic information. Reduction of pediatric CT radiation dose and optimization of CT scan protocols in a pediatric patient require a team approach involving the radiologist, technologist, and physicist. The radiologist is responsible for ensuring that every imaging study in pediatric patients is thoughtful, appropriate, and medically indicated for that child. The radiologic technologist is responsible for ensuring that protocols are adjusted to use child-size parameters and verifying that proper scan technique factors are set for each pediatric scan. The medical
298 Journal of the American College of Radiology/Vol. 11 No. 3 March 2014
Table 2. Useful websites related to CT scanning and dose reduction www.imagegently.org www.cancer.gov/cancertopics/causes radiation-riskspediatric-CT www.acr.org http://hps.org www.rsna.org www.radiologyinfo.org http://www.cancer.gov/cancertopics/causes%20radiationrisks-%20pediatric-CT https://rpop.iaea.org/RPoP/RPoP/Content/index.htm
physicist is responsible for improving image quality by providing guidance on appropriate pediatric techniques and ensuring that young patients are imaged using radiation doses as low as reasonably achievable (ALARA). There are several excellent web-based resources that can be useful in reducing CT radiation dose in children (Table 2). When scheduling a pediatric CT scan, or any other patient scan, it is important to consider and review the number of previous radiation studies that have been performed to ensure that the study requested is essential and that cumulative radiation doses are not excessive [14]. It also is important to assess the reasons for the requested CT scan and follow ACR Appropriateness Criteria and Practice Guidelines related to pediatric CT studies [15].
It is important to perform only necessary CT scans after considering alternative imaging modalities, such as MRI or ultrasound, whenever possible. Children can be frightened and anxious, so it is important to create a childfriendly environment for the patient that is comforting. After it has been decided that a CT scan is necessary, scan only the organ or anatomical region indicated. Decrease the kVp for smaller sized patients and use mAs modulation, and the appropriate field-of-view for scanning [16]. Use iterative reconstruction methods to further reduce dose in patients. Importantly, eliminate multiphase CT studies whenever possible. Pre- and postcontrast studies may not provide a benefit over only postcontrast imaging, and avoiding precontrast studies might offer a potential 50% dose reduction. Reduce mAs even further for repeat studies, such as might be used for verification of a shunt placement when the diagnostic requirements may be different from those of an initial imaging study, without compromising diagnostic information. Following these strategies can result in improved diagnostic quality and reduced radiation dose to our youngest and most vulnerable patients. Table 3 summarizes the key concepts discussed. TAKE-HOME POINTS
Use a team approach to review and optimize scan protocols and determine best practice by implementing child-size imaging guidelines.
Table 3. Strategies for dose reduction in pediatric CT studies (adapted from [20]) Patient Management Factors to Reduce Radiation Dose: Use a team approach to review and optimize scan protocols and determine best practice for each patient. Triage all examinations to eliminate inappropriate referrals and consider the use of alternative modalities, such as conventional radiography, sonography, or MRI whenever possible. Define and implement child-size imaging guidelines. Develop guidelines to compare protocol dose indicators with diagnostic reference levels. Educate patients by explaining the risks of ionizing radiation and why the potential benefits from CT scanning procedures may outweigh the risks. Technical Factors to Reduce CT Radiation Dose: Optimize CT scan parameter settings based on patient weight or diameter and anatomic region of interest to optimally balance image quality and radiation dose. Develop and use a chart or table of tube-current settings based on patient weight (preferably diameter) and anatomical region of interest. Utilize mAs modulation to reduce tube current whenever possible and set dose limits to optimize image quality and reduce radiation dose. Increase table increment (axial scanning) or pitch (helical scanning) to optimize tube-current and pitch settings for diagnostic requirements. CT scanner vendors can offer recommendations to optimize image quality and reduce dose. Reduce the number of multiple phase scans (eg, precontrast) with contrast material unless medically essential. Ensure that CT dose (CTDIvol) and DLP data are visible on the display console and recorded in the patient medical record. Educational Considerations: Educate physician colleagues regarding alternative studies and how to optimize CT scan parameters to reduce radiation dose. Make sure health care professionals are knowledgeable and prepared to discuss with parents why CT is sometimes the best examination to diagnose conditions in children. Educate ordering physicians to request only necessary CT examinations and first consider alternative modalities such as ultrasound or MRI. Emerging Trends for CT Dose Management: Guidelines and standards are evolving that may lead to regulations for monitoring lifetime cumulative radiation dose from medical imaging sources. More attention will be focused on cancer risks for pediatric patients undergoing multiple CT scans. Additional support will go to CT research and clinical studies. New government regulations are possible, with Europe leading the way.
