Radiation Protection Dosimetry Advance Access published November 4, 2014 Radiation Protection Dosimetry (2014), pp. 1–6
doi:10.1093/rpd/ncu321
RADIATION EXPOSURE DURING PAEDIATRIC CT IN SUDAN: CT DOSE, ORGAN AND EFFECTIVE DOSES
*Corresponding author: [email protected], [email protected] Received 1 May 2014; revised 20 September 2014; accepted 24 September 2014 The purpose of this study was to assess the magnitude of radiation exposure during paediatric CT in Sudanese hospitals. Doses were determined from CT acquisition parameters using CT-Expo 2.1 dosimetry software. Doses were evaluated for three patient ages (0– 1, 1–5 and 5 –10 y) and two common procedures (head and abdomen). For children aged 0– 1 y, volume CT air kerma index (Cvol ), air Kerma–length product and effective dose (E) values were 19.1 mGy, 265 mGy.cm and 3.1 mSv, respectively, at head CT and those at abdominal CT were 8.8 mGy, 242 mGy.cm and 7.7 mSv, respectively. Those for children aged 1 –5 y were 22.5 mGy, 305 mGy.cm and 1.1 mSv, respectively, at head CT and 12.6 mGy, 317 mGy.cm, and 5.1 mSv, respectively, at abdominal CT. Dose values and variations were comparable with those reported in the literature. Organ equivalent doses vary from 7.5 to 11.6 mSv for testes, from 9.0 to 10.0 mSv for ovaries and from 11.1 to 14.3 mSv for uterus in abdominal CT. The results are useful for dose optimisation and derivation of national diagnostic reference levels.
INTRODUCTION The use of CT has increased rapidly in the past two decades, as a result of recent advances, particularly multidetector technology, which provided increased and more diverse applications. However, there is also the potential for inappropriate use and unnecessary radiation dose. By their nature, CT examinations contribute disproportionately to the collective diagnostic radiation dose to the population; for example, in UK, it has been estimated that 9 % of diagnostic radiology procedures are CT examinations, but their contribution to the collective dose is 47 %(1). Because of the technological advancement added to the clear benefit to the examined individuals, the frequency of CT examinations is increasing worldwide and the types of examination using CT are also becoming more numerous. The proportion of childhood CT examinations is, therefore, increasing(2). At the same time, wide variations in patient dose have been reported, indicating the need for dose optimisation to limit CT exposure by following the ALARA (as low as reasonably achievable) principle(3, 4). There is a variety of strategies to limit radiation dose in CT, including performing only necessary examinations, limiting the region of coverage and adjusting individual CT settings based on indication,
region imaged and size of the child. Radiation dose in CT can be reduced as well as in other X-ray procedures using reference dose levels. Radiation protection in paediatrics radiology particularly in CT deserves unique considerations in children since they are considerably more sensitive to radiation than adults and have a longer life expectancy. As a result, the risk for developing a radiationrelated cancer can be several times higher for a young child compared with an adult exposed to an identical CT scan(5). Recently, several surveys were carried out in Sudan to study the doses of radiation that patients were exposed to during diagnostic procedures(4, 6 – 8). The common goal of these surveys was to establish a national patient exposure databank, increase radiation protection awareness in the medical field and use the results to formulate national diagnostic reference levels (DRLs). As a continuation of these efforts, the authors conducted a dose survey to assess radiation exposure of children in head and abdominal CT examinations. It was motivated by the increased concern about the risk of developing cancer for CT examinations during childhood and will complement the authors’ previous study on adult CT(4). Paediatric CT doses in some Sudanese hospitals were presented
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
I. I. Suliman1,2,*, H. M. Khamis2, T. H. Ombada3, K. Alzimami4, M. Alkhorayef4 and A. Sulieman5 1 Department of Radiology and Molecular Imaging, Medical Physics Section, College of Medicine and Health Sciences, Sultan Qaboos University, PO Box 35, 123 Al-Khod, Oman 2 Sudan Atomic Energy Commission, Radiation Safety Institute, PO Box 3001, Khartoum, Sudan 3 Faculty of Science and Technology, Department of Medical Physics, Al-Neelain University, PO Box 12702, Khartoum, Sudan 4 Department of Radiological Sciences, Applied Medical Sciences College, King Saud University, PO Box 10219, Riyadh 11433, Saudi Arabia 5 Radiology and Medical Imaging Department, College of Applied Medical Sciences, Salman Bin Abdulaziz University, PO Box 422, Alkharj 11943, Saudi Arabia
I. I. SULIMAN ET AL.
MATERIALS AND METHODS This study was conducted in 2011 and consisted of a survey of scan parameters and equipment information in six CT clinics in Khartoum (Sudan). Data were used to assess doses for 132 children who underwent head and abdominal CT (abdomen–pelvis) examinations. The local ethics committees of all participating institutions approved the study protocol. Children are separated in three age groups 0–1, 1–5 and 5 –10 y. In each centre, one of the authors collected information in regard to: † † †
CT equipment-specific information W Made/model/year of installation, W Number of slices. Patient demographic data W Age and gender. CT scan parameters W kV, mA, rotation time and scan time (spiral mode), W Scan length (start and end of scan region), W Number of slices, slice thickness, collimation and pitch.
Dosimetric calculations CT Expo software was used to calculate common CT dose descriptors: (i) CT air kerma index (Cw) and volume air kerma index (CVOL) provides an indication of the average absorbed dose in the scanned region, (ii) CT air kerma –length product (PKL,CT) the integrated absorbed dose along a line parallel to the axis of rotation for the complete CT examination, and (iii) effective dose (E): a method for comparing patient doses from different diagnostic procedures (effective dose)(4, 10 – 12). In practice, assessment of CT air kerma index can be made using a pencil ionisation chamber with an active length of 100 mm. These measurements are carried out either free-in-air (Cair) or at the centre (C100,p) and at the periphery (C100,p) of the standard head or body CT dosimetry phantom from which weighted CT dose index is obtained Cw ¼ ð1=3Þ CK;PMMA; 100; c þ ð2=3ÞCK; PMMA; 100; p (10 – 12).
