| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Full Paper |
Department of 1 Medical Physics and 2 Radiology, Medical School, Aristotelian University of Thessaloniki, Thessaloniki 54124, Greece
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
In Europe, and subsequently in Greece, it is a legal requirement to keep the radiation dose "as low as reasonably achievable", giving consideration to social and economic factors (the ALARA principle).
For children the probability of late effects occurring from radiation exposure is 2 to 3 times higher than for adults [1]. Tissue sensitivity in children is higher and the probability that there may be late radiation effects is also higher, as a consequence of the longer life expectation. Therefore, the attention given to radiation protection and quality assurance in paediatric radiology has been increased. The European Commission has published a report for quality assurance and reference doses from some simple radiographic paediatric examinations [2] but dose values for complex examinations such as micturating cystourethrography (MCU) are not cited.
The purpose of this study was to determine the effective dose and to estimate the risk from the MCU paediatric radiological examination. The quantity "effective dose" which is a weighted sum of mean absorbed doses in selected organs, was calculated. Effective dose can be calculated from entrance dose conversion factors (Cf). Such factors have been calculated for adults elsewhere [3–5] but they cannot be applied in paediatric radiology, because children are smaller. Entrance surface dose (ESD) can be measured with thermoluminescent dosimeters (TLDs) or with dose–area product (DAP). Cf values for paediatric MCU examinations, using DAP entrance dose have been calculated [6] and the National Radiological Protection Board (NRPB) has computed extensive Cf data for a set of commonly used paediatric radiographs [7]. However these factors are exposure specific and cannot be applied to this work.
In the present study Monte Carlo simulations of MCU examinations were performed for each of the patients using the appropriate mathematical phantoms. TLDs were used to measure ESD. Absorbed organ doses were calculated for separate radiological views as well as conversion factors for each gender and age group and the effective dose for boys and girls for each age group.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
According to the protocol, contrast medium was inserted into the urinary bladder via the urethra. At the beginning of the MCU examination, a brief fluoroscopic view was obtained in order to see the position of the urinary bladder. An anteroposterior (AP) image was obtained when the urinary bladder was full of contrast medium, followed by two oblique (OBL) views. The last view was OBL for male patients or AP for female patients and was obtained during micturition. Additional views could also be obtained if the radiologist in charge asked for them. All the radiological views were taken with the cassette inside the bucky. The oblique views were set at an angle of 45° right and 45° left of an axis normal to the radiological table, while the patient remained supine. Upper extremities were always outside the radiation field.
For each child, besides the age, external physical characteristics were also recorded, namely the height, weight, AP diameter as well as the distance of the centre of the light field to symphysis. Also for each radiological view the exposure factors were recorded in order to be used in Monte Carlo simulations.
The criteria for choosing the appropriate mathematical phantom for each child were primarily the physical dimensions of the child and secondarily the age. This was done because of the variation in size of children of the same age.
The mean exposure factors used and the mean number of films for each age group are shown in Table 1
. The classification of children was made according to the mathematical phantom used in order to simulate each child (Table 2).
|
|
All the paediatric MCU examinations were done on a General Medical Merate (Seriate/Bergamo, Italy) DX12 MTT90 radiographic-fluoroscopic system.
ESD for each view was measured with TLD-100 dosimeters (LiF:Mg:Ti). The TLDs were put in the centre of the radiation field, over and underneath the patient. Calibration of the TLD was done with the same radiographic system at 80 kVp, in order to avoid energy dependence of the TLD readings. A set of 30 TLD chips with ±5% accuracy was used for measurements.
The TLDs were read with a Harshaw-Bicron (Bicron, OH) TLD reader with a reading cycle of 300°C and a 10°C s-1 heating rate. After deconvolution of the glow curve, the sum of peak 3, peak 4 and peak 5 was taken as the TLD reading.
Monte Carlo simulations
The Monte Carlo code MCNP-4A [8] was used for calculation of the organ doses. This is a neutron–photon transport code, which simulates photon and electron interactions with tissue. The type of interaction is determined statistically, according to physical laws.
For each Monte Carlo simulation, an input file was created with the following information:
a. the geometry of the problem
b. the elementary composition of tissue types and their density
c. the position and characteristics of the source
d. the type of answers we want to receive from the program (tallies)
For each patient and every radiological view, an input file was created and a Monte Carlo simulation run was performed. A typical simulation run involved about 200 000 starting photons. The relative standard error of the result was less than 1% and each run passed the MCNP-4A statistical tests. Also the variance reduction technique "geometry splitting with Russian roulette" was used so that internal organs with small physical dimensions could receive adequate numbers of photons and thus pass the statistical tests.
Anthropomorphic phantoms
We used mathematical phantoms representing male and female paediatric patients of various ages: newborn, 1 and 5 years of age. The mathematical phantoms are based on the ORNL phantom family [9]. Simple mathematical equations (planes, ellipsoid, tori and cones), represent external and internal organs of the body. These equations have age dependent values that must be introduced in the "geometry description" section of the MCNP input file.