Nelson/Reducing Pediatric CT Radiation Dose 299
Triage all examinations to eliminate inappropriate referrals and consider the use of alternative modalities such as conventional radiography, sonography, or MRI whenever possible. Optimize CT scan parameter settings based on patient weight or diameter and anatomic region of interest to optimally balance image quality and radiation dose using mAs modulation to reduce tube current whenever possible and set dose limits to optimize image quality. The primary factors that can reduce dose include kVp, mAs, pitch, and using current modulation and iterative reconstruction algorithms whenever possible. Reduce the number of multiple phase scans (eg, precontrast) with contrast material unless medically essential. Educate ordering physicians to request only necessary CT examinations and first consider alternative modalities such as ultrasound or MRI. ACKNOWLEDGMENTS Supported in part by UC Center for Health Quality & Innovation Program, Standardization and Optimization of Computed Tomography Patient Radiation Dose Across the University of California Medical Centers. REFERENCES 1. International Commission on Radiological Protection. 1990 recommendations of the international commission on radiological protection (Report 60). Ann ICRP 1991;21. 2. McCollough CH, Schueler BA. Calculation of effective dose. Med Phys 2000;27:828-37. 3. Boone JM. The trouble with CTDI100. Med Phys 2007;34:1364-71. 4. Seibert JA, Boone JM, Wootton-Gorges, Lamba R. Dose is not always what it seems: where very misleading values can result from CT dose index and dose length product. J Am Coll Radiol 2014;39:233-7. 5. International Commission on Radiological Protection. 2007 recommendations of the international commission on radiological protection (ICRP Publication 103). Ann ICRP 2007;37:1-332.
6. Schilham A, van der Molen AJ, Prokop M, de Jong HW. Overranging at multi-section CT: an underestimated source of excess radiation exposure. Radiographics 2010;30:1057-67. 7. Gies M, Kalender WA, Wolf H, et al. Dose reduction in CT by anatomically adapted tube current modulation: simulation studies. Med Phys 1999;26:2235-47. 8. Graser A, Wintersperger BJ, Suess C, et al. Dose reduction and image quality in MDCT colonography using tube currentmodulation. AJR Am J Roentgenol 2006;187:695-701. 9. Jin KN, Park EA, Shin CL, Lee W, Chung JW, Park JH. Retrospective versus prospective ECG-gated dual source CT in pediatric patients with congenital heart disease: comparison of image quality and radiation dose. Int J Cardiovasc Imaging 2010;26:63-73. 10. Vorona GA, Ceschin RC, Clayton BL, Sutcavage T, Tadros SS, Panigraphy A. Reducing abdominal CT radiation dose with the adaptive statistical iterative reconstruction technique in children: a feasibility study. Pediatr Radiol 2011;41:1174-82. 11. American Association of Physicists in Medicine. AAPM position statement on the use of bismuth shielding for the purpose of dose reduction in CT scanning. American Association of Physicists, College Park, MD, 2012. Available at: http://www.aapm.org/publicgeneral/ BismuthShielding.pdf. 12. Raman SP, Johnson PT, Deshmukh S, Mahesh M, Grant KL, Fishman EK. CT dose reduction applications: available tools on the latest generation of CT scanners. J Am Coll Radiol 2013;10:37-41. 13. Brenner DJ, Hall EJ. Computed tomography: an increasing source of radiation exposure. N Engl J Med 2007;357:2277-84. 14. Donnelly LF. Reducing radiation dose associated with pediatric CT by decreasing unnecessary examinations. AJR Am J Roentgen 2005;184: 655-7. 15. ACR Practice Guideline for The Performance of Pediatric Computed Tomography (CT), American College of Radiology, Reston VA, 2008. http://www.acr.org/w/media/ACR/Documents/PGTS/guidelines/CT_ Pediatric.pdf. 16. Huda W, Bushong SC. In x-ray computed tomography, technique factors should be selected appropriate to patient size. Med Phys 2001;28:1543-5. 17. Cody DD, Kofler JM. Optimizing CT Image Protocols with Respect To Image Quality and Radiation Dose, AAPM Summer School, San Diego, 2012. 18. Yu L, Liu X, Leng S, Kofler JM, Ramierz-Giraldo, et al. Radiation dose reduction in computed tomography: techniques and future perspectives. Imaging Med 2009;1:65-84. 19. Shrimpton PC, Hillier MC, Lewis MA, et al. National survey of doses from CT in the UK: 2003. Br J Radiol 2006;79:968-80. 20. Colang JE, Killion JB, Vano E. Patient dose from CT: a literature review, Radiol Technol 2007;79:17-26.