For the purpose of dose estimation, the ratio of Cw and Cair is defined for the standard head PH ¼ Cw; H =Cair and body PB ¼ Cw; B =Cair dosimetry phantom(9). For certain scanners, the value of Cair is measured at some reference kV values. The values of Cw for any other scanner are therefore obtained using Equation 1(9, 13): Cw ¼ Cair :
U Uref:
kU :kOB :P
ð1Þ
where P ¼ nCw =nCair and is determined separately for head and body; ðU=Uref: ÞkU is the voltage correction factor, kOB is the slice collimation and over beaming correction factor. The overall energy delivered by a given scan protocol is better represented using CT air kerma –length product PKL,CT. PKL; CT ¼ CVOL :Ltot
ð2Þ
where Ltot is the total length and is calculated as: Ltot ¼ jfirst last slice positionj þ hrec þ DL), where DL is the scanner over ranging, hrec is slice reconstruction, DL depends on the scan length and scanner type and varies from 0 to 9.2 cm for head CT examinations and from 4.0 to 17 cm in abdominal CT. The authors have used CT-Expo Version 2.1 software tool for dose calculations(13) and CT-Expo tools— based on Monte Carlo data published by the Research Center for Environment and Health in Germany—for dose calculation. Dose estimation is done based on mathematical phantoms for adult (ADAM and EVA), CHILD and BABY. The software allows the calculations of the CT dose descriptors (Cvol and PKL,CT), organ doses and effective dose in accordance with new recommendations of the international commission for radiological protection(14, 15). One limitation of dose estimation using CT-Expo software is that it has only two sizes of paediatric phantoms: a baby phantom of 7 weeks old and a child phantom of 5– 7 y old resulting in underestimations or overestimations for older and younger children, respectively(13, 15). However, the authors preferred to use CT-Expo software for the authors’ calculations compared with ImPACT CT dosimetry software where doses have to be calculated first for adults and then conversion coefficients are used relating adults’ dose to children, which could results in much more uncertainties. RESULTS Table 1 summaries the characteristic performance parameters for the CT systems included in the study. Three CT scanners were from Siemens (two 16-slice and one dual-slice scanners); 4-slice scanner from
Page 2 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
alongside with doses from other 40 countries in an International Atomic Energy Agency project (2). Detailed dosimetric data as presented in the current study are important for providing patient exposure data for optimisation and formulation of national DRLs. Organ and effective doses presented highlight the magnitude of risk in CT examination of children as compared with adult and other X-ray procedures. The study strengthens the ongoing international efforts on radiation protection in radiology particularly in high-dose modalities such as CT. This study provides practical applications of dosimetry formalisms given in ref. (9).
RADIATION EXPOSURE DURING PAEDIATRIC CT IN SUDAN Table 1. Summary of characteristic performance parameters for the CT systems used for dose calculation in this study. Code
Make
Model
Uref (kV)
Number of slices
Head mode nCw
SI SII SIII T G N
Siemens Siemens Siemens Toshiba GE Neusoft
Emotion Duo Sensation 16 Sensation 16 Acquilion CTe Neusoft
2 16 16 4 1 2
(mGy mAs21)
130 120 120 120 120 120
0.215 0.184 0.184 0.198 0.154 0.127
Body mode PH
nCw
(mGy mAs21)
0.71 0.76 0.76 0.72 0.71 **
PB
0.215 0.131 0.131 0.198 0.154 0.083
0.71 0.77 0.77 0.72 0.71 **
Age group
Head Mean
Cvol (mGy) 0 –1 y 19.1 1 –5 y 22.5 5 –10 y 29.3 PKL,CT (mGy . cm) 0 –1 y 265 1 –5 y 305 5 –10 y 407 E (mSv) 0 –1 y 3.05 1 –5 y 1.06 5 –10 y 1.23
Abdomen
s
Range
Third quartile
Max/min
Mean
s
Range
Third quartile
Max/min
8.3 11.4 15.1
12.0–33.2 11.9–44.6 20.7–51.9
19.3 21.2 30.8
2.8 3.8 2.5
8.8 12.6 11.8
1.4 1.3 3.7
7.9– 10.4 10.5– 13.6 5.6– 15.3
9.3 13.6 13.3
1.3 1.3 2.7
76 122 115
181–383 155–523 304–572
285 316 433
2.1 3.4 1.9
242 317 314
81 78 123
187– 335 196– 393 154– 340
269 358 340
1.8 2.0 2.2
0.53 0.32 0.37
2.65–3.82 0.65–1.45 0.67–1.64
3.15 1.25 1.40
1.4 2.2 2.5
7.73 5.08 5.00
1.54 1.79 2.27
6.70– 9.50 3.20– 7.80 2.51– 6.00
8.18 5.76 5.25
1.4 2.4 2.4
Table 3. Median and range (min –Max) of acquisition parameters for paediatric head and abdomen CT examination. Examination
Head Abdomen
Group age (y)
1 5 10 1 5 10
N
12 22 15 27 34 22
kV
mAs
L
Pitch
Median
Range
Median
Range
Median
Range
Median
Range
120 120 125 100 120 120
110–120 120–140 120–140 90–110 110–130 110–120
105 132 143 131 95 82
99– 231 104– 228 123– 241 80– 147 77– 112 45– 179
11.6 12.4 12.9 24.8 26.4 29.6
3.2–14.0 8.9–16.3 9.2–16.1 23.7 –33.3 21.8 –29.5 19.1 –37.9
1.2 1.3 1.2 1.4 1.4 1.4
1.0–1.3 0.9–1.8 1 –1.6 0.8–1.8 1.1–1.5 1.0–1.8
Toshiba, single-slice scanner from General electric and Neusoft. Comparable values are presented. The Neusoft scanner is not included in CT-Expo software library. This scanner is therefore calibrated and the obtained normalised Cvol are used for dose calculation. Table 2 presents doses for head and abdominal CT examinations of children arranged in three age groups (0 –1, 1–5 and 5–10 y) and the descriptive statistics of mean doses averaged over the participating centres.
