- Email: [email protected]
How Much Does Lead Shielding during Fluoroscopy Reduce Radiation Dose to Out-of-Field Body Parts?
Journal of Medical Imaging and Radiation Sciences
Journal of Medical Imaging and Radiation Sciences xx (2016) 1-7
Journal de l’imagerie médicale et des sciences de la radiation
www.elsevier.com/locate/jmir
How Much Does Lead Shielding during Fluoroscopy Reduce Radiation Dose to Out-of-Field Body Parts? Andrew S. Phelps, MD*, Robert G. Gould, ScD, Jesse L. Courtier, MD, Peter A. Marcovici, MD, Christina Salani, RT(R)(ARRT) and John D. MacKenzie, MD Department of Radiology & Biomedical Imaging, UCSF Benioff Children’s Hospital, University of California, San Francisco, California, USA
ABSTRACT Background: Fluoroscopy technologists routinely place a lead shield between the x-ray table and the patient’s gonads, even if the gonads are not directly in the x-ray field. Internal scatter radiation is the greatest source of radiation to out-of-field body parts, but a shield placed between the patient and the x-ray source will not block internal scatter. Prior nonfluoroscopy research has shown that there is a small reduction in radiation dose when shielding the leakage radiation that penetrates through the collimator shutters. The goal of this in vitro study was to determine if there was any radiation dose reduction when shielding leakage radiation during fluoroscopy. Methods: This was an in vitro comparison study of radiation doses using different collimation and shielding strategies during fluoroscopy. Ionization chamber measurements were obtained during fluoroscopy of an acrylic block with and without collimation and shielding. Ionization chamber readings were taken in-field at 0 cm and out-of-field at 7.5, 10, and 12.5 cm from beam center. Results: Collimation reduced 87% of the out-of-field radiation dose, and the remaining measurable dose was because of internal scatter. The radiation dose contribution from leakage radiation was negligible, as there was not any measurable radiation dose difference when shielding leakage radiation, with P value of .48. Conclusion: These results call into question the clinical utility of routinely shielding out-of-field body parts during fluoroscopy.
RESUME Contexte : Les technologues en fluoroscopie place toujours un ecran de plomb entre les la table de radiographie et les gonades du patient, m^eme si celles-ci ne sont pas directement dans le champ de rayonnement. La diffusion interne est la principale source de radiation pour les parties du corps hors-champ, mais un ecran place entre le patient et la source ne bloquera pas la diffusion interne. Les recherches anterieures dans d’autres champs que la fluoroscopie ont demontre que l’ecranage des fuites de radiation par l’obturateur du collimateur permet une faible reduction de la dose de rayonnement. Le but de cetteetude in-vitro etait de determiner si l’ecranage des fuites de radiation pendant la fluoroscopie pouvait permettre une reduction de la dose de rayonnement. Methodologie : Il s’agit d’une etude comparative in-vitro des doses de rayonnement pour differentes strategies de collimation et d’ecranage durant la fluoroscopie. Les mesures de chambre d’ionisation ont ete prises durant la fluoroscopie d’un bloc d’acrylique avec et sans collimation et ecranage. Les lectures de chambre d’ionisation ont ete prises dans le champ a o cm du centre du faisceau et hors du champ a 7,5, 10 et 12,5 cm du centre du faisceau. Resultats : La collimation a permis de reduire la dose de rayonnement hors-champ de 87 %, la dose mesurable restante etant attribuable a la diffusion interne. La contribution des fuites de radiation a la dose de rayonnement etait negligeable, l’ecranage des fuites de radiation ne montrant aucune difference mesurable dans la dose de rayonnement, avec une valeur p de 0,48. Conclusion : Ces resultats remettent en question l’utilite clinique de l’ecranage de routine des parties du corps hors-champ durant la fluoroscopie.
Keywords: Collimation; dose; shielding; scatter; radiation
Funding sources: None. * Corresponding author: Andrew S. Phelps, MD, Department of Radiology & Biomedical Imaging, Benioff Children’s Hospital C1758N, 1975 4th Street, San Francisco, CA 94158, USA. E-mail address: [email protected] (A.S. Phelps). 1939-8654/$ - see front matter Ó 2016 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jmir.2015.12.082
Introduction Fluoroscopy technologists routinely place a lead shield between the x-ray table and the patient’s gonads, even if the gonads are not directly in the x-ray field. For example, a
technologist might shield the gonads during an esophagram although the x-ray beam is collimated to the chest. The area excluded by collimation is referred to as ‘‘out-of-field.’’ The justification for out-of-field shielding is to block the x-rays that penetrate through the collimator shutters. This source of radiation is referred to as ‘‘leakage radiation.’’ Nonfluoroscopy research has shown that leakage radiation has a small, but measurable contribution to patient radiation dose [1–10]. To the authors’ knowledge, there has not been a study evaluating shielding of leakage radiation in fluoroscopy. The authors hypothesized that shielding leakage radiation during fluoroscopy should reduce radiation dose. There are a number of methods recommended to reduce radiation dose during radiography [11, 12]. Lead shielding is one method to reduce radiation dose to body parts included in the direct x-ray beam [10, 13–21]. The area included in the direct x-ray beam is referred to as ‘‘in-field.’’ Shielding also reduces radiation dose to body parts excluded from direct x-ray beam by using collimation. The area excluded by the direct x-ray beam is referred to as ‘‘out-of-field.’’ However, the radiation dose reduction from shielding out-of-field body parts is small, unless the shield can be curved to block internal scatter originating from in-field body parts [1–9]. An example of this would be shielding the arms from scatter radiation exiting the patient’s torso by placing a shield between the arms and torso. There are four sources of out-of-field radiation: (1) internal scatter within the patient, (2) off-focus radiation from electrons hitting the edges of the anode, (3) leakage radiation through the collimator shutters, and (4) leakage radiation through the tube housing (Figure 1) [22]. Of these sources, internal scatter is the leading contributor to radiation dose [6, 7, 10]. Internal scatter originates from the patient and is not image-forming. In contrast, the three nonscatter radiation sources originate from the x-ray source and are image-forming (Figure 2). Of the three nonscatter radiation sources, leakage
Figure 1. This diagram shows the different sources of out-of-field radiation, which are indicated with shades of gray. The anode is indicated with an asterisk. The ‘‘out-of-field’’ radiation sources are: scatter radiation (1), offfocus radiation (2), collimator leak (3), and tube leak (4).