The objective of this article is to provide a brief review of CT scanning radiation sensitivity in children and explain CT scan parameters that affect radiation dose. We discuss key factors influencing radiation dose and study quality and how these factors can be used to optimize scan protocols with the goal of reducing pediatric CT radiation dose without compromising diagnostic quality. Finally, we provide some practical tips for reducing radiation doses to children. Key Words: Pediatric, CT, radiation, dose, dose reduction, automatic exposure control, protocol optimization J Am Coll Radiol 2014;11:292-299. Copyright © 2014 American College of Radiology
INTRODUCTION
New CT technology has facilitated more rapid and sophisticated patient scanning. The potential for increased patient radiation dose resulting from this technology makes it essential to ensure that each CT scan performed is both medically indicated and uses optimized scanning protocols. Children are not small adults and have quite different imaging needs. Their cells are dividing in different ways and are more radiation sensitive, and their body proportions are different. Radiation’s effects on children’s cells also are cumulative and radiation-related risk increases with each dose. Thus, it is important to optimize scan protocols to the individual to maximize diagnostic information and minimize radiation dose. There are a number of strategies that are available to help patients, technologists, and physicians understand and optimize pediatric CT scanning. Although the radiation dose from a single CT scan is relatively low, when adult-sized imaging techniques are used, children receive a higher radiation dose. Because children’s cells are more susceptible to the effects of radiation, there is a slightly increased risk for developing cancer from stochastic, or nondeterministic, effects. Also, there may be changes to children’s genes from radiation doses received over the course of their lifetimes with the possibility that genetic effects may be passed on to their offspring. Deterministic effects also can occur above certain dose thresholds and typically are the most readily visible effects that we identify, such as hair loss at high doses. Some studies have suggested that pediatric
Department of Radiology, University of California, San Diego, La Jolla, California. Corresponding author and reprints: Thomas R. Nelson, PhD, Department of Radiology, M-0610, University of California, San Diego, La Jolla, CA, 92093-0610; e-mail: [email protected].
292
CT scan doses are higher than necessary, so we need to be particularly vigilant in how we optimize scanning protocols to minimize dose. Defining CT scan radiation risk is complex for several reasons. First, risk assessment involves a population of patients; second, there is a risk versus benefit assessment for each CT study; and third, there is an intrinsic risk for developing disease that is independent of radiation. The risk of cancer induction for individuals increases dramatically with decreasing age, becoming an important concern in younger individuals and women below the age of 20, thus, the necessity of carefully evaluating the risk versus benefit of pediatric CT scanning. CT RADIATION DOSE
Radiation dose is measured in units of absorbed dose, which is the amount of energy absorbed per unit of mass and has units of Gray (1 J/kg). Because some tissues are more radiation sensitive than others, it is sometimes useful to determine the effective dose (E) that has units of the Sievert (Sv) to estimate the stochastic risk from radiation. E is estimated from the absorbed dose for the critical organ (mGy) multiplied by an organ-specific conversion factor (k) that depends on the radiosensitivity of the organ and the age of the individual being scanned. Although E is not a true measure of dose, it makes it possible to compare biologic effects between different types of diagnostic examinations or different acquisition parameters [1,2]. Additionally, comparison of a CT scan effective dose to natural background dose (w3 mSv/year) can help patients assess the relative risk of a CT scan. CT radiation dose is measured using a parameter called volume CT dose index (CTDIvol) [3] that is reported in milliGray (mGy). CTDIvol is measured in 1 of 2 acrylic phantoms that represent the head (16 cm) or the body (32 cm) under standard scan conditions. The ª 2014 American College of Radiology 1546-1440/14/$36.00 http://dx.doi.org/10.1016/j.jacr.2013.10.011
Nelson/Reducing Pediatric CT Radiation Dose 293
Fig 1. This example shows CTDIvol values for 3 different phantom sizes (32 cm, 16 cm and 10 cm diameter). The measured CTDIvol increases as the phantom gets smaller for the same technical parameters. The reported console CTDIvol depends on the scan parameters selected particularly the field-of-view (FOV). The CTDIvol for a 16 cm phantom measurement has good agreement with the console value for the 16 cm FOV but is off if the FOV does not match the actual size of the object being scanned (red box). A similar situation occurs with the 32 cm phantom where the CTDIvol measurement has good agreement with the console value for the 32 cm FOV (brown box). Note that the 10 cm phantom measured CTDIvol value is significantly larger than the console value, pointing out the necessity of using the correct FOV for the patient’s size and anatomy. CTDIvol ¼ volume CT dose index.