The Cartesian coordinate system used and the centre of the axis was placed at the base of the trunk. Each phantom consisted of four different materials, representing bone tissue, lung tissue, contrast medium and soft tissue.
The mathematical phantoms were male and female, each one with the appropriate gender specific organs, both having skin and all the internal organs in the trunk. The head, the heart and the oesophagus were not simulated, because they were always out of the radiation field.
Geometry of the irradiation
Imaging during paediatric MCU examination at AXEPA hospital involved the abdominal area, between diaphragm and symphysis.
In Monte Carlo simulations the X-ray unit was represented by a point source. Three different radiological views were modelled for each examination, one AP and two OBL. The OBL views were at an angle of 45° right (R-OBL) and 45° left (L-OBL) of the axis normal to the radiological table.
Only the X-ray source, the patient (phantom) and a volume of air were modelled. Details of the examination room and the equipment were not modelled. The focus to table distance was always 100 cm. The AP diameter of each patient was known and focus to skin distance for each patient was calculated.
X-ray spectrum
The X-ray unit used for MCU paediatric examinations was a GMM DX12, with a 6-pulse generator and a tube with maximum 150 kVp. Automatic exposure control was not used for the radiographs.
The half value layer (HVL) for the kVp used was measured for each radiological view. The mass attenuation coefficients (µ/
) for various photon energies (Eph) for Al are known from Johns and Cunningham [10] and the density of Al is =2.699 g cm-3. From these data we calculated the linear attenuation coefficient (µ) and the HVL for each photon energy (Eph) used for MCU examinations. The relation between HVL and Eph is shown in Figure 1.
|
In order to evaluate this approximation we calculated the difference between the polyenergetic and equivalent monoenergetic X-ray beam, assuming a typical anode angle of 12° and ripple 30%. The measured HVL for 60 kVp was 2.65 mm Al and the equivalent energy Eph=33.8 keV. We calculated the organ doses for a mathematical phantom representing a 5-year-old child taking the X-ray spectra from IPEMB Report No. 78 [11]. The maximum organ dose difference between monoenergetic and polyenergetic X-ray beam results was 10% and the effective dose difference was 3.6%, which was considered to be an acceptable approximation. Differences of this order of magnitude have also been reported by other authors [12].
Calculation of absorbed organ doses and effective doses
Monte Carlo simulations of the anthropomorphic phantoms were used to calculate the absorbed doses to the various organs.
In order to calculate organ doses, tally F6 from MCNP-4A was used. The output of this tally is MeV g-1 per starting particle and gives the energy imparted at each organ divided by the mass of the organ. Using this output we calculated the mean absorbed dose in the organ.
The number of photons (
) emitted from the source is:
|
|
)air is the mass absorption coefficient (m2 kg-1). The dose was measured for each view, after each MCU examination, with an ionization chamber free in air.
The absorbed organ dose was calculated from the following equation:
|
|
is the photon number. The effective dose was calculated by multiplying each organ dose with the ICRP [13] tissue-weighting factor. In the male phantoms the testes were taken as the gonads, in the females the ovaries.
The mean effective dose conversion factors were calculated from the mean of TLD measurements of the entrance dose and the Monte Carlo calculated effective dose to the computational phantoms.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
.
|
. The organ doses for the radiosensitive organs, calculated with the three mathematical phantoms representing a newborn child, a 1-year-old and a 5-year-old child, for the three different types of views are depicted in Figures 2, 3 and 4.
|
|
|
|
.
|
, respectively.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The mean exposure factors used for MCU examination in this study were 56.8–65.3 kVp and 11.4–15.4 mAs for AP and OBL views (Table 1). Other authors [14] report mean exposure values for AP and OBL views, 70 kVp and 8 mAs for children up to 1 year old and 77 kVp for OBL views for children 2–5 years old. The European Commission [2] suggests 65–90 kV for paediatric MCU examination.
The number of radiological views depends on the technique used at various radiological centres. In this study the mean number of views per MCU examination was 5, 2 AP views and 3 OBL. Almen et al [14] report a total of 6 views per examination, 4 AP and 2 OBL, Gonzalez et al [15] report 3 AP views and 2 OBL, a total of 5 views per examination. Schultz et al [6] report 1 AP view and 2 OBL, a total of 3 views per MCU examination.
The measured ESD (Table 3) from this work can be compared with other studies. Chapple et al [16] report a mean dose of 7.36 mGy for 22 children 0.1–1.0 year old and 11.56 mGy for 20 children 1–5 years old. No details are given for exposure factors and number of views. The study of Almen [14] reports a mean dose of 0.63 mGy for AP view and a mean dose of 0.64 mGy for OBL view for children up to 1.0 years old. For 15 children 2–5 years old the mean dose per AP view was 1.0 mGy and 1.8 mGy per OBL view. The mean ESD for the total examination was 5.8 mGy. The mean entrance dose for AP view in this study was greater than the dose reported by Almen [14], due to the selection of 10 kV smaller and 6 mAs greater mean exposure factors. The result was a less penetrating radiation and greater absorption of radiation by the patient.