2
radiation through the collimator shutters contributes the most to radiation dose. Collimator shutters are thin, which allows for higher mobility, but it also allows some of the x-rays to pass through to out-of-field body parts. In contrast, tube housing is very thick, so x-ray leakage from the tube housing is minimal. The contribution to radiation dose from off-focus radiation is also minimal, because it only affects a thin rim at the boundary of the in-field and out-of-field areas [22]. Leakage radiation through collimators results in image formation of out-of-field body parts. An example of this is shown in Figure 2A and B. Figure 2A shows a lower-extremity limb length survey in a 23-month-old female with gait abnormality. The examination in this patient was performed erect with digital radiography in two exposures, lower and upper, which were then stitched together. In Figure 2B, the same stitched radiograph is rewindowed to reveal the patient’s torso with a breast shield present. In this example, the detector was not quite large enough to include all the legs, so a second exposure was required to cover the pelvis and upper femurs. Collimation was used during the second exposure to exclude the torso, yet image formation of the torso still occurred on the detector. This is because of radiation leakage through the collimator shutters. The breast shield blocks image formation from the leakage radiation. Radiology technologists will routinely see the image-forming effects of leakage radiation to out-of-field body parts. Seeing the out-of-field image-formation reinforces a technologist’s desire to shield radiosensitive out-of-field body parts, as was done for the breasts in the clinical example previously mentioned. A technologist sees how a shield can block image formation from leakage radiation and assumes that shielding must also reduce skinentrance dose to the patient. Radiologists, who interpret the radiographic images and report on the radiation dose, may not even know if out-of-field shielding was performed, because the technologists will routinely crop out the out-offield image with postprocessing software [23]. It is important for radiologists to be aware of this shielding practice, as the radiologists are ultimately responsible for the care of patients in the radiology department. Shielding out-of-field body parts has been shown to provide only a small reduction in radiation dose, but it may be beneficial for pediatric patients. Radiation dose reduction is clinically relevant in children because the lifetime risks of ionizing radiation are higher for children than for adults [24–27]. The most conservative approach is to assume any amount of radiation is harmful, and the effects of radiation are cumulative. The principle of ALARA stands for ‘‘as low as reasonably achievable’’ and acknowledges the tradeoff between patient safety and diagnostic accuracy. Although computed tomography contributes the most to overall diagnostic radiation dose [28, 29], fluoroscopy is still a significant source of radiation in children as it is a frequent examination. To the authors’ knowledge, there is not any prior pediatric fluoroscopy research investigating out-of-field shielding, likely the consequence of ethical barriers. The current recommendations for reducing radiation to children during fluoroscopy
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
Figure 2. (A) Lower-extremity limb length survey in a 23-month-old female with gait abnormality. (B) Same stitched radiograph after adjustment of the window and level, revealing image formation of the out-of-field torso with a breast shield present.
focus on operator experience with children and the use of proper fluoroscopic settings, such as low frame rate and low magnification [30]. Also, the recommendations for pediatric gonadal shielding are based on research in pelvic radiography, where gonads are in-field [10, 14, 15, 18, 31]. For institutions serving a large pediatric population and performing many fluoroscopic studies, any reduction in fluoroscopic radiation dose is desirable. The goal of this project was to measure the effect of shielding leakage radiation on air-exposure radiation during fluoroscopy. The study hypothesis was that shielding leakage radiation would reduce airexposure radiation. Materials and Methods This was an in vitro experiment, and animals were not used. For the sake of clinical correlation, a single clinical radiograph of a child was studied and presented in the figures.
This clinical radiograph was obtained in the course of routine clinical care. The institutional review board approved the retrospective evaluation and publication of the image. The materials used included: fluoroscopy unit (AXIOM Luminos; Siemens USA, Washington DC), ionization chamber (Model 1515; Radcal, Monrovia, CA), acrylic block (9.75 25 25 cm), wooden blocks (5 10 30 cm), flexible vinylcovered lead shield (0.5-mm lead thickness). The fluoroscopy source-to-table distance was 66 cm. The ionization chamber volume was 6 mL. The ionization chamber is a gas-filled radiation detector capable of quantifying exposure to x-rays across a wide range of energies. All the equipment was up to date with quality control. The experimental setup is shown in Figure 3. The flexible shield was placed in two positions, flat or curved. The intent of the ‘‘flat shield’’ position was to block leakage radiation (Figure 3A and B). The intent of the ‘‘curved shield’’ position was to block scatter radiation
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
3
Figure 3. Photographs of experimental setup. In (A) and (B) the ionization chamber is shown in the ‘‘in-field’’ position, where it is exposed to the primary beam and centered under the acrylic block. In both (A) and (B) the flexible shield (shown with asterisk) is placed flat on table. In (C) the ionization chamber is shown in the ‘‘out-of-field’’ position, and the flexible shield (shown with asterisk) has been curved up on one side to block scatter traveling from the blocks to the ionization chamber. For the actual acquisitions, the image intensifier was in its lowest position; it was raised in these photos to better show the setup.
originating from the acrylic and wooden blocks (Figure 3C). The speculated geometry of the leakage and scattered radiation in relation to different shield positions is shown in 4
Figure 4. The acrylic block was placed on top of two wooden blocks, to allow space for the ionization chamber. The ionization chamber was positioned on the fluoroscopy tabletop, just
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
Figure 4. This shows the experimental setup superimposed with the speculated scatter radiation geometry. The radiation scatter from the acrylic and wooden blocks can only be blocked with the curved shield.
below the acrylic block. The ionization chamber position was chosen to simulate skin-entrance dose, as x-rays in fluoroscopy originate from beneath the table and course upward. The image intensifier was lowered to its lowest position at 24 cm from tabletop. Ionization chamber measurements were taken in-field at the center of the beam, defined as 0 cm, and out-of-field at 7.5, 10, and 12.5 cm. The out-of-field distances were equally spaced between the outer edge of the wood block and outer most edge of the uncollimated image. The image intensifier was placed in its lowest position to maximize image size. There were two image field dimensions used. The image field dimensions were 11 23 cm when the collimation was adjusted to include the out-of-field ionization chamber positions in the image. This larger image field will be hereforth referred to as ‘‘without collimator.’’ The image field dimensions were 11 7.5 cm when collimation was adjusted to exclude the out-of-field ionization chamber positions from the image. This smaller image field will be hereforth referred to as ‘‘with collimator.’’ There were four possible setup permutations at each of the measurement distances: (1) without collimator or shield, (2) with collimator and without shield, (3) with collimator and flat shield, and (4) with collimator