CT scanner console reports the CTDIvol based on manufacturer factory measurements for each scan protocol configuration selected by the technologist. The medical physicist verifies the scanner console values by measuring the dose in the acrylic phantoms for each CT scanner as a part of routine quality control performance assessment. The CTDIvol only provides an estimate of patient dose whose accuracy depends on how close the patient size is to the phantom size (Fig. 1). The article by Seibert et al in this journal highlights how important the selection of phantom size is in dose estimation [4]. CTDIvol is a slice-specific dose measurement. To obtain a better overall estimate of the dose to the organ or patient, the dose-length product (DLP) is used. The DLP is the product of the length of the radiated scan volume and the average CTDIvol over that distance. It has units in milliGray-centimeter (mGy-cm). The CTDIvol and DLP depend on the patient’s anatomy and the corresponding scan’s technical factors over the entire scan area. As can be seen in Figure 2, CTDIvol varies with the anatomy being scanned. The DLP is an integration of individual CTDIvol values throughout the anatomy. E is the DLP multiplied by the conversion factor (k) that takes into account organ size and its radiosensitivity [2,5]. As can be appreciated in Table 1, k depends on the region of the body being scanned and the age of the individual being scanned. CT SCAN PARAMETERS AFFECTING RADIATION DOSE
Although new CT technology offers the capability to use subsecond rotational speeds plus multislice detectors to
image patients rapidly, an often underappreciated side effect is that faster scanning invites more frequent scanning of patients and more sophisticated studies with potentially higher radiation doses. There are several parameters that contribute to the radiation dose from modern CT scanners, including beam energy, beam current, detector configuration, pitch, dose length, and the number of imaging phases over the same anatomy. Beam Energy
The x-ray tube potential (kVp) is selected when setting up the scan protocol and determines the x-ray photon
Fig 2. CTDIvol values for a chest and abdominal CT scan that uses mA modulation to optimize the scan parameters based on the anatomy being imaged. The dose-length product is the sum of all the individual CTDIvol values. CTDIvol ¼ volume CT dose index.
294 Journal of the American College of Radiology/Vol. 11 No. 3 March 2014
Table 1. Normalized effective dose (mSv) per doselength product (mGy-cm) for adults and pediatric patients (k [mSv]/[mGy-cm]) Region 0 1 5 10 of Body Years Year Years Years Adult Head and neck Head Neck Chest Abdomen and pelvis Trunk
.0013 .011 .017 .039 .049
.0085 .0067 .012 .026 .030
.0057 .0040 .011 .018 .020
.0042 .0032 .0079 .013 .015
.0031 .0021 .0059 .014 .015
.044
.028
.019
.014
.015
Adult head and neck and pediatric patient k factors assume 16-cm head CT dose phantom data. All other k factors assume 32-cm diameter CT body phantom [18,19], based on ICRP report 60 [1].
energy. The radiation dose changes when changing between 2 different kVps and is approximately proportional to the square of the percentage change in tube potential, all other parameters being fixed, and so can have a large impact on radiation dose. Decreasing kVp in children can reduce the radiation dose and may improve soft tissue contrast (Fig. 3). However, there are several factors that should be taken into account when lower kVp techniques are considered. First, the milliAmpere second (mAs) will likely have to be increased to keep noise levels constant; second, a weight- or sizebased kVp/mAs/dose technique chart should be used to determine when a lower kVp is appropriate; third, a lower kVp may require longer scan times because of mAs limits that can increase motion artifacts; and fourth, a lower kVp may increase iodine conspicuity but not necessarily improve other soft tissue contrast. Tube Current
The tube current determines the number of electrons accelerated across the x-ray tube and, thus, the number of x-rays produced (x-ray fluence). Tube current is given in units of mA. Dose is directly proportional to the product of the mA and the slice scan time (s) or mAs. Taking into account the patient’s body size makes it possible to achieve significant reductions in dose by reducing mAs (Fig. 4).
Fig 3. Reducing kVp from 120 kVp to 100 kVp can reduce dose by w23%, which is especially important in children and also may improve soft tissue contrast as shown in this example (adapted from [17]). CTDIvol ¼ volume CT dose index.