Table 3 shows the median and range (Min –Max) of acquisition parameters for paediatric head and abdomen CT examination. Median tube voltage varied from 100 to 125 kV with 120 being the most frequently used tube voltage. Therefore, the normalised CT air kerma index values are given for this reference kV. When tube voltage deviates from the reference value (120 kV), correction factor is applied to account for differences in tube voltages (Equation 1).
Page 3 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
Table 2. Descriptive statistics mean, s, range (Min –Max), third quartile, max/min of average doses of the participating centres.
I. I. SULIMAN ET AL.
In Table 5, age-specific conversion coefficients E/PKL,CT (mSv mGy21 cm21) derived from correlation between patient dose and body size (age). DISCUSSION As presented in Table 2, both Cvol and PKL,CT increase with age whereas effective dose decreases. Large dose variations were observed in the 5–10 y age category, which is expected due to the large age range. These variations are clear evidence that dose reduction is possible without substantial degradation of image quality. In light of the results obtained, several dose optimisation measures are possible. Concerning tube voltage, nearly all hospitals use 120 kV, a value usually used in adult CT(4). Significant decreases in dose can be achieved with lower kVp selections, e.g. decreasing the tube voltage to 80 or 100 kVp for smaller patients coupled with appropriate choice of mAs(19). As shown in Table 3, lower kVp values were used successfully for 1-y-old children in
Table 4. CT dose descriptors in terms of Cvol (mGy), PKL,CT (mGy.cm) and E (mSv) obtained in the present work compared with the data reported in Belgium(16) (B), Canada(17) (C) and Australia(18) (U). Age group
Cvol (mGy) 0 –1 1 –5 5 –10 PKL,CT (mGy.cm) 0 –1 1 –5 5 –10 E (mSv) 0 –1 1 –5 5 –10
Head CT
Abdomen CT
Sudan (this study)
B
C
U
Sudan (this study)
B
C
U
19.1 22.5 29.3
20.5 26.5
** ** **
30 45 50
8.8 12.6 11.8
16.4 17.7 20.8
** ** **
12 13 20
265 305 407
168.8 225.2 592
543 610 639
270 470 620
241.7 317.2 314.0
324 387 714
371 420 595
200 230 370
3.1 1.06 1.23
0.8 0.82 1.63
3.6 2.4 2.0
** ** **
7.73 5.08 5.00
8.2 9.8 11.9
8.4 8.9 5.9
** ** **
Figure 1. The correlation between CT air kerma–length product (PKL,CT) and child age in: (a) head CT and (b) abdomen CT.
Page 4 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
As shown in Table 3, low tube current per rotation is used, which could be the main reason for relative low doses in this study. In Table 4, the results obtained have been compared with values of other recent studies(16 – 18). It can be observed that the values of Cvol and PKL,CT for head and abdominal CT reported in this study were lower than the values reported by Pages et al.(16) for the same types of procedure. The values of E were comparable in both studies. This is mainly because E was evaluated in this study based on recent ICRP recommendations(14). In comparison with similar studies, current dose variations are lower compared with those reported by Pages et al.(16) in Belgium who found the ratio of highest to lowest E up to 4 for head CT and up to 6.6 for abdomen CT. The correlation between patient dose and child’s body size (age) is graphically presented in Figure 1a and b. The determination coefficients (R 2) are 0.17 and 0.25 for head and abdominal CT, respectively.
RADIATION EXPOSURE DURING PAEDIATRIC CT IN SUDAN Table 5. Conversion coefficients, E/PKL,CT (mSv mGy21 cm21) values as a function of child size (age) obtained using CT-Expo software. Examination Head CT Abdominal CT
Newborn
1y
5y
10 y
8.5 28.3
7.7 26.5
4.6 19.9
0.6 10.5
Figure 2. (a) Organ equivalent doses in head CT. (b) Organ equivalent doses in abdominal CT.
CT are presented in Figure 2a and b, respectively. In head CT, mean organ equivalent dose to the lens of the eye and the thyroid for 1-y-old child are 17 and 9 mSv, respectively. In abdominal CT, mean organ equivalent dose to the testes, the ovaries and the uterus for 1-y-old child are 8, 9 and 14 mSv, respectively. These results are comparable with those previously reported by Fujii et al.(17), who found organ doses in the range of 3–16 mGy for children with abdominal CT. As pointed out, lower effective mAs values used in multi-slice CT scanners delivered lower exposure doses than those in the earlier single-slice or axial scanners, and therefore, the incidence rate of radiation-induced cancer for children will be significantly lowered(17). In radiodiagnostic, an important optimisation measure is the establishment and the use of DRLs(6). Radiation protection of the patient in the country would be greatly assisted if present doses are used to suggest DRLs. In doing so, the authors have taken mean patient doses from each centre. Third quartile of the mean hospital doses are used as the DRLs presented in Table 6. When these DRLs are exceeded in paediatrics CT practice, the hospital or the region of study should act appropriately to bring doses below what is considered acceptable.
Page 5 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
both head and abdominal CT. For other age categories, the values of kV are almost constant and comparable with what is usually used for adult patients. In those cases, decreasing tube current to an acceptable level could be exercised as a way for dose reduction. Modern multi-slice CT scanners are equipped with tube current modulation features where radiation dose is adjusted according to the patient attenuation. In this study, two multi-slice CT units (Siemens Sensation 16) were employing automatic current modulation whereas the remaining units used paediatric protocol provided by the vendors. It is difficult to draw conclusions on the differences on the dose performance because of small size and also both scanners are of the same make. An important aspect of CT Expo Dosimetry software is that it incorporates factors such as scanner over-ranging and overbeaming in dose calculations. Both factors increase patient dose by increasing the exposed area and reconstruction slices, respectively. As shown in Table 3, relatively increased pitch (median 1.2–1.4) was used in the current study, and this partially explains the low doses seen. CT scanners from Siemens use pitch-modified mAs. Therefore, a pitch value of unity was used in dose calculations of these scanners. In conformity to our previous findings, we have demonstrated a correlation (Figure 1a and b) between body size (age) and the average radiation dose(8). As seen, patient age explains 23 % (head CT) and 25 % (abdominal CT) in dose variations. Optimization of technique parameters according to body size is therefore important. Patient age –dose correlation shown in the present study is relatively low because of differences in the dose performance of different scanner models included. In Table 5, conversion coefficients (E/PKL,CT) are given derived from the correlation between patient dose and body size (age). These coefficients permit PKL,CT values to be converted into a corresponding value of the paediatric patient E. Higher conversion coefficients are shown for younger children emphasising risk increase during childhood exposure. Radiation-induced cancer is dependent upon which organ is irradiated. In CT Expo software, organ equivalent doses are given as the output file alongside with other CT dose descriptors. Organ equivalent doses calculated in this study for head and abdominal
I. I. SULIMAN ET AL. Table 6. Suggested DRLs for head and abdominal CT examinations of children in Sudan. 4. Age group
Abdomen CT
19.3 21.2 30.8
9.3 13.6 13.3
285 316 433
269 358 340
5.