and curved shield. A total of 16 setup permutations were performed. For each setup permutation, 3 ionization chamber measurements were obtained. Each measurement lasted 10 seconds, and during that time a single fluoroscopic exposure was obtained. In addition, 6 control measurements were obtained, each for 10 seconds, without fluoroscopic exposure. Auto exposure was left on to simulate actual clinical usage. The grid was kept on to simulate higher radiation dose situations. For the described experimental setup, the actual energy profile with collimation was: 81 peak kVp, 54 mA, 9.0 ms, 0.48 mAs. The energy profile without collimation was: 77 kVp, 58 mA, 7.3 ms, 0.42 mAs. Descriptive statistics were performed with averages, standard deviations, and relative ratios. For the sake of simplicity, the ionization chamber measurements are hereforth referred to as ‘‘dose.’’ A corrected dose was calculated by subtracting background dose from each measured dose. To graph the relationships between the different dose reductions, relative ratios were calculated for each permutation. The averaged corrected dose for a permutation was divided by the averaged corrected dose at 0 cm without collimator and without shield. Unpaired two-tailed t-tests were performed comparing unshielded and shielded doses at 7.5, 10, and 12.5 cm. Results When shielding was absent, collimation and increased distance from the center reduced the dose. Background dose was small at 0.41 microgray with standard deviation 0.03 and was subtracted from all the experimental measurements. The dose in the center of the beam increased when the periphery was collimated, from 89.89 to 102.92, which is a difference of þ14% (Table 1). This increase was attributed to autoexposure control from the image intensifier. The dose decreased exponentially with increased distance from the beam center (Figure 5). The single largest dose reduction seen in this study was 87% when collimation was applied (Figures 5 and 6). After collimation, curved shielding reduced the dose further, whereas flat shielding mostly did not. The only significant dose reduction with flat shielding was observed at
Table 1 Effect of Distance, Collimation, and Shielding on Exposure to Ionizing Radiation Dose (microgray),* SD, and Relative Ratiosy
Setup
In-field
Out-of-field (Collimated Area)
0cm
No Yes Yes Yes
collimator, collimator, collimator, collimator,
no shield no shield yes shield (flat) yes shield (curved)
7.5 cm
10 cm
12.5 cm
Dose
SD
Ratio
Dose
SD
Ratio
Dose
SD
Ratio
Dose
SD
Ratio
89.89 102.92 102.52 102.35
0.30 0.21 0.38 0.06
1.000 1.145 1.141 1.139
59.72 9.86 10.08 0.31
0.15 0.22 0.13 0.03
0.660 0.110 0.112 0.003
48.22 5.33 5.26 0.77
0.15 0.04 0.05 0.04
0.536 0.059 0.059 0.009
34.15 3.47 3.09 0.68
0.25 0.05 0.10 0.10
0.380 0.039 0.034 0.008
SD, standard deviation. Exposure (in microgray) was measured from ionization chamber positioned at 0, 7.5, 10, and 12.5 cm away from center of beam (see Figures 3 and 4 for setup). Each dose represents the corrected average of 3 independent 10-second readings. y Each dose is converted into a ratio (bolded) that is defined relative to the 0 cm/no-collimator/no-shield dose (underlined). These ratios are also graphed in Figure 5. *
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
5
Figure 5. Radiation dose reduction from collimation and shielding at different distances from center of beam.
12.5 cm with a 11% dose reduction and P value of .004 (Table 1). Flat shielding did not significantly change the dose at 7.5 or 10 cm, with P values of .2 and .13, respectively. In contrast to the effects of flat shielding, the dose decreases with curved shielding were all large and significant. The dose reductions were 97%, 86%, and 80% at 7.5, 10, and 12.5 cm, respectively. The P value for each comparison was .0001. When averaging the doses from all out-of-field distances, the curved shielding reduced dose by 91% with P value of .0005, whereas the flat shielding reduced dose by 1% with P value of .48 (Figure 6). Discussion After proper collimation during fluoroscopy, the remaining out-of-field radiation dose was mostly from internal scatter. This result is consistent with previous research [1–3, 6–9]. The contribution from leakage radiation was not significant, as shielding of leakage radiation did not result in a significant dose reduction beyond that achieved with collimation alone. The lack of any detectable benefit from shielding the leakage radiation during fluoroscopy supported the null hypothesis. The results from this study are different from prior nonfluoroscopy research that was able to demonstrate a radiation dose reduction when shielding leakage radiation [6–8]. If there is not a benefit to shielding leakage radiation during fluoroscopy, can one shield scatter radiation? In this study, shielding scatter radiation was achieved by curving the shield in between the acrylic block and the ionization chamber.
Figure 6. Dose reduction before and after collimation and shielding.
6
Unfortunately, unlike the simple setup of this in vitro study, there are very few clinical scenarios where one can place a shield in between the in-field and out-of-field body parts. The authors could not find any fluoroscopy literature addressing this problem; however, examples were found in dental radiography and radiation therapy. In dental radiography, a gonad shield can be placed in the patient’s lap to block the scatter radiation exiting the face and reentering the pelvis [5, 9]. In pelvic radiation therapy, spherical gonadal shields have been described [2], but spherical shields are not practical for routine diagnostic studies. For diagnostic fluoroscopy imaging, which uses much smaller radiation doses than radiation therapy, any small radiation dose reduction benefit from a spherical gonad shield is outweighed by patient cooperation and comfort. Fortunately, the radiation dose from scatter and leakage radiation is small, and out-of-field shielding is not the primary strategy to reduce patient radiation dose. More effective means of reducing in-field radiation dose include: minimizing exposure time, imaging the correct body part, and tight collimation [10, 16, 27, 32]. There are multiple limitations in this study: - The main limitation is the narrow scope of the experimental design. The study did not include real patients because standard clinical practice would make it ethically challenging to justify measuring nonshielded radiation doses in patients. - Only a single fluoroscopic unit was used. Computed and digital radiography units, which might have variations in collimator thickness and x-ray spectrum, were not tested. - Only a single radiation detector was used, and a larger ionization chamber or a solid state detector would have been more sensitive to smaller dose reductions. - An anthropomorphic phantom was not used, and the scatter geometry of wooden and acrylic blocks may not reflect in vivo situations. For example, the estimated skin entrance doses were overestimated because of scatter from the wooden blocks. - Only one thickness of lead shielding was used and beam hardened x-rays penetrating the collimator might pass through the shield and ionization chamber more easily than scattered radiation. - The maximum distance from image center was 12.5 cm, which was limited by the size of the image intensifier. In larger patients, both the scattered and leakage radiation dose to out-of-field body parts might be negligible because of the inverse square rule. - Automatic exposure control was used rather than manually controlling tube voltage and current. The use of automatic exposure control reduces experimental reproducibility; however, the intent of this study was to replicate day-to-day fluoroscopic use. Of the eight statistical comparisons performed, three of them had insignificant P values; however, a retrospective power analysis showed that two of these three comparisons were not sufficiently powered to reject the null hypothesis. There could have been a small radiation dose difference