Fig 4. A plot showing the relative beam current across a range of body sizes for both head and abdominal studies as a function of patient age. Adjusting mAs for patient body size can offer a potentially significant reduction in radiation dose. CTDIvol ¼ volume CT dose index.
Multiple Detector CT Systems
Modern CT scanners use multiple detector systems that first became available in the mid 1990s. Multiple detector CT scanners significantly reduce scan times while providing equal resolution in X, Y, and Z directions. The number of data channels times the detector width for each data channel determines the beam width, or aperture, collimation. As the aperture increases, the relative CTDIvol decreases, with the result that it is possible to achieve significant dose savings by using the widest possible aperture subject to considerations related to pitch as is discussed in the next section. Pitch
During scanning, the table moves through the gantry either in a continuous mode for helical scanning or in a step-and-shoot mode for axial scanning. The scan pitch is the ratio of the table feed per table gantry rotation divided by the beam width (Fig. 5). When the table moves the same distance as the beam width, the pitch is equal to one. The dose is inversely proportional to the pitch, so a pitch less than one means that the beam
Fig 5. The effect of changing the pitch of the scan. (Left) With a pitch less than 1, the beam width has some overlap at each view angle from rotation to rotation and a commensurate increase in dose. (Middle) With a pitch equal to 1, there is no overlap of beam width at each view angle and no view angles that are not covered at certain table positions. (Right) With a pitch greater than 1, some view angles are not covered by the beam width at certain table positions leading to possible gaps in scan data although there is a decrease in dose (adapted from [11]).
Nelson/Reducing Pediatric CT Radiation Dose 295
Effective Tube Current Time Product
The effective tube current time product is mAs divided by the pitch, also reported in mAs. Dose is directly proportional to the effective tube current time product. Lowering the mAs is the most direct way to reduce the radiation dose. However, lowering the mAs too much can produce a noisy, lower quality image resulting in a misdiagnosis or requiring a repeated scan. Fig 6. The consequences of helical over-ranging for scan coverage and dose. The use of a smaller aperture can reduce the dose for short helical acquisitions. A large aperture has significant excess dose at the ends of the acquisition and is not recommended unless a long acquisition is used, in which case the scan will complete more rapidly with a larger aperture. A short scan range will obtain dose-reduction benefits if a smaller aperture is used, because it will reduce overscan excess radiation dose.
overlaps previously radiated tissue with each rotation and the dose goes up. A pitch greater than one means that the dose will go down, but some tissue will not be imaged completely, potentially reducing image quality or introducing gaps that may miss important anatomy. Helical Over-Ranging
Helical over-ranging can be an important issue for pediatric patients who have shorter acquisition distances than adults. Over-ranging refers to the part of the scan that does not include a complete rotation at a given slice location [6] (Fig. 6). The dose from helical over-ranging is far greater for children than for adults and typically occurs at the end of the acquisition. Generally, it is better to perform a single helical scan acquisition rather than multiple scans. Fig 7. This figure shows the multiple scan phases for a brain perfusion scan. The perfusion scan has a significantly increased dose (CTDIvol 16.4 mGy) (DLP 3,500 mGy-cm) compared with a single scan (CTDIvol 45.9 mGy) (DLP 0.6 mGy-cm), even when dose-reduction methods are being used. Contrast intensity time distributions are useful; however, a significantly higher dose can result and, thus, scan protocol optimization is essential. CTDIvol ¼ volume CT dose index. DLP ¼ dose length product.
Exposure Length
The exposure length is the distance the patient anatomy in the Z direction is exposed to the x-ray beam. Exposure length always should be optimized by limiting the scan to the organ of interest. Scan Phases
The number of scan phases refers to the number of times the same patient anatomy is radiated, such as might occur in a multiphase liver study or a brain perfusion scan (Fig. 7). Multiphase scans significantly increase the dose to an organ, and most reported CT overdose events have been the result of multiphase scans. The CT perfusion of the head study (Fig. 7) shows multiple acquisitions of the same anatomy. This can be useful to produce intensity time distributions for a contrast material. However, that portion of anatomy scanned multiple times receives a significantly higher dose, and, therefore, scan protocol optimization is essential. The use of multiphase CT scanning in children should be limited to absolute necessity. ENHANCED DOSE-REDUCTION STRATEGIES
There are several enhanced dose-reduction strategies that can be used to help optimize scan protocols and reduce patient dose, including using localizer images to
296 Journal of the American College of Radiology/Vol. 11 No. 3 March 2014
Fig 8. This figure shows longitudinal tube current modulation through the chest and abdomen (upper left). Incorporation of angular tube current modulation can further reduce mA depending on patient dimension (upper right). Finally, combining these both results in optimum mAs reduction using both rotation and also longitudinal position (bottom) (adapted from [11]).
optimize the kVp and mAs to customize the scan to an individual patient’s anatomy. Faster computers facilitate more sophisticated image reconstruction algorithms, such as iterative reconstruction, and also can improve image quality while permitting reduced mAs and, therefore, reduced-dose imaging. Automatic kVp Selection
Localizer images can be used to estimate the patient attenuation and body size to select an optimum kVp to reduce radiation dose and improve soft tissue contrast. Larger patients generally require higher kVp, whereas smaller patients can be imaged using reduced kVp. Studies involving contrast also can select specific kVp values to enhance contrast material visibility. Tube Current Modulation
Scanner dose control with automatic tube current modulation uses localizer images to adjust the mAs based on patient anatomy (Fig. 8) [7,8], but requires the
patient to be precisely centered in the gantry to work properly. A consequence of mAs modulation is that although the CTDIvol dose will decrease for smaller patients, it will increase for larger patients. As mentioned, it is very important to position the patient at the center of rotation. When the patient is located below the center rotation, the anterior-posterior (AP) projection localizer image will appear to show a smaller patient than is actually present, leading to a reduction in mAs and possibly poor image quality. Conversely, if the patient is located above the center of rotation, the AP projection localizer image will appear to show a larger patient than is actually present, potentially leading to an increase in mAs and dose. Therefore, it is essential to confirm that the patient is centered in the gantry. ECG-based tube current modulation also can be used for cardiac studies to turn on radiation only during a specific time in the cardiac cycle for which imaging is desired [9].
Nelson/Reducing Pediatric CT Radiation Dose 297
Fig 9. This figure shows a typical multidetector CT renal perfusion study image using 160 mAs with filtered back projection (FBP) that is diagnostic (A), 16mAs with filtered back projection that is very noisy and difficult to interpret (B), and 16 mAs with iterative reconstruction that produced a diagnostic quality image (C) [18].
Organ-based tube current modulation is used to decrease tube current over radiation-sensitive areas, such as the breast, and increase it in other areas that are less radiation sensitive. This strategy must use sufficient average mAs to produce a diagnostic quality image so areas other than the “reduced dose” target area may receive a higher than normal dose. Iterative Reconstruction Algorithms
Using more sophisticated reconstruction algorithms also can reduce CT radiation dose. Traditional CT reconstruction is based on filtered back projection. Some improvements in image quality and signal-to-noise ratio can be obtained by adjusting the filtered back projection algorithms. There are significant limitations in how far these methods can be used to improve image quality to assist in dose reduction, as the increased statistical noise from lower mAs requires more severe filtering that can decrease spatial resolution. Iterative reconstruction algorithms use information acquired during the scan that is incorporated into repeated reconstruction steps to ultimately converge on an image with less noise and better image quality compared with filtered back projection. As a result, iterative reconstruction allows acquisition reconstruction of diagnostic quality images at far lower radiation doses (Fig. 9) [10]. In-Plane Shielding
Patient dose can be limited in some cases by using inplane shielding. In-plane shielding is controversial, but when properly used, it potentially reduces the dose to sensitive organs [11]. Shielding can only be applied after the localizer images are acquired; otherwise, it could increase the dose by giving incorrect values for automatic dose-modulation algorithms.