Most of the mean organ doses are greater when simulated from a 5-year-old mathematical phantom (Figures 2–4). This happens because the exposure factors and the mean ESD are greater for these children. In some cases the lung or the breast of a newborn or 1-year-old child received a greater dose than the 5-year-old child, due to inappropriate collimation of the beam. The sum of the remainder organ doses was larger for newborn children, because the radiation field was proportionally larger for them than for older children. Also the remainder organs for newborn children are closer to the skin and receive more dose than for larger children.
The effective dose conversion factors (Cf) (Table 5) were greater for newborn female than newborn male patients and this is consistent with the study of Schultz et al [6]. An explanation of this can be the contribution of breast and ovaries, which both have large weighting factors, while for male patients the testes were sometimes outside the radiation field. The exact positioning and size of the radiation field is crucial to the total amount of dose received by the internal and external organs. Because newborn babies have small physical dimensions, small differences in collimation have great influence on the Cf values. This difference of Cf factors is less pronounced in larger children represented by a 5-year-old mathematical phantom.
The Cf factors for L-OBL views are greater than the Cf factors for R-OBL views. This is because the stomach and lower large intestine receive greater doses in L-OBL views and both have a weighting factor 0.12.
The Cf factors decrease as age increases (Table 5). Newborn children have Cf factors almost twice those of 1-year-old children. For children belonging to the 5-year-old group the Cf factors for AP and OBL views are almost the same. This happens for both male and female patients. For children belonging to newborn and 1-year-old groups there is a great range of Cf values between 0.12 and 1.95 mSv (Gy·cm2)-1. This is due to different radiation fields chosen for each examination and the different physical growth of each child. The results are consistent with other studies [6].
The Cf values cannot be used generally, because they depend on exposure factors, the technical parameters of the radiological machine and the dimensions and position of the light field. For this reason the results cannot be compared with NRPB-R279 [7] because this study used different light field dimensions and geometry of irradiation.
The mean effective dose (E) per examination for male patients is 0.86±0.31 mSv and for female patients 0.76±0.28 mSv. Male patients received 5.2 views each, while female patients received 4.7 views. This explains the lower effective dose for female patients. For both sexes the mean effective dose per view is 0.16 mSv. The mean effective dose per view is approximately the same for the three groups newborn, 1 and 5 years old (Figures 5 and 6). This happens because the radiation field includes more radiosensitive organs for smaller patients but the exposure factors are greater for children of greater size.
Comparisons of the results with three other studies are given in Table 6.
|
For this study the effective dose was calculated using the weighting factors of ICRP No. 60 [13]. Using the calculated effective dose values, it can be predicted that one case of lethal cancer will present per 23 000 MCU examinations for males and one case per 26 000 examinations for females assuming a fatal cancer probability of 5.0 x 10-2 Sv-1 which is recommended by the ICRP for the total population. This estimated number of lethal cases could be 2–3 times larger according to the various projection models for the estimation of the lethal cancer probabilities for children (ICRP 60).
|
Conclusion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
The effective dose conversion factors (Cf) were greater for newborn female than newborn male patients.
The Cf factors for L-OBL views are greater than the Cf factors for R-OBL views. The Cf factors decrease as the child's age increases. Newborn children have Cf factors almost twice those for children belonging to the 1-year-old group. For children belonging to the 5-year-old group the Cf factors for AP and OBL views are almost the same. For children belonging to newborn and 1-year-old groups there is a great range of Cf values between 0.12 mSv (Gy·cm2)-1 and 1.95 mSv (Gy·cm2)-1.
The mean effective dose (E) per examination for male patients is 0.86±0.31 mSv and for female patients 0.76±0.28 mSv. For both sexes the mean effective dose per view is 0.16 mSv and is approximately the same for the three children's groups newborn, 1 and 5 years old.
Compared with other studies the effective dose given at MCU examination is in the middle of the reported range. Some measures that can reduce the dose even further are proper collimation, selected tube voltage greater than 65 kV and an X-ray unit with automatic exposure control.
Received for publication December 6, 2002. Revision received May 20, 2003. Accepted for publication July 8, 2003.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  V. L. Ward, K. J. Strauss, C. E. Barnewolt, D. Zurakowski, V. Venkatakrishnan, F. H. Fahey, R. L. Lebowitz, and G. A. Taylor Pediatric Radiation Exposure and Effective Dose Reduction during Voiding Cystourethrography Radiology, December 1, 2008; 249(3): 1002 - 1009. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Sulieman, K Theodorou, M Vlychou, T Topaltzikis, D Kanavou, I Fezoulidis, and C Kappas Radiation dose measurement and risk estimation for paediatric patients undergoing micturating cystourethrography Br. J. Radiol., September 1, 2007; 80(957): 731 - 737. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |