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
when shielding leakage radiation, but the sample size was insufficient to detect the difference. That said, a radiation dose reduction this small is not clinically relevant. Is there any harm to unnecessary shielding? The most likely potential harm is from hospital spread of infection. Radiology cassettes, markers, and shields are all potential fomites, and although cleaning protocols minimize the infectious risk, the risk is not zero [33–37]. Also, previous research with Monte Carlo simulations has shown how out-of-field shielding can increase skin entrance radiation dose [8]. Therefore, if shielding leakage radiation does not provide a significant overall radiation dose reduction benefit, it should not be performed. Conclusions Leakage radiation during fluoroscopy had a negligible effect on out-of-field dose in this in vitro study. This result calls into question the clinical benefit of shielding out-offield body parts. Furthermore, in vivo investigation is warranted to confirm the results of this study. References [1] Hawking, N., & Sharp, T. (2013). Decreasing radiation exposure on pediatric portable chest radiographs. Radiol Technol 85, 9–16. [2] Singhal, M., Kapoor, A., & Singh, D., et al. (2014). Scattered radiation to gonads: role of testicular shielding for para-aortic and homolateral illiac nodal radiotherapy. J Egypt Natl Canc Inst 26, 99–101. [3] Liebmann, M., L€ ullau, T., Kluge, A., Poppe, B., & von Boetticher, H. (2014). Patient radiation protection covers for head CT scans–a clinical evaluation of their effectiveness. Rofo 186(11), 1022–1027. [4] Barcham, N., Egan, I., & Dowd, S. (1997). Gonadal protection methods in neonatal chest radiography. Radiol Technol 69, 157–161. [5] Stenstr€om, B., Rehnmark-Larsson, S., Julin, P., & Richter, S. (1983). Radiation shielding in dental radiography. Swed Dent J 7, 85–91. [6] Platten, C., Purfield, D., & Fife, I. (2004). Investigating the effect of patient lead shielding on radiation dose when placed outside the primary beam. 11th International Congress of the International Radiation Protection Association. [7] Daniels, C., & Furey, E. (2008). The effectiveness of surface lead shielding of gonads outside the primary x-ray beam. J Med Imaging Radiat Sci 39, 189–191. [8] Matyagin Y, Collins P, Ruwoldt S, Chew S, West J. Effectiveness of gonad shields: a Monte Carlo evaluation. Royal Australian and New Zealand College of Radiologists Combined Science Meeting 2014. [9] Wood, R., Harris, A., van der Merwe, E., & Nortje, C. (1991). The leaded apron revisited: does it reduce gonadal radiation dose in dental radiology? Oral Surg Oral Med Oral Pathol 71, 642–646. [10] Slovis, T., & Strauss, K. (2013). Gonadal shielding for neonates. Pediatr Radiol 43, 1265–1266. [11] Frush, D. (2011). Radiation, thoracic imaging, and children: radiation safety. Radiol Clin North Am 49, 1053–1069. [12] Andersen, P. J., Andersen, P., & van der Kooy, P. (1982). Dose reduction in radiography of the spine in scoliosis. Acta Radiol Diagn (Stockh) 23, 251–253. [13] Nguyen, K., Schlaifer, A., & Smith, D., et al. (2012). In automated fluoroscopy settings, does shielding affect radiation exposure to surrounding unshielded tissues? J Endourol 26, 1489–1493. [14] Webster, E., & Merrill, O. (1957). Measurements of gonadal dose in radiographic examinations. N Engl J Med 257, 811–819. [15] Bishop, H., Webber, M., & O’Loughlin, B. (1959). Reducing gonad irradiation in pediatric diagnosis. Calif Med 90, 20–25.
[16] Don, S., Macdougall, R., & Strauss, K., et al. (2013). Image gently campaign back to basics initiative: ten steps to help manage radiation dose in pediatric digital radiography. AJR Am J Roentgenol 200, 431– 436. [17] Hohl, C., Mahnken, A., & Klotz, E., et al. (2005). Radiation dose reduction to the male gonads during MDCT: the effectiveness of a lead shield. AJR Am J Roentgenol 184, 128–130. [18] Winfeld, M., Strubel, N., & Pinkney, L., et al. (2013). Relative distribution of pertinent findings on portable neonatal abdominal radiographs: can we shield the gonads? Pediatr Radiol 43, 1295–1302. [19] Frantzen, M., Robben, S., Postma, A., Zoetelief, J., Wildberger, J., & Kemerink, G. (2012). Gonad shielding in paediatric pelvic radiography: disadvantages prevail over benefit. Insights Imaging 3, 23–32. [20] Makri, T., Yakoumakis, E., & Papadopoulou, D., et al. (2006). Radiation risk assessment in neonatal radiographic examinations of the chest and abdomen: a clinical and Monte Carlo dosimetry study. Phys Med Biol 51, 5023–5033. [21] Elbakri, I. (2013). Estimation of dose-area product-to-effective dose conversion factors for neonatal radiography using PCXMC. Radiat Prot Dosimetry 158, 43–50. [22] Bushberg, J., Seibert, J., Leidholdt, E., & Boone, J. (2011). The Essential Physics of Medical Imaging, (3rd ed.). Lippincott Williams & Wilkins. [23] Morrison, G., John, S., & Goske, M., et al. (2011). Pediatric digital radiography education for radiologic technologists: current state. Pediatr Radiol 41, 602–610. [24] Journy, N., Ancelet, S., & Rehel, J., et al. (2014). Predicted cancer risks induced by computed tomography examinations during childhood, by a quantitative risk assessment approach. Radiat Environ Biophys 53, 39–54. [25] Baysson, H., Etard, C., Brisse, H., & Bernier, M. (2012). Diagnostic radiation exposure in children and cancer risk: current knowledge and perspectives. Arch Pediatr 19, 64–73. [26] Bernier, M., Rehel, J., & Brisse, H., et al. (2012). Radiation exposure from CT in early childhood: a French large-scale multicentre study. Br J Radiol 85, 53–60. [27] Goske, M., Frush, D., Brink, J., Kaste, S., Butler, P., & Pandharipande, P. (2014). Curbing potential radiation-induced cancer risks in oncologic imaging: perspectives from the ‘image gently’ and ‘image wisely’ campaigns. Oncology (Williston Park) 28, 232–238. [28] Wiest, P., Locken, J., Heintz, P., & Mettler, F. J. (2002). CT scanning: a major source of radiation exposure. Semin Ultrasound CT MR 23, 402–410. [29] Brenner, D., & Hall, E. (2007). Computed tomography–an increasing source of radiation exposure. N Engl J Med 357, 2277–2284. [30] Strauss, K., & Kaste, S. (2006). The ALARA (as low as reasonably achievable) concept in pediatric interventional and fluoroscopic imaging: striving to keep radiation doses as low as possible during fluoroscopy of pediatric patients–a white paper executive summary. Radiology 240, 621–622. [31] Sikand, M., Stinchcombe, S., & Livesley, P. (2003). Study on the use of gonadal protection shields during paediatric pelvic X-rays. Ann R Coll Surg Engl 85, 422–425. [32] Soboleski, D., Theriault, C., Acker, A., Dagnone, V., & Manson, D. (2006). Unnecessary irradiation to non-thoracic structures during pediatric chest radiography. Pediatr Radiol 36, 22–25. [33] Hodges, A. (2001). Radiographic markers: friend or fomite? Radiol Technol 73, 183–185. [34] Grogan, B., Cranston, W., Lopez, D., Furbee, C., Murray, C., & Hsu, J.Skeletal Trauma Research Consortium (2011). Do protective lead garments harbor harmful bacteria? Orthopedics 34, 765–767. [35] Kim, J., Kim, H., & Park, J., et al. (2012). Contamination of X-ray cassettes with methicillin-resistant Staphylococcus aureus and methicillinresistant Staphylococcus haemolyticus in a radiology department. Ann Lab Med 32, 206–209. [36] Haskin, M., & McGinley, J. (1972). Hospital ecology and nosocomial infections. A systems analysis of fomites with infection potential and vectors of microbial cross contamination in hospital radiology departments. Radiol Clin North Am 10, 583–588. [37] Layer, P. (1998). Disposable shielding for neonates. Radiol Technol 70, 125.
