Entrance Surface Dose Measurement and Lifetime Attribute Risk Analysis from Postero-anterior Chest X-ray Imaging via Direct and Indirect Measurement

Entrance Surface Dose Measurement and Lifetime Attribute Risk Analysis from Postero-anterior Chest X-ray Imaging via Direct and Indirect Measurement

1,2K^amal, I., ³S^alehhon, n., ³h^aShIm, S., ⁴m^uSa, Y. a^nd1,*a^bdul K^arIm, m. K.

ABSTRACT

The aimed of this study is to estimate the entrance surface dose (ESD) of routine chest X-ray (CXR) examination and to compare the direct and indirect measurement. This study also extended to relate with the patients body thickness and established lifetime attributable risk (LAR) of the examinations. A total of 101 CXR examinations in posteroanterior (PA) projections were selected as subjects and data, such as patient habitus and scanning acquisition parameters were recordedff. The mean ESD value obtained from the TLD-100 and the software calculation was 0.31 mGy and 0.17 mGy, respectively. The percentage deviation obtained ranged from 25.5% to 61.3%. In comparison with the national diagnostic reference level (DRL), the mean values of ESD obtained from this study were lower by a factor of 3. Furthermore, it was observed that the mean absorbed dose of the adrenals, kidneys, lungs, oesophagus, and heart walls were lower as compared to the ESD value. The examinations were only associated with very low risk of cancer incidence and mortality. Hence, the results of this study suggested a need for standardisations of the personnel training to perform X-ray examinations according to ALARA principles.

Keywords: Chest X-ray examinations; Entrance surface dose; TLD-100; radiation risk

1School of Health Sciences, KPJ Healthcare University College, Persiaran Seriemas, Kota Seriemas, 71800 Nilai, Negeri Sembilan, Malaysia

2Department of Physics, Faculty of Science, Universiti Putra Malaysia 43400 Serdang, Selangor Malaysia

3Department of Physics, Faculty of Science, Universiti Teknologi Malaysia 81310 Johor Bahru, Malaysia

4Department of Physics, Ahmadu Bello University, Zaira Nigeria, local 810211 Nigeria

*Corresponding author: Abdul Karim, M. K.

Tel: +60192140612 Fax: +60397693237

Email: mkhalis@upm.edu.my Received: 15 October 2020

Accepted for publication: 8 December 2020

Publisher: Malaysian Association of Medical Physics (MAMP) http://www.mamp.org.my/

https://www.facebook.com/MedicalPhysicsMalaysia

INTRODUCTION

Medical radiation has been acknowledge as the largest man-made source of public exposure to the ionising radiation. Although the radiation dose exposure from the X-ray examinations are relatively low, the frequency of the examinations has increase drastically and become a concern among publics. National Radiological Protection Board (NRPB) of United Kingdom in 2009 for example, has promotes the worldwide interest in patient dose measurement with the publication of Patient Dose Report in Diagnostic Radiology (Hart et al. 2009) and 208 000 dose-area product measurements along with 187 000 fluoroscopy times for diagnostic examinations or interventional procedures. In addition, patient dose data for dental X-ray examinations were included for the first time in the series of 5-yearly reviews. This article presents a summary of a key output from the NPDD - national reference doses. These are based on the third quartile values of the dose distributions for 30 types of diagnostic X-ray examination and 8 types of interventional procedure on adults, and for 4 types of X-ray examination on children. The reference doses are approximately 16% lower than the corresponding values in the previous (2000. Several agencies and authorities also has directed the initiatives to assess and report the levels of exposure to ionising radiation and their effects by establishing Diagnostic Reference Levels (DRLs) (Abdul Karim et al. 2016). According to United Nations Scientific Committee on the Effects of Atomic Radiation (UNSCEAR) report in 2000, uncontrolled use of ionizing radiation may become carcinogenic, or in other words, having the potential to cause cancer (Harun et al. 2020;

Muhammad et al. 2019; UNSCEAR, 2011).

In Malaysia, the national dose survey was conducted twice by the Malaysian government from 1993 to 1995 and from 2005 to 2009. National dose survey done by Ministry of Health (MOH) serves as reference for all centres to ensure the use of ionising radiation is monitored and optimised (Ministry of Health Malaysia, 2013). Although ionizing radiation is relevant and allow for the diagnosis of a disease, the amount of dose exposure sometimes are more than what is necessary.

Optimized dose in medical imaging are basically balance between the image quality and radiation dose, that sufficient for the extent of investigation to meet the specific clinical objectives (Harun et al. 2020). Hence, for establishing DRLs of medical imaging, the radiation dose in two dimension (2D) projection imaging is based on directly measurable dose quantity, entranced surface dose (ESD) (Shahbazi-Gahrouei, 2006). ESD is defined by International Atomic Energy Agency (IAEA) in its Technical Reports Series No. 457 as the absorbed dose in air, including the contribution from backscatter factor, assessed at a point on the entrance surface of a specified object (IAEA, 2007)

In this study, the ESD from the TLD-100 and the CalDose_X Version 5.0 software were compared to identify the accuracy of the radiation dose delivered to a patient. Subsequently, the measured values were compared to the national and international diagnostic reference levels (DRLs) recommended by different organisations. This study would provide important insight regarding the radiation dose and the risk associated with CXR examinations of the current practices.

EXPERIMENTAL METHODS

In this prospective study, the ethical approval was granted by our institution research ethics committee, with no patient consent required (NMRR-14-606-20966). The study involved a total of 101 subjects who underwent PA CXR examinations in five different public clinics in Malaysia. For every PA chest examination, each of the patients was instructed to stand facing the wal detector.

As suggested by the name of the procedure, the X-rays were irradiated from the back (posterior) of the patient’s body to the front (anterior) with three dosimeters were attached at the centre of the back of the body before examinations were performed.

DOSE MEASUREMENT AND CALIBRATION Two methods were utilized for direct and indirect ESD measurement using thermoluminesence dosimeter (TLD) (Harshaw, USA) and the CalDose_X (Ver.5.0, Brazil) software, respectively. Basically, TLD contains lithium

fluoride doped with magnesium and titanium (LiF: Mg, Ti). Its relative proportion of ⁶Li and 7Li is 7.5% and 92.5%, respectively with dimensions of 3.2 mm × 3.2 mm and 0.9 mm thickness. The dosimeter has effective atomic number of 8.2 and density is 2.64 g cm^-3 and has been well-known for its reliability, high in sensitivity and long-term stability (Abdul Karim et al. 2019) . The dosimeter were manufactured by the Harshaw Chemicals and is the most frequently used family of thermoluminescence (TL) phosphors.

All chips were calibrated at the Secondary Standard Dosimetry Laboratory (SSDL) in Malaysia. The calibration determine the TL response of the dosimeter to a measured exposure or an absorbed dose of radiation of a clearly defined energy (Freire et al. 2008). The best range of doses for medical imaging exposure is between 5 mGy to 10 mGy and therefore, 8 mGy was chosen as the calibration dose in this study (Abdul Karim et al.

2016). After irradiating with the calibration dose, all TLDs went for the readouts process using the Harshaw Thermoluminescence Dosimetry (TLD) Reader Model 3500. Each individual calibration coefficient (CC) for TLDs can be found as in Eqn. (1):

(1)

where Q_rad is gamma energy of 8 mGy from Cs-137, M is the reading from the event, M_background is the reading background of the TLD reader, and C_background is the background reading of the TLDs in the environment exposure. TLD require annealing process to be optimized for measurement process. To ensure a good repeatability, an oven with a good temperature control system and a long-term effectiveness was used. The TLDs were annealed for 1 hour at 400°C, 2 hours at 100°C, and