6.
7.
CONCLUSIONS The results of the present study were comparable or below the reference dose levels for the same type of examination and age group. Even if sample size was relatively small, hospital used low mAs and kV for younger children (0–1 year). This was partly achieved by using automatic current modulation or CT protocol for children. Some hospitals used 120 kV for examination of neonates as well adults. These centres were advised on the possibility of dose saving by using lower tube potential for small children. Dose performance is further improved by the fast rotation time (low mAs) of the multi-slice CT added to the increased pitch. As shown, factors such as overbeaming and overhanging can lead to a significant increase in dose, and this reflects the important of using Dosimetry software for accurate dose estimates. Current results are important for providing patient exposure data for optimisation and derivation of national DRLs. DRLs for the selected CT examinations of children are provided.
8. 9.
10.
11. 12. 13. 14.
ACKNOWLEDGEMENTS The authors extend their appreciation to the College of Applied Medical Sciences Research Center and the Deanship of Scientific Research at King Saud University for funding this research. The first author thanks the International Center for Theoretical Physics for the Associate Award.
15.
16.
REFERENCES
17.
1. Shrimpton, P. C., Hillier, M. C., Lewis, M. A. and Dunn, M. National survey of doses from CT in the UK: 2003 Br. J. Radiol. 79, 968– 980 (2006). 2. Vassileva, J., Rehani, M., Applegate, K., Ahmed, N. A., Humoud, Al-Dhuhli and Al-Naemi, H. M. IAEA survey of pediatrics computed tomography practice in 40 countries in Asia, Europe, Latin America and Africa: procedures and protocols. Eur. Radiol. 23, 623 – 631 (2013). 3. Moorin, R. E., Forsyth, R., Gibson, D. J. and Fox, R. Radiation dosimetry assessment of routine CT scanning
18. 19.
Page 6 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
Cvol (mGy) 0 –1 y 1 –5 y 5 –10 y PKL,CT (mGy.cm) 0 –1 y 1 –5 y 5 –10 y
Head CT
protocols used in Western Australia. J. Radiol. Prot. 33, 295– 312 (2013). Suliman, I. I., Abdalla, S. E., Nada, A. A., Galal, M. A. and IsamSalih Survey of computed tomography technique and radiation dose in Sudanese hospitals. Eur. J. Radiol. 80(3), e544– e551 (2001). United Nations Scientific Committee on Effect of Atomic Radiation. Sources and effects of ionizing radiation. Report to the General Assembly with Scientific Annex, United Nations (2000). Suliman, I. I. and Mohammedzein, T. S. Estimation of adult patient doses for common diagnostic X-ray examinations in Wad-madani, Sudan: derivation of local diagnostic reference levels. Australas. Phys. Eng. Sci. Med. 37, 425–429 (2014). Suliman, I. I. and Elawed, S. O. Radiation measurements for optimisations of chest X-ray examinations of children in general radiography hospitals. Radiat. Prot. Dosim. 156(3), 310– 314 (2013). Suliman, I. I. and Elshiekh, E. H. A. Radiation doses from some common pediatric X-ray examinations in Sudan. Rad. Prot. Dosim. 132(1), 64– 72 (2008). Brix, G., Lechel, U., Veit, R., Truckenbrodt, R., Stamm, G., Coppenrath, E. M., Griebel, J. and Nagel, H. D. Assessment of a theoretical formalism for dose estimation in CT: an anthropomorphic phantom study. Eur. Radiol. 14, 1275–1284 (2004). International Commission on Radiation Measurements and Units. Patient dosimetry for X-rays used in Medical Imaging. ICRU report No. 74, ICRU, Bethesda, MD (2006). International Atomic Energy Agency. Dosimetry in diagnostic radiology: an international code of practice. IAEA TRS. No 457, IAEA (2007). Commission for European Community (CEC). European guidelines on quality criteria for computed tomography. CEC report EUR 16262 EN, CEC (1999). Stamm, G. and Nagel, H. D. CT-Expoein neuartiges Programm zur Dosisevaluierung in der CT. Fortschr Rontgenstr 174, 1570– 1576 (2002). International Commission on Radiological Protection. The 2007 Recommendations of the International Commission on Radiological Protection. ICRP Publication 103, Ann. ICRP 37 ICRP (2007). Bernier, M.-O., Rehel, J.-L., Brisse, H. J., Wu-Zhou, X., Caer-Lorho, S., Jacob, S., Chateil, J. F., Aubert, B. and Laurier, D. Radiation exposure from CT in early childhood: a French large-scale multicentre study. B. J. Radiol. 85, 53– 60 (2012). Pages, J., Buls, N. and Osteaux, M. CT doses in children: a multicentre study. Br. J. Radiol. 76, 803 –811 (2003). Fujii, K., Aoyama, T., Koyama, S. and Kawaura, C. Comparative evaluation of organ and effective doses for pediatrics patients with those for adults in chest and abdominal CT examinations. Br. J. Radiol. 80, 657–667 (2007). Australian Radiation Protection and Nuclear Safety Agency. National diagnostic reference level fact sheet. ARPANA (2011). International Commission on Radiological Protection. Radiological protection in pediatrics diagnostic and interventional radiology. ICRP Publication 121. Ann. ICRP 42(2). ICRP (2013).
doi:10.1093/rpd/ncu321
RADIATION EXPOSURE DURING PAEDIATRIC CT IN SUDAN: CT DOSE, ORGAN AND EFFECTIVE DOSES
*Corresponding author: [email protected], [email protected] Received 1 May 2014; revised 20 September 2014; accepted 24 September 2014 The purpose of this study was to assess the magnitude of radiation exposure during paediatric CT in Sudanese hospitals. Doses were determined from CT acquisition parameters using CT-Expo 2.1 dosimetry software. Doses were evaluated for three patient ages (0– 1, 1–5 and 5 –10 y) and two common procedures (head and abdomen). For children aged 0– 1 y, volume CT air kerma index (Cvol ), air Kerma–length product and effective dose (E) values were 19.1 mGy, 265 mGy.cm and 3.1 mSv, respectively, at head CT and those at abdominal CT were 8.8 mGy, 242 mGy.cm and 7.7 mSv, respectively. Those for children aged 1 –5 y were 22.5 mGy, 305 mGy.cm and 1.1 mSv, respectively, at head CT and 12.6 mGy, 317 mGy.cm, and 5.1 mSv, respectively, at abdominal CT. Dose values and variations were comparable with those reported in the literature. Organ equivalent doses vary from 7.5 to 11.6 mSv for testes, from 9.0 to 10.0 mSv for ovaries and from 11.1 to 14.3 mSv for uterus in abdominal CT. The results are useful for dose optimisation and derivation of national diagnostic reference levels.
INTRODUCTION The use of CT has increased rapidly in the past two decades, as a result of recent advances, particularly multidetector technology, which provided increased and more diverse applications. However, there is also the potential for inappropriate use and unnecessary radiation dose. By their nature, CT examinations contribute disproportionately to the collective diagnostic radiation dose to the population; for example, in UK, it has been estimated that 9 % of diagnostic radiology procedures are CT examinations, but their contribution to the collective dose is 47 %(1). Because of the technological advancement added to the clear benefit to the examined individuals, the frequency of CT examinations is increasing worldwide and the types of examination using CT are also becoming more numerous. The proportion of childhood CT examinations is, therefore, increasing(2). At the same time, wide variations in patient dose have been reported, indicating the need for dose optimisation to limit CT exposure by following the ALARA (as low as reasonably achievable) principle(3, 4). There is a variety of strategies to limit radiation dose in CT, including performing only necessary examinations, limiting the region of coverage and adjusting individual CT settings based on indication,
region imaged and size of the child. Radiation dose in CT can be reduced as well as in other X-ray procedures using reference dose levels. Radiation protection in paediatrics radiology particularly in CT deserves unique considerations in children since they are considerably more sensitive to radiation than adults and have a longer life expectancy. As a result, the risk for developing a radiationrelated cancer can be several times higher for a young child compared with an adult exposed to an identical CT scan(5). Recently, several surveys were carried out in Sudan to study the doses of radiation that patients were exposed to during diagnostic procedures(4, 6 – 8). The common goal of these surveys was to establish a national patient exposure databank, increase radiation protection awareness in the medical field and use the results to formulate national diagnostic reference levels (DRLs). As a continuation of these efforts, the authors conducted a dose survey to assess radiation exposure of children in head and abdominal CT examinations. It was motivated by the increased concern about the risk of developing cancer for CT examinations during childhood and will complement the authors’ previous study on adult CT(4). Paediatric CT doses in some Sudanese hospitals were presented
# The Author 2014. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
I. I. Suliman1,2,*, H. M. Khamis2, T. H. Ombada3, K. Alzimami4, M. Alkhorayef4 and A. Sulieman5 1 Department of Radiology and Molecular Imaging, Medical Physics Section, College of Medicine and Health Sciences, Sultan Qaboos University, PO Box 35, 123 Al-Khod, Oman 2 Sudan Atomic Energy Commission, Radiation Safety Institute, PO Box 3001, Khartoum, Sudan 3 Faculty of Science and Technology, Department of Medical Physics, Al-Neelain University, PO Box 12702, Khartoum, Sudan 4 Department of Radiological Sciences, Applied Medical Sciences College, King Saud University, PO Box 10219, Riyadh 11433, Saudi Arabia 5 Radiology and Medical Imaging Department, College of Applied Medical Sciences, Salman Bin Abdulaziz University, PO Box 422, Alkharj 11943, Saudi Arabia
I. I. SULIMAN ET AL.
MATERIALS AND METHODS This study was conducted in 2011 and consisted of a survey of scan parameters and equipment information in six CT clinics in Khartoum (Sudan). Data were used to assess doses for 132 children who underwent head and abdominal CT (abdomen–pelvis) examinations. The local ethics committees of all participating institutions approved the study protocol. Children are separated in three age groups 0–1, 1–5 and 5 –10 y. In each centre, one of the authors collected information in regard to: † † †
CT equipment-specific information W Made/model/year of installation, W Number of slices. Patient demographic data W Age and gender. CT scan parameters W kV, mA, rotation time and scan time (spiral mode), W Scan length (start and end of scan region), W Number of slices, slice thickness, collimation and pitch.
Dosimetric calculations CT Expo software was used to calculate common CT dose descriptors: (i) CT air kerma index (Cw) and volume air kerma index (CVOL) provides an indication of the average absorbed dose in the scanned region, (ii) CT air kerma –length product (PKL,CT) the integrated absorbed dose along a line parallel to the axis of rotation for the complete CT examination, and (iii) effective dose (E): a method for comparing patient doses from different diagnostic procedures (effective dose)(4, 10 – 12). In practice, assessment of CT air kerma index can be made using a pencil ionisation chamber with an active length of 100 mm. These measurements are carried out either free-in-air (Cair) or at the centre (C100,p) and at the periphery (C100,p) of the standard head or body CT dosimetry phantom from which weighted CT dose index is obtained Cw ¼ ð1=3Þ CK;PMMA; 100; c þ ð2=3ÞCK; PMMA; 100; p (10 – 12).
For the purpose of dose estimation, the ratio of Cw and Cair is defined for the standard head PH ¼ Cw; H =Cair and body PB ¼ Cw; B =Cair dosimetry phantom(9). For certain scanners, the value of Cair is measured at some reference kV values. The values of Cw for any other scanner are therefore obtained using Equation 1(9, 13): Cw ¼ Cair :
U Uref:
kU :kOB :P
ð1Þ
where P ¼ nCw =nCair and is determined separately for head and body; ðU=Uref: ÞkU is the voltage correction factor, kOB is the slice collimation and over beaming correction factor. The overall energy delivered by a given scan protocol is better represented using CT air kerma –length product PKL,CT. PKL; CT ¼ CVOL :Ltot
ð2Þ
where Ltot is the total length and is calculated as: Ltot ¼ jfirst last slice positionj þ hrec þ DL), where DL is the scanner over ranging, hrec is slice reconstruction, DL depends on the scan length and scanner type and varies from 0 to 9.2 cm for head CT examinations and from 4.0 to 17 cm in abdominal CT. The authors have used CT-Expo Version 2.1 software tool for dose calculations(13) and CT-Expo tools— based on Monte Carlo data published by the Research Center for Environment and Health in Germany—for dose calculation. Dose estimation is done based on mathematical phantoms for adult (ADAM and EVA), CHILD and BABY. The software allows the calculations of the CT dose descriptors (Cvol and PKL,CT), organ doses and effective dose in accordance with new recommendations of the international commission for radiological protection(14, 15). One limitation of dose estimation using CT-Expo software is that it has only two sizes of paediatric phantoms: a baby phantom of 7 weeks old and a child phantom of 5– 7 y old resulting in underestimations or overestimations for older and younger children, respectively(13, 15). However, the authors preferred to use CT-Expo software for the authors’ calculations compared with ImPACT CT dosimetry software where doses have to be calculated first for adults and then conversion coefficients are used relating adults’ dose to children, which could results in much more uncertainties. RESULTS Table 1 summaries the characteristic performance parameters for the CT systems included in the study. Three CT scanners were from Siemens (two 16-slice and one dual-slice scanners); 4-slice scanner from
Page 2 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
alongside with doses from other 40 countries in an International Atomic Energy Agency project (2). Detailed dosimetric data as presented in the current study are important for providing patient exposure data for optimisation and formulation of national DRLs. Organ and effective doses presented highlight the magnitude of risk in CT examination of children as compared with adult and other X-ray procedures. The study strengthens the ongoing international efforts on radiation protection in radiology particularly in high-dose modalities such as CT. This study provides practical applications of dosimetry formalisms given in ref. (9).
RADIATION EXPOSURE DURING PAEDIATRIC CT IN SUDAN Table 1. Summary of characteristic performance parameters for the CT systems used for dose calculation in this study. Code
Make
Model
Uref (kV)
Number of slices
Head mode nCw
SI SII SIII T G N
Siemens Siemens Siemens Toshiba GE Neusoft
Emotion Duo Sensation 16 Sensation 16 Acquilion CTe Neusoft
2 16 16 4 1 2
(mGy mAs21)
130 120 120 120 120 120
0.215 0.184 0.184 0.198 0.154 0.127
Body mode PH
nCw
(mGy mAs21)
0.71 0.76 0.76 0.72 0.71 **
PB
0.215 0.131 0.131 0.198 0.154 0.083
0.71 0.77 0.77 0.72 0.71 **
Age group
Head Mean
Cvol (mGy) 0 –1 y 19.1 1 –5 y 22.5 5 –10 y 29.3 PKL,CT (mGy . cm) 0 –1 y 265 1 –5 y 305 5 –10 y 407 E (mSv) 0 –1 y 3.05 1 –5 y 1.06 5 –10 y 1.23
Abdomen
s
Range
Third quartile
Max/min
Mean
s
Range
Third quartile
Max/min
8.3 11.4 15.1
12.0–33.2 11.9–44.6 20.7–51.9
19.3 21.2 30.8
2.8 3.8 2.5
8.8 12.6 11.8
1.4 1.3 3.7
7.9– 10.4 10.5– 13.6 5.6– 15.3
9.3 13.6 13.3
1.3 1.3 2.7
76 122 115
181–383 155–523 304–572
285 316 433
2.1 3.4 1.9
242 317 314
81 78 123
187– 335 196– 393 154– 340
269 358 340
1.8 2.0 2.2
0.53 0.32 0.37
2.65–3.82 0.65–1.45 0.67–1.64
3.15 1.25 1.40
1.4 2.2 2.5
7.73 5.08 5.00
1.54 1.79 2.27
6.70– 9.50 3.20– 7.80 2.51– 6.00
8.18 5.76 5.25
1.4 2.4 2.4
Table 3. Median and range (min –Max) of acquisition parameters for paediatric head and abdomen CT examination. Examination
Head Abdomen
Group age (y)
1 5 10 1 5 10
N
12 22 15 27 34 22
kV
mAs
L
Pitch
Median
Range
Median
Range
Median
Range
Median
Range
120 120 125 100 120 120
110–120 120–140 120–140 90–110 110–130 110–120
105 132 143 131 95 82
99– 231 104– 228 123– 241 80– 147 77– 112 45– 179
11.6 12.4 12.9 24.8 26.4 29.6
3.2–14.0 8.9–16.3 9.2–16.1 23.7 –33.3 21.8 –29.5 19.1 –37.9
1.2 1.3 1.2 1.4 1.4 1.4
1.0–1.3 0.9–1.8 1 –1.6 0.8–1.8 1.1–1.5 1.0–1.8
Toshiba, single-slice scanner from General electric and Neusoft. Comparable values are presented. The Neusoft scanner is not included in CT-Expo software library. This scanner is therefore calibrated and the obtained normalised Cvol are used for dose calculation. Table 2 presents doses for head and abdominal CT examinations of children arranged in three age groups (0 –1, 1–5 and 5–10 y) and the descriptive statistics of mean doses averaged over the participating centres.
Table 3 shows the median and range (Min –Max) of acquisition parameters for paediatric head and abdomen CT examination. Median tube voltage varied from 100 to 125 kV with 120 being the most frequently used tube voltage. Therefore, the normalised CT air kerma index values are given for this reference kV. When tube voltage deviates from the reference value (120 kV), correction factor is applied to account for differences in tube voltages (Equation 1).
Page 3 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
Table 2. Descriptive statistics mean, s, range (Min –Max), third quartile, max/min of average doses of the participating centres.
I. I. SULIMAN ET AL.
In Table 5, age-specific conversion coefficients E/PKL,CT (mSv mGy21 cm21) derived from correlation between patient dose and body size (age). DISCUSSION As presented in Table 2, both Cvol and PKL,CT increase with age whereas effective dose decreases. Large dose variations were observed in the 5–10 y age category, which is expected due to the large age range. These variations are clear evidence that dose reduction is possible without substantial degradation of image quality. In light of the results obtained, several dose optimisation measures are possible. Concerning tube voltage, nearly all hospitals use 120 kV, a value usually used in adult CT(4). Significant decreases in dose can be achieved with lower kVp selections, e.g. decreasing the tube voltage to 80 or 100 kVp for smaller patients coupled with appropriate choice of mAs(19). As shown in Table 3, lower kVp values were used successfully for 1-y-old children in
Table 4. CT dose descriptors in terms of Cvol (mGy), PKL,CT (mGy.cm) and E (mSv) obtained in the present work compared with the data reported in Belgium(16) (B), Canada(17) (C) and Australia(18) (U). Age group
Cvol (mGy) 0 –1 1 –5 5 –10 PKL,CT (mGy.cm) 0 –1 1 –5 5 –10 E (mSv) 0 –1 1 –5 5 –10
Head CT
Abdomen CT
Sudan (this study)
B
C
U
Sudan (this study)
B
C
U
19.1 22.5 29.3
20.5 26.5
** ** **
30 45 50
8.8 12.6 11.8
16.4 17.7 20.8
** ** **
12 13 20
265 305 407
168.8 225.2 592
543 610 639
270 470 620
241.7 317.2 314.0
324 387 714
371 420 595
200 230 370
3.1 1.06 1.23
0.8 0.82 1.63
3.6 2.4 2.0
** ** **
7.73 5.08 5.00
8.2 9.8 11.9
8.4 8.9 5.9
** ** **
Figure 1. The correlation between CT air kerma–length product (PKL,CT) and child age in: (a) head CT and (b) abdomen CT.
Page 4 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
As shown in Table 3, low tube current per rotation is used, which could be the main reason for relative low doses in this study. In Table 4, the results obtained have been compared with values of other recent studies(16 – 18). It can be observed that the values of Cvol and PKL,CT for head and abdominal CT reported in this study were lower than the values reported by Pages et al.(16) for the same types of procedure. The values of E were comparable in both studies. This is mainly because E was evaluated in this study based on recent ICRP recommendations(14). In comparison with similar studies, current dose variations are lower compared with those reported by Pages et al.(16) in Belgium who found the ratio of highest to lowest E up to 4 for head CT and up to 6.6 for abdomen CT. The correlation between patient dose and child’s body size (age) is graphically presented in Figure 1a and b. The determination coefficients (R 2) are 0.17 and 0.25 for head and abdominal CT, respectively.
RADIATION EXPOSURE DURING PAEDIATRIC CT IN SUDAN Table 5. Conversion coefficients, E/PKL,CT (mSv mGy21 cm21) values as a function of child size (age) obtained using CT-Expo software. Examination Head CT Abdominal CT
Newborn
1y
5y
10 y
8.5 28.3
7.7 26.5
4.6 19.9
0.6 10.5
Figure 2. (a) Organ equivalent doses in head CT. (b) Organ equivalent doses in abdominal CT.
CT are presented in Figure 2a and b, respectively. In head CT, mean organ equivalent dose to the lens of the eye and the thyroid for 1-y-old child are 17 and 9 mSv, respectively. In abdominal CT, mean organ equivalent dose to the testes, the ovaries and the uterus for 1-y-old child are 8, 9 and 14 mSv, respectively. These results are comparable with those previously reported by Fujii et al.(17), who found organ doses in the range of 3–16 mGy for children with abdominal CT. As pointed out, lower effective mAs values used in multi-slice CT scanners delivered lower exposure doses than those in the earlier single-slice or axial scanners, and therefore, the incidence rate of radiation-induced cancer for children will be significantly lowered(17). In radiodiagnostic, an important optimisation measure is the establishment and the use of DRLs(6). Radiation protection of the patient in the country would be greatly assisted if present doses are used to suggest DRLs. In doing so, the authors have taken mean patient doses from each centre. Third quartile of the mean hospital doses are used as the DRLs presented in Table 6. When these DRLs are exceeded in paediatrics CT practice, the hospital or the region of study should act appropriately to bring doses below what is considered acceptable.
Page 5 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
both head and abdominal CT. For other age categories, the values of kV are almost constant and comparable with what is usually used for adult patients. In those cases, decreasing tube current to an acceptable level could be exercised as a way for dose reduction. Modern multi-slice CT scanners are equipped with tube current modulation features where radiation dose is adjusted according to the patient attenuation. In this study, two multi-slice CT units (Siemens Sensation 16) were employing automatic current modulation whereas the remaining units used paediatric protocol provided by the vendors. It is difficult to draw conclusions on the differences on the dose performance because of small size and also both scanners are of the same make. An important aspect of CT Expo Dosimetry software is that it incorporates factors such as scanner over-ranging and overbeaming in dose calculations. Both factors increase patient dose by increasing the exposed area and reconstruction slices, respectively. As shown in Table 3, relatively increased pitch (median 1.2–1.4) was used in the current study, and this partially explains the low doses seen. CT scanners from Siemens use pitch-modified mAs. Therefore, a pitch value of unity was used in dose calculations of these scanners. In conformity to our previous findings, we have demonstrated a correlation (Figure 1a and b) between body size (age) and the average radiation dose(8). As seen, patient age explains 23 % (head CT) and 25 % (abdominal CT) in dose variations. Optimization of technique parameters according to body size is therefore important. Patient age –dose correlation shown in the present study is relatively low because of differences in the dose performance of different scanner models included. In Table 5, conversion coefficients (E/PKL,CT) are given derived from the correlation between patient dose and body size (age). These coefficients permit PKL,CT values to be converted into a corresponding value of the paediatric patient E. Higher conversion coefficients are shown for younger children emphasising risk increase during childhood exposure. Radiation-induced cancer is dependent upon which organ is irradiated. In CT Expo software, organ equivalent doses are given as the output file alongside with other CT dose descriptors. Organ equivalent doses calculated in this study for head and abdominal
I. I. SULIMAN ET AL. Table 6. Suggested DRLs for head and abdominal CT examinations of children in Sudan. 4. Age group
Abdomen CT
19.3 21.2 30.8
9.3 13.6 13.3
285 316 433
269 358 340
5.
6.
7.
CONCLUSIONS The results of the present study were comparable or below the reference dose levels for the same type of examination and age group. Even if sample size was relatively small, hospital used low mAs and kV for younger children (0–1 year). This was partly achieved by using automatic current modulation or CT protocol for children. Some hospitals used 120 kV for examination of neonates as well adults. These centres were advised on the possibility of dose saving by using lower tube potential for small children. Dose performance is further improved by the fast rotation time (low mAs) of the multi-slice CT added to the increased pitch. As shown, factors such as overbeaming and overhanging can lead to a significant increase in dose, and this reflects the important of using Dosimetry software for accurate dose estimates. Current results are important for providing patient exposure data for optimisation and derivation of national DRLs. DRLs for the selected CT examinations of children are provided.
8. 9.
10.
11. 12. 13. 14.
ACKNOWLEDGEMENTS The authors extend their appreciation to the College of Applied Medical Sciences Research Center and the Deanship of Scientific Research at King Saud University for funding this research. The first author thanks the International Center for Theoretical Physics for the Associate Award.
15.
16.
REFERENCES
17.
1. Shrimpton, P. C., Hillier, M. C., Lewis, M. A. and Dunn, M. National survey of doses from CT in the UK: 2003 Br. J. Radiol. 79, 968– 980 (2006). 2. Vassileva, J., Rehani, M., Applegate, K., Ahmed, N. A., Humoud, Al-Dhuhli and Al-Naemi, H. M. IAEA survey of pediatrics computed tomography practice in 40 countries in Asia, Europe, Latin America and Africa: procedures and protocols. Eur. Radiol. 23, 623 – 631 (2013). 3. Moorin, R. E., Forsyth, R., Gibson, D. J. and Fox, R. Radiation dosimetry assessment of routine CT scanning
18. 19.
Page 6 of 6
Downloaded from http://rpd.oxfordjournals.org/ at Florida International University on November 18, 2015
Cvol (mGy) 0 –1 y 1 –5 y 5 –10 y PKL,CT (mGy.cm) 0 –1 y 1 –5 y 5 –10 y
Head CT
protocols used in Western Australia. J. Radiol. Prot. 33, 295– 312 (2013). Suliman, I. I., Abdalla, S. E., Nada, A. A., Galal, M. A. and IsamSalih Survey of computed tomography technique and radiation dose in Sudanese hospitals. Eur. J. Radiol. 80(3), e544– e551 (2001). United Nations Scientific Committee on Effect of Atomic Radiation. Sources and effects of ionizing radiation. Report to the General Assembly with Scientific Annex, United Nations (2000). Suliman, I. I. and Mohammedzein, T. S. Estimation of adult patient doses for common diagnostic X-ray examinations in Wad-madani, Sudan: derivation of local diagnostic reference levels. Australas. Phys. Eng. Sci. Med. 37, 425–429 (2014). Suliman, I. I. and Elawed, S. O. Radiation measurements for optimisations of chest X-ray examinations of children in general radiography hospitals. Radiat. Prot. Dosim. 156(3), 310– 314 (2013). Suliman, I. I. and Elshiekh, E. H. A. Radiation doses from some common pediatric X-ray examinations in Sudan. Rad. Prot. Dosim. 132(1), 64– 72 (2008). Brix, G., Lechel, U., Veit, R., Truckenbrodt, R., Stamm, G., Coppenrath, E. M., Griebel, J. and Nagel, H. D. Assessment of a theoretical formalism for dose estimation in CT: an anthropomorphic phantom study. Eur. Radiol. 14, 1275–1284 (2004). International Commission on Radiation Measurements and Units. Patient dosimetry for X-rays used in Medical Imaging. ICRU report No. 74, ICRU, Bethesda, MD (2006). International Atomic Energy Agency. Dosimetry in diagnostic radiology: an international code of practice. IAEA TRS. No 457, IAEA (2007). Commission for European Community (CEC). European guidelines on quality criteria for computed tomography. CEC report EUR 16262 EN, CEC (1999). Stamm, G. and Nagel, H. D. CT-Expoein neuartiges Programm zur Dosisevaluierung in der CT. Fortschr Rontgenstr 174, 1570– 1576 (2002). International Commission on Radiological Protection. The 2007 Recommendations of the International Commission on Radiological Protection. ICRP Publication 103, Ann. ICRP 37 ICRP (2007). Bernier, M.-O., Rehel, J.-L., Brisse, H. J., Wu-Zhou, X., Caer-Lorho, S., Jacob, S., Chateil, J. F., Aubert, B. and Laurier, D. Radiation exposure from CT in early childhood: a French large-scale multicentre study. B. J. Radiol. 85, 53– 60 (2012). Pages, J., Buls, N. and Osteaux, M. CT doses in children: a multicentre study. Br. J. Radiol. 76, 803 –811 (2003). Fujii, K., Aoyama, T., Koyama, S. and Kawaura, C. Comparative evaluation of organ and effective doses for pediatrics patients with those for adults in chest and abdominal CT examinations. Br. J. Radiol. 80, 657–667 (2007). Australian Radiation Protection and Nuclear Safety Agency. National diagnostic reference level fact sheet. ARPANA (2011). International Commission on Radiological Protection. Radiological protection in pediatrics diagnostic and interventional radiology. ICRP Publication 121. Ann. ICRP 42(2). ICRP (2013).