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
7
Journal of Medical Imaging and Radiation Sciences xx (2016) 1-7
Journal de l’imagerie médicale et des sciences de la radiation
www.elsevier.com/locate/jmir
How Much Does Lead Shielding during Fluoroscopy Reduce Radiation Dose to Out-of-Field Body Parts? Andrew S. Phelps, MD*, Robert G. Gould, ScD, Jesse L. Courtier, MD, Peter A. Marcovici, MD, Christina Salani, RT(R)(ARRT) and John D. MacKenzie, MD Department of Radiology & Biomedical Imaging, UCSF Benioff Children’s Hospital, University of California, San Francisco, California, USA
ABSTRACT Background: Fluoroscopy technologists routinely place a lead shield between the x-ray table and the patient’s gonads, even if the gonads are not directly in the x-ray field. Internal scatter radiation is the greatest source of radiation to out-of-field body parts, but a shield placed between the patient and the x-ray source will not block internal scatter. Prior nonfluoroscopy research has shown that there is a small reduction in radiation dose when shielding the leakage radiation that penetrates through the collimator shutters. The goal of this in vitro study was to determine if there was any radiation dose reduction when shielding leakage radiation during fluoroscopy. Methods: This was an in vitro comparison study of radiation doses using different collimation and shielding strategies during fluoroscopy. Ionization chamber measurements were obtained during fluoroscopy of an acrylic block with and without collimation and shielding. Ionization chamber readings were taken in-field at 0 cm and out-of-field at 7.5, 10, and 12.5 cm from beam center. Results: Collimation reduced 87% of the out-of-field radiation dose, and the remaining measurable dose was because of internal scatter. The radiation dose contribution from leakage radiation was negligible, as there was not any measurable radiation dose difference when shielding leakage radiation, with P value of .48. Conclusion: These results call into question the clinical utility of routinely shielding out-of-field body parts during fluoroscopy.
RESUME Contexte : Les technologues en fluoroscopie place toujours un ecran de plomb entre les la table de radiographie et les gonades du patient, m^eme si celles-ci ne sont pas directement dans le champ de rayonnement. La diffusion interne est la principale source de radiation pour les parties du corps hors-champ, mais un ecran place entre le patient et la source ne bloquera pas la diffusion interne. Les recherches anterieures dans d’autres champs que la fluoroscopie ont demontre que l’ecranage des fuites de radiation par l’obturateur du collimateur permet une faible reduction de la dose de rayonnement. Le but de cetteetude in-vitro etait de determiner si l’ecranage des fuites de radiation pendant la fluoroscopie pouvait permettre une reduction de la dose de rayonnement. Methodologie : Il s’agit d’une etude comparative in-vitro des doses de rayonnement pour differentes strategies de collimation et d’ecranage durant la fluoroscopie. Les mesures de chambre d’ionisation ont ete prises durant la fluoroscopie d’un bloc d’acrylique avec et sans collimation et ecranage. Les lectures de chambre d’ionisation ont ete prises dans le champ a o cm du centre du faisceau et hors du champ a 7,5, 10 et 12,5 cm du centre du faisceau. Resultats : La collimation a permis de reduire la dose de rayonnement hors-champ de 87 %, la dose mesurable restante etant attribuable a la diffusion interne. La contribution des fuites de radiation a la dose de rayonnement etait negligeable, l’ecranage des fuites de radiation ne montrant aucune difference mesurable dans la dose de rayonnement, avec une valeur p de 0,48. Conclusion : Ces resultats remettent en question l’utilite clinique de l’ecranage de routine des parties du corps hors-champ durant la fluoroscopie.
Keywords: Collimation; dose; shielding; scatter; radiation
Funding sources: None. * Corresponding author: Andrew S. Phelps, MD, Department of Radiology & Biomedical Imaging, Benioff Children’s Hospital C1758N, 1975 4th Street, San Francisco, CA 94158, USA. E-mail address: [email protected] (A.S. Phelps). 1939-8654/$ - see front matter Ó 2016 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jmir.2015.12.082
Introduction Fluoroscopy technologists routinely place a lead shield between the x-ray table and the patient’s gonads, even if the gonads are not directly in the x-ray field. For example, a
technologist might shield the gonads during an esophagram although the x-ray beam is collimated to the chest. The area excluded by collimation is referred to as ‘‘out-of-field.’’ The justification for out-of-field shielding is to block the x-rays that penetrate through the collimator shutters. This source of radiation is referred to as ‘‘leakage radiation.’’ Nonfluoroscopy research has shown that leakage radiation has a small, but measurable contribution to patient radiation dose [1–10]. To the authors’ knowledge, there has not been a study evaluating shielding of leakage radiation in fluoroscopy. The authors hypothesized that shielding leakage radiation during fluoroscopy should reduce radiation dose. There are a number of methods recommended to reduce radiation dose during radiography [11, 12]. Lead shielding is one method to reduce radiation dose to body parts included in the direct x-ray beam [10, 13–21]. The area included in the direct x-ray beam is referred to as ‘‘in-field.’’ Shielding also reduces radiation dose to body parts excluded from direct x-ray beam by using collimation. The area excluded by the direct x-ray beam is referred to as ‘‘out-of-field.’’ However, the radiation dose reduction from shielding out-of-field body parts is small, unless the shield can be curved to block internal scatter originating from in-field body parts [1–9]. An example of this would be shielding the arms from scatter radiation exiting the patient’s torso by placing a shield between the arms and torso. There are four sources of out-of-field radiation: (1) internal scatter within the patient, (2) off-focus radiation from electrons hitting the edges of the anode, (3) leakage radiation through the collimator shutters, and (4) leakage radiation through the tube housing (Figure 1) [22]. Of these sources, internal scatter is the leading contributor to radiation dose [6, 7, 10]. Internal scatter originates from the patient and is not image-forming. In contrast, the three nonscatter radiation sources originate from the x-ray source and are image-forming (Figure 2). Of the three nonscatter radiation sources, leakage
Figure 1. This diagram shows the different sources of out-of-field radiation, which are indicated with shades of gray. The anode is indicated with an asterisk. The ‘‘out-of-field’’ radiation sources are: scatter radiation (1), offfocus radiation (2), collimator leak (3), and tube leak (4).
2
radiation through the collimator shutters contributes the most to radiation dose. Collimator shutters are thin, which allows for higher mobility, but it also allows some of the x-rays to pass through to out-of-field body parts. In contrast, tube housing is very thick, so x-ray leakage from the tube housing is minimal. The contribution to radiation dose from off-focus radiation is also minimal, because it only affects a thin rim at the boundary of the in-field and out-of-field areas [22]. Leakage radiation through collimators results in image formation of out-of-field body parts. An example of this is shown in Figure 2A and B. Figure 2A shows a lower-extremity limb length survey in a 23-month-old female with gait abnormality. The examination in this patient was performed erect with digital radiography in two exposures, lower and upper, which were then stitched together. In Figure 2B, the same stitched radiograph is rewindowed to reveal the patient’s torso with a breast shield present. In this example, the detector was not quite large enough to include all the legs, so a second exposure was required to cover the pelvis and upper femurs. Collimation was used during the second exposure to exclude the torso, yet image formation of the torso still occurred on the detector. This is because of radiation leakage through the collimator shutters. The breast shield blocks image formation from the leakage radiation. Radiology technologists will routinely see the image-forming effects of leakage radiation to out-of-field body parts. Seeing the out-of-field image-formation reinforces a technologist’s desire to shield radiosensitive out-of-field body parts, as was done for the breasts in the clinical example previously mentioned. A technologist sees how a shield can block image formation from leakage radiation and assumes that shielding must also reduce skinentrance dose to the patient. Radiologists, who interpret the radiographic images and report on the radiation dose, may not even know if out-of-field shielding was performed, because the technologists will routinely crop out the out-offield image with postprocessing software [23]. It is important for radiologists to be aware of this shielding practice, as the radiologists are ultimately responsible for the care of patients in the radiology department. Shielding out-of-field body parts has been shown to provide only a small reduction in radiation dose, but it may be beneficial for pediatric patients. Radiation dose reduction is clinically relevant in children because the lifetime risks of ionizing radiation are higher for children than for adults [24–27]. The most conservative approach is to assume any amount of radiation is harmful, and the effects of radiation are cumulative. The principle of ALARA stands for ‘‘as low as reasonably achievable’’ and acknowledges the tradeoff between patient safety and diagnostic accuracy. Although computed tomography contributes the most to overall diagnostic radiation dose [28, 29], fluoroscopy is still a significant source of radiation in children as it is a frequent examination. To the authors’ knowledge, there is not any prior pediatric fluoroscopy research investigating out-of-field shielding, likely the consequence of ethical barriers. The current recommendations for reducing radiation to children during fluoroscopy
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
Figure 2. (A) Lower-extremity limb length survey in a 23-month-old female with gait abnormality. (B) Same stitched radiograph after adjustment of the window and level, revealing image formation of the out-of-field torso with a breast shield present.
focus on operator experience with children and the use of proper fluoroscopic settings, such as low frame rate and low magnification [30]. Also, the recommendations for pediatric gonadal shielding are based on research in pelvic radiography, where gonads are in-field [10, 14, 15, 18, 31]. For institutions serving a large pediatric population and performing many fluoroscopic studies, any reduction in fluoroscopic radiation dose is desirable. The goal of this project was to measure the effect of shielding leakage radiation on air-exposure radiation during fluoroscopy. The study hypothesis was that shielding leakage radiation would reduce airexposure radiation. Materials and Methods This was an in vitro experiment, and animals were not used. For the sake of clinical correlation, a single clinical radiograph of a child was studied and presented in the figures.
This clinical radiograph was obtained in the course of routine clinical care. The institutional review board approved the retrospective evaluation and publication of the image. The materials used included: fluoroscopy unit (AXIOM Luminos; Siemens USA, Washington DC), ionization chamber (Model 1515; Radcal, Monrovia, CA), acrylic block (9.75 25 25 cm), wooden blocks (5 10 30 cm), flexible vinylcovered lead shield (0.5-mm lead thickness). The fluoroscopy source-to-table distance was 66 cm. The ionization chamber volume was 6 mL. The ionization chamber is a gas-filled radiation detector capable of quantifying exposure to x-rays across a wide range of energies. All the equipment was up to date with quality control. The experimental setup is shown in Figure 3. The flexible shield was placed in two positions, flat or curved. The intent of the ‘‘flat shield’’ position was to block leakage radiation (Figure 3A and B). The intent of the ‘‘curved shield’’ position was to block scatter radiation
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
3
Figure 3. Photographs of experimental setup. In (A) and (B) the ionization chamber is shown in the ‘‘in-field’’ position, where it is exposed to the primary beam and centered under the acrylic block. In both (A) and (B) the flexible shield (shown with asterisk) is placed flat on table. In (C) the ionization chamber is shown in the ‘‘out-of-field’’ position, and the flexible shield (shown with asterisk) has been curved up on one side to block scatter traveling from the blocks to the ionization chamber. For the actual acquisitions, the image intensifier was in its lowest position; it was raised in these photos to better show the setup.
originating from the acrylic and wooden blocks (Figure 3C). The speculated geometry of the leakage and scattered radiation in relation to different shield positions is shown in 4
Figure 4. The acrylic block was placed on top of two wooden blocks, to allow space for the ionization chamber. The ionization chamber was positioned on the fluoroscopy tabletop, just
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
Figure 4. This shows the experimental setup superimposed with the speculated scatter radiation geometry. The radiation scatter from the acrylic and wooden blocks can only be blocked with the curved shield.
below the acrylic block. The ionization chamber position was chosen to simulate skin-entrance dose, as x-rays in fluoroscopy originate from beneath the table and course upward. The image intensifier was lowered to its lowest position at 24 cm from tabletop. Ionization chamber measurements were taken in-field at the center of the beam, defined as 0 cm, and out-of-field at 7.5, 10, and 12.5 cm. The out-of-field distances were equally spaced between the outer edge of the wood block and outer most edge of the uncollimated image. The image intensifier was placed in its lowest position to maximize image size. There were two image field dimensions used. The image field dimensions were 11 23 cm when the collimation was adjusted to include the out-of-field ionization chamber positions in the image. This larger image field will be hereforth referred to as ‘‘without collimator.’’ The image field dimensions were 11 7.5 cm when collimation was adjusted to exclude the out-of-field ionization chamber positions from the image. This smaller image field will be hereforth referred to as ‘‘with collimator.’’ There were four possible setup permutations at each of the measurement distances: (1) without collimator or shield, (2) with collimator and without shield, (3) with collimator and flat shield, and (4) with collimator
and curved shield. A total of 16 setup permutations were performed. For each setup permutation, 3 ionization chamber measurements were obtained. Each measurement lasted 10 seconds, and during that time a single fluoroscopic exposure was obtained. In addition, 6 control measurements were obtained, each for 10 seconds, without fluoroscopic exposure. Auto exposure was left on to simulate actual clinical usage. The grid was kept on to simulate higher radiation dose situations. For the described experimental setup, the actual energy profile with collimation was: 81 peak kVp, 54 mA, 9.0 ms, 0.48 mAs. The energy profile without collimation was: 77 kVp, 58 mA, 7.3 ms, 0.42 mAs. Descriptive statistics were performed with averages, standard deviations, and relative ratios. For the sake of simplicity, the ionization chamber measurements are hereforth referred to as ‘‘dose.’’ A corrected dose was calculated by subtracting background dose from each measured dose. To graph the relationships between the different dose reductions, relative ratios were calculated for each permutation. The averaged corrected dose for a permutation was divided by the averaged corrected dose at 0 cm without collimator and without shield. Unpaired two-tailed t-tests were performed comparing unshielded and shielded doses at 7.5, 10, and 12.5 cm. Results When shielding was absent, collimation and increased distance from the center reduced the dose. Background dose was small at 0.41 microgray with standard deviation 0.03 and was subtracted from all the experimental measurements. The dose in the center of the beam increased when the periphery was collimated, from 89.89 to 102.92, which is a difference of þ14% (Table 1). This increase was attributed to autoexposure control from the image intensifier. The dose decreased exponentially with increased distance from the beam center (Figure 5). The single largest dose reduction seen in this study was 87% when collimation was applied (Figures 5 and 6). After collimation, curved shielding reduced the dose further, whereas flat shielding mostly did not. The only significant dose reduction with flat shielding was observed at
Table 1 Effect of Distance, Collimation, and Shielding on Exposure to Ionizing Radiation Dose (microgray),* SD, and Relative Ratiosy
Setup
In-field
Out-of-field (Collimated Area)
0cm
No Yes Yes Yes
collimator, collimator, collimator, collimator,
no shield no shield yes shield (flat) yes shield (curved)
7.5 cm
10 cm
12.5 cm
Dose
SD
Ratio
Dose
SD
Ratio
Dose
SD
Ratio
Dose
SD
Ratio
89.89 102.92 102.52 102.35
0.30 0.21 0.38 0.06
1.000 1.145 1.141 1.139
59.72 9.86 10.08 0.31
0.15 0.22 0.13 0.03
0.660 0.110 0.112 0.003
48.22 5.33 5.26 0.77
0.15 0.04 0.05 0.04
0.536 0.059 0.059 0.009
34.15 3.47 3.09 0.68
0.25 0.05 0.10 0.10
0.380 0.039 0.034 0.008
SD, standard deviation. Exposure (in microgray) was measured from ionization chamber positioned at 0, 7.5, 10, and 12.5 cm away from center of beam (see Figures 3 and 4 for setup). Each dose represents the corrected average of 3 independent 10-second readings. y Each dose is converted into a ratio (bolded) that is defined relative to the 0 cm/no-collimator/no-shield dose (underlined). These ratios are also graphed in Figure 5. *
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
5
Figure 5. Radiation dose reduction from collimation and shielding at different distances from center of beam.
12.5 cm with a 11% dose reduction and P value of .004 (Table 1). Flat shielding did not significantly change the dose at 7.5 or 10 cm, with P values of .2 and .13, respectively. In contrast to the effects of flat shielding, the dose decreases with curved shielding were all large and significant. The dose reductions were 97%, 86%, and 80% at 7.5, 10, and 12.5 cm, respectively. The P value for each comparison was .0001. When averaging the doses from all out-of-field distances, the curved shielding reduced dose by 91% with P value of .0005, whereas the flat shielding reduced dose by 1% with P value of .48 (Figure 6). Discussion After proper collimation during fluoroscopy, the remaining out-of-field radiation dose was mostly from internal scatter. This result is consistent with previous research [1–3, 6–9]. The contribution from leakage radiation was not significant, as shielding of leakage radiation did not result in a significant dose reduction beyond that achieved with collimation alone. The lack of any detectable benefit from shielding the leakage radiation during fluoroscopy supported the null hypothesis. The results from this study are different from prior nonfluoroscopy research that was able to demonstrate a radiation dose reduction when shielding leakage radiation [6–8]. If there is not a benefit to shielding leakage radiation during fluoroscopy, can one shield scatter radiation? In this study, shielding scatter radiation was achieved by curving the shield in between the acrylic block and the ionization chamber.
Figure 6. Dose reduction before and after collimation and shielding.
6
Unfortunately, unlike the simple setup of this in vitro study, there are very few clinical scenarios where one can place a shield in between the in-field and out-of-field body parts. The authors could not find any fluoroscopy literature addressing this problem; however, examples were found in dental radiography and radiation therapy. In dental radiography, a gonad shield can be placed in the patient’s lap to block the scatter radiation exiting the face and reentering the pelvis [5, 9]. In pelvic radiation therapy, spherical gonadal shields have been described [2], but spherical shields are not practical for routine diagnostic studies. For diagnostic fluoroscopy imaging, which uses much smaller radiation doses than radiation therapy, any small radiation dose reduction benefit from a spherical gonad shield is outweighed by patient cooperation and comfort. Fortunately, the radiation dose from scatter and leakage radiation is small, and out-of-field shielding is not the primary strategy to reduce patient radiation dose. More effective means of reducing in-field radiation dose include: minimizing exposure time, imaging the correct body part, and tight collimation [10, 16, 27, 32]. There are multiple limitations in this study: - The main limitation is the narrow scope of the experimental design. The study did not include real patients because standard clinical practice would make it ethically challenging to justify measuring nonshielded radiation doses in patients. - Only a single fluoroscopic unit was used. Computed and digital radiography units, which might have variations in collimator thickness and x-ray spectrum, were not tested. - Only a single radiation detector was used, and a larger ionization chamber or a solid state detector would have been more sensitive to smaller dose reductions. - An anthropomorphic phantom was not used, and the scatter geometry of wooden and acrylic blocks may not reflect in vivo situations. For example, the estimated skin entrance doses were overestimated because of scatter from the wooden blocks. - Only one thickness of lead shielding was used and beam hardened x-rays penetrating the collimator might pass through the shield and ionization chamber more easily than scattered radiation. - The maximum distance from image center was 12.5 cm, which was limited by the size of the image intensifier. In larger patients, both the scattered and leakage radiation dose to out-of-field body parts might be negligible because of the inverse square rule. - Automatic exposure control was used rather than manually controlling tube voltage and current. The use of automatic exposure control reduces experimental reproducibility; however, the intent of this study was to replicate day-to-day fluoroscopic use. Of the eight statistical comparisons performed, three of them had insignificant P values; however, a retrospective power analysis showed that two of these three comparisons were not sufficiently powered to reject the null hypothesis. There could have been a small radiation dose difference
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
when shielding leakage radiation, but the sample size was insufficient to detect the difference. That said, a radiation dose reduction this small is not clinically relevant. Is there any harm to unnecessary shielding? The most likely potential harm is from hospital spread of infection. Radiology cassettes, markers, and shields are all potential fomites, and although cleaning protocols minimize the infectious risk, the risk is not zero [33–37]. Also, previous research with Monte Carlo simulations has shown how out-of-field shielding can increase skin entrance radiation dose [8]. Therefore, if shielding leakage radiation does not provide a significant overall radiation dose reduction benefit, it should not be performed. Conclusions Leakage radiation during fluoroscopy had a negligible effect on out-of-field dose in this in vitro study. This result calls into question the clinical benefit of shielding out-offield body parts. Furthermore, in vivo investigation is warranted to confirm the results of this study. References [1] Hawking, N., & Sharp, T. (2013). Decreasing radiation exposure on pediatric portable chest radiographs. Radiol Technol 85, 9–16. [2] Singhal, M., Kapoor, A., & Singh, D., et al. (2014). Scattered radiation to gonads: role of testicular shielding for para-aortic and homolateral illiac nodal radiotherapy. J Egypt Natl Canc Inst 26, 99–101. [3] Liebmann, M., L€ ullau, T., Kluge, A., Poppe, B., & von Boetticher, H. (2014). Patient radiation protection covers for head CT