Patient Immobilization
Obtaining high-quality pediatric CT scans requires immobilizing infants and children during the acquisition to reduce motion artifacts and avoid repeat scans. One approach is to use different acquisition protocols with faster scan acquisition parameters. Alternatively, immobilization devices may be used to limit movement. OVERALL STRATEGIES TO MINIMIZE CT RADIATION DOSE IN PEDIATRIC PATIENTS
New scanner technology offers a variety of opportunities to reduce radiation dose while obtaining improved diagnostic information [12]. Automated tube current modulation, automated tube potential selection, and iterative reconstruction all can reduce patient dose while providing high-quality diagnostic studies. It is important to remember that new scan technology also offers new challenges by inviting overuse and more complex studies, including dynamic studies with repeated scanning that can produce very high doses [13]. The scan protocol needs to be optimized to the patient by making adjustments for patient age and size using dosereduction scan techniques whenever possible while still obtaining diagnostic information. Reduction of pediatric CT radiation dose and optimization of CT scan protocols in a pediatric patient require a team approach involving the radiologist, technologist, and physicist. The radiologist is responsible for ensuring that every imaging study in pediatric patients is thoughtful, appropriate, and medically indicated for that child. The radiologic technologist is responsible for ensuring that protocols are adjusted to use child-size parameters and verifying that proper scan technique factors are set for each pediatric scan. The medical
298 Journal of the American College of Radiology/Vol. 11 No. 3 March 2014
Table 2. Useful websites related to CT scanning and dose reduction www.imagegently.org www.cancer.gov/cancertopics/causes radiation-riskspediatric-CT www.acr.org http://hps.org www.rsna.org www.radiologyinfo.org http://www.cancer.gov/cancertopics/causes%20radiationrisks-%20pediatric-CT https://rpop.iaea.org/RPoP/RPoP/Content/index.htm
physicist is responsible for improving image quality by providing guidance on appropriate pediatric techniques and ensuring that young patients are imaged using radiation doses as low as reasonably achievable (ALARA). There are several excellent web-based resources that can be useful in reducing CT radiation dose in children (Table 2). When scheduling a pediatric CT scan, or any other patient scan, it is important to consider and review the number of previous radiation studies that have been performed to ensure that the study requested is essential and that cumulative radiation doses are not excessive [14]. It also is important to assess the reasons for the requested CT scan and follow ACR Appropriateness Criteria and Practice Guidelines related to pediatric CT studies [15].
It is important to perform only necessary CT scans after considering alternative imaging modalities, such as MRI or ultrasound, whenever possible. Children can be frightened and anxious, so it is important to create a childfriendly environment for the patient that is comforting. After it has been decided that a CT scan is necessary, scan only the organ or anatomical region indicated. Decrease the kVp for smaller sized patients and use mAs modulation, and the appropriate field-of-view for scanning [16]. Use iterative reconstruction methods to further reduce dose in patients. Importantly, eliminate multiphase CT studies whenever possible. Pre- and postcontrast studies may not provide a benefit over only postcontrast imaging, and avoiding precontrast studies might offer a potential 50% dose reduction. Reduce mAs even further for repeat studies, such as might be used for verification of a shunt placement when the diagnostic requirements may be different from those of an initial imaging study, without compromising diagnostic information. Following these strategies can result in improved diagnostic quality and reduced radiation dose to our youngest and most vulnerable patients. Table 3 summarizes the key concepts discussed. TAKE-HOME POINTS
Use a team approach to review and optimize scan protocols and determine best practice by implementing child-size imaging guidelines.
Table 3. Strategies for dose reduction in pediatric CT studies (adapted from [20]) Patient Management Factors to Reduce Radiation Dose: Use a team approach to review and optimize scan protocols and determine best practice for each patient. Triage all examinations to eliminate inappropriate referrals and consider the use of alternative modalities, such as conventional radiography, sonography, or MRI whenever possible. Define and implement child-size imaging guidelines. Develop guidelines to compare protocol dose indicators with diagnostic reference levels. Educate patients by explaining the risks of ionizing radiation and why the potential benefits from CT scanning procedures may outweigh the risks. Technical Factors to Reduce CT Radiation Dose: Optimize CT scan parameter settings based on patient weight or diameter and anatomic region of interest to optimally balance image quality and radiation dose. Develop and use a chart or table of tube-current settings based on patient weight (preferably diameter) and anatomical region of interest. Utilize mAs modulation to reduce tube current whenever possible and set dose limits to optimize image quality and reduce radiation dose. Increase table increment (axial scanning) or pitch (helical scanning) to optimize tube-current and pitch settings for diagnostic requirements. CT scanner vendors can offer recommendations to optimize image quality and reduce dose. Reduce the number of multiple phase scans (eg, precontrast) with contrast material unless medically essential. Ensure that CT dose (CTDIvol) and DLP data are visible on the display console and recorded in the patient medical record. Educational Considerations: Educate physician colleagues regarding alternative studies and how to optimize CT scan parameters to reduce radiation dose. Make sure health care professionals are knowledgeable and prepared to discuss with parents why CT is sometimes the best examination to diagnose conditions in children. Educate ordering physicians to request only necessary CT examinations and first consider alternative modalities such as ultrasound or MRI. Emerging Trends for CT Dose Management: Guidelines and standards are evolving that may lead to regulations for monitoring lifetime cumulative radiation dose from medical imaging sources. More attention will be focused on cancer risks for pediatric patients undergoing multiple CT scans. Additional support will go to CT research and clinical studies. New government regulations are possible, with Europe leading the way.
Nelson/Reducing Pediatric CT Radiation Dose 299
Triage all examinations to eliminate inappropriate referrals and consider the use of alternative modalities such as conventional radiography, sonography, or MRI whenever possible. Optimize CT scan parameter settings based on patient weight or diameter and anatomic region of interest to optimally balance image quality and radiation dose using mAs modulation to reduce tube current whenever possible and set dose limits to optimize image quality. The primary factors that can reduce dose include kVp, mAs, pitch, and using current modulation and iterative reconstruction algorithms whenever possible. Reduce the number of multiple phase scans (eg, precontrast) with contrast material unless medically essential. Educate ordering physicians to request only necessary CT examinations and first consider alternative modalities such as ultrasound or MRI. ACKNOWLEDGMENTS Supported in part by UC Center for Health Quality & Innovation Program, Standardization and Optimization of Computed Tomography Patient Radiation Dose Across the University of California Medical Centers. REFERENCES 1. International Commission on Radiological Protection. 1990 recommendations of the international commission on radiological protection (Report 60). Ann ICRP 1991;21. 2. McCollough CH, Schueler BA. Calculation of effective dose. Med Phys 2000;27:828-37. 3. Boone JM. The trouble with CTDI100. Med Phys 2007;34:1364-71. 4. Seibert JA, Boone JM, Wootton-Gorges, Lamba R. Dose is not always what it seems: where very misleading values can result from CT dose index and dose length product. J Am Coll Radiol 2014;39:233-7. 5. International Commission on Radiological Protection. 2007 recommendations of the international commission on radiological protection (ICRP Publication 103). Ann ICRP 2007;37:1-332.
6. Schilham A, van der Molen AJ, Prokop M, de Jong HW. Overranging at multi-section CT: an underestimated source of excess radiation exposure. Radiographics 2010;30:1057-67. 7. Gies M, Kalender WA, Wolf H, et al. Dose reduction in CT by anatomically adapted tube current modulation: simulation studies. Med Phys 1999;26:2235-47. 8. Graser A, Wintersperger BJ, Suess C, et al. Dose reduction and image quality in MDCT colonography using tube currentmodulation. AJR Am J Roentgenol 2006;187:695-701. 9. Jin KN, Park EA, Shin CL, Lee W, Chung JW, Park JH. Retrospective versus prospective ECG-gated dual source CT in pediatric patients with congenital heart disease: comparison of image quality and radiation dose. Int J Cardiovasc Imaging 2010;26:63-73. 10. Vorona GA, Ceschin RC, Clayton BL, Sutcavage T, Tadros SS, Panigraphy A. Reducing abdominal CT radiation dose with the adaptive statistical iterative reconstruction technique in children: a feasibility study. Pediatr Radiol 2011;41:1174-82. 11. American Association of Physicists in Medicine. AAPM position statement on the use of bismuth shielding for the purpose of dose reduction in CT scanning. American Association of Physicists, College Park, MD, 2012. Available at: http://www.aapm.org/publicgeneral/ BismuthShielding.pdf. 12. Raman SP, Johnson PT, Deshmukh S, Mahesh M, Grant KL, Fishman EK. CT dose reduction applications: available tools on the latest generation of CT scanners. J Am Coll Radiol 2013;10:37-41. 13. Brenner DJ, Hall EJ. Computed tomography: an increasing source of radiation exposure. N Engl J Med 2007;357:2277-84. 14. Donnelly LF. Reducing radiation dose associated with pediatric CT by decreasing unnecessary examinations. AJR Am J Roentgen 2005;184: 655-7. 15. ACR Practice Guideline for The Performance of Pediatric Computed Tomography (CT), American College of Radiology, Reston VA, 2008. http://www.acr.org/w/media/ACR/Documents/PGTS/guidelines/CT_ Pediatric.pdf. 16. Huda W, Bushong SC. In x-ray computed tomography, technique factors should be selected appropriate to patient size. Med Phys 2001;28:1543-5. 17. Cody DD, Kofler JM. Optimizing CT Image Protocols with Respect To Image Quality and Radiation Dose, AAPM Summer School, San Diego, 2012. 18. Yu L, Liu X, Leng S, Kofler JM, Ramierz-Giraldo, et al. Radiation dose reduction in computed tomography: techniques and future perspectives. Imaging Med 2009;1:65-84. 19. Shrimpton PC, Hillier MC, Lewis MA, et al. National survey of doses from CT in the UK: 2003. Br J Radiol 2006;79:968-80. 20. Colang JE, Killion JB, Vano E. Patient dose from CT: a literature review, Radiol Technol 2007;79:17-26.