scans–a clinical evaluation of their effectiveness. Rofo 186(11), 1022–1027. [4] Barcham, N., Egan, I., & Dowd, S. (1997). Gonadal protection methods in neonatal chest radiography. Radiol Technol 69, 157–161. [5] Stenstr€om, B., Rehnmark-Larsson, S., Julin, P., & Richter, S. (1983). Radiation shielding in dental radiography. Swed Dent J 7, 85–91. [6] Platten, C., Purfield, D., & Fife, I. (2004). Investigating the effect of patient lead shielding on radiation dose when placed outside the primary beam. 11th International Congress of the International Radiation Protection Association. [7] Daniels, C., & Furey, E. (2008). The effectiveness of surface lead shielding of gonads outside the primary x-ray beam. J Med Imaging Radiat Sci 39, 189–191. [8] Matyagin Y, Collins P, Ruwoldt S, Chew S, West J. Effectiveness of gonad shields: a Monte Carlo evaluation. Royal Australian and New Zealand College of Radiologists Combined Science Meeting 2014. [9] Wood, R., Harris, A., van der Merwe, E., & Nortje, C. (1991). The leaded apron revisited: does it reduce gonadal radiation dose in dental radiology? Oral Surg Oral Med Oral Pathol 71, 642–646. [10] Slovis, T., & Strauss, K. (2013). Gonadal shielding for neonates. Pediatr Radiol 43, 1265–1266. [11] Frush, D. (2011). Radiation, thoracic imaging, and children: radiation safety. Radiol Clin North Am 49, 1053–1069. [12] Andersen, P. J., Andersen, P., & van der Kooy, P. (1982). Dose reduction in radiography of the spine in scoliosis. Acta Radiol Diagn (Stockh) 23, 251–253. [13] Nguyen, K., Schlaifer, A., & Smith, D., et al. (2012). In automated fluoroscopy settings, does shielding affect radiation exposure to surrounding unshielded tissues? J Endourol 26, 1489–1493. [14] Webster, E., & Merrill, O. (1957). Measurements of gonadal dose in radiographic examinations. N Engl J Med 257, 811–819. [15] Bishop, H., Webber, M., & O’Loughlin, B. (1959). Reducing gonad irradiation in pediatric diagnosis. Calif Med 90, 20–25.
[16] Don, S., Macdougall, R., & Strauss, K., et al. (2013). Image gently campaign back to basics initiative: ten steps to help manage radiation dose in pediatric digital radiography. AJR Am J Roentgenol 200, 431– 436. [17] Hohl, C., Mahnken, A., & Klotz, E., et al. (2005). Radiation dose reduction to the male gonads during MDCT: the effectiveness of a lead shield. AJR Am J Roentgenol 184, 128–130. [18] Winfeld, M., Strubel, N., & Pinkney, L., et al. (2013). Relative distribution of pertinent findings on portable neonatal abdominal radiographs: can we shield the gonads? Pediatr Radiol 43, 1295–1302. [19] Frantzen, M., Robben, S., Postma, A., Zoetelief, J., Wildberger, J., & Kemerink, G. (2012). Gonad shielding in paediatric pelvic radiography: disadvantages prevail over benefit. Insights Imaging 3, 23–32. [20] Makri, T., Yakoumakis, E., & Papadopoulou, D., et al. (2006). Radiation risk assessment in neonatal radiographic examinations of the chest and abdomen: a clinical and Monte Carlo dosimetry study. Phys Med Biol 51, 5023–5033. [21] Elbakri, I. (2013). Estimation of dose-area product-to-effective dose conversion factors for neonatal radiography using PCXMC. Radiat Prot Dosimetry 158, 43–50. [22] Bushberg, J., Seibert, J., Leidholdt, E., & Boone, J. (2011). The Essential Physics of Medical Imaging, (3rd ed.). Lippincott Williams & Wilkins. [23] Morrison, G., John, S., & Goske, M., et al. (2011). Pediatric digital radiography education for radiologic technologists: current state. Pediatr Radiol 41, 602–610. [24] Journy, N., Ancelet, S., & Rehel, J., et al. (2014). Predicted cancer risks induced by computed tomography examinations during childhood, by a quantitative risk assessment approach. Radiat Environ Biophys 53, 39–54. [25] Baysson, H., Etard, C., Brisse, H., & Bernier, M. (2012). Diagnostic radiation exposure in children and cancer risk: current knowledge and perspectives. Arch Pediatr 19, 64–73. [26] Bernier, M., Rehel, J., & Brisse, H., et al. (2012). Radiation exposure from CT in early childhood: a French large-scale multicentre study. Br J Radiol 85, 53–60. [27] Goske, M., Frush, D., Brink, J., Kaste, S., Butler, P., & Pandharipande, P. (2014). Curbing potential radiation-induced cancer risks in oncologic imaging: perspectives from the ‘image gently’ and ‘image wisely’ campaigns. Oncology (Williston Park) 28, 232–238. [28] Wiest, P., Locken, J., Heintz, P., & Mettler, F. J. (2002). CT scanning: a major source of radiation exposure. Semin Ultrasound CT MR 23, 402–410. [29] Brenner, D., & Hall, E. (2007). Computed tomography–an increasing source of radiation exposure. N Engl J Med 357, 2277–2284. [30] Strauss, K., & Kaste, S. (2006). The ALARA (as low as reasonably achievable) concept in pediatric interventional and fluoroscopic imaging: striving to keep radiation doses as low as possible during fluoroscopy of pediatric patients–a white paper executive summary. Radiology 240, 621–622. [31] Sikand, M., Stinchcombe, S., & Livesley, P. (2003). Study on the use of gonadal protection shields during paediatric pelvic X-rays. Ann R Coll Surg Engl 85, 422–425. [32] Soboleski, D., Theriault, C., Acker, A., Dagnone, V., & Manson, D. (2006). Unnecessary irradiation to non-thoracic structures during pediatric chest radiography. Pediatr Radiol 36, 22–25. [33] Hodges, A. (2001). Radiographic markers: friend or fomite? Radiol Technol 73, 183–185. [34] Grogan, B., Cranston, W., Lopez, D., Furbee, C., Murray, C., & Hsu, J.Skeletal Trauma Research Consortium (2011). Do protective lead garments harbor harmful bacteria? Orthopedics 34, 765–767. [35] Kim, J., Kim, H., & Park, J., et al. (2012). Contamination of X-ray cassettes with methicillin-resistant Staphylococcus aureus and methicillinresistant Staphylococcus haemolyticus in a radiology department. Ann Lab Med 32, 206–209. [36] Haskin, M., & McGinley, J. (1972). Hospital ecology and nosocomial infections. A systems analysis of fomites with infection potential and vectors of microbial cross contamination in hospital radiology departments. Radiol Clin North Am 10, 583–588. [37] Layer, P. (1998). Disposable shielding for neonates. Radiol Technol 70, 125.
A.S. Phelps et al./Journal of Medical Imaging and Radiation Sciences
-
(2016) 1-7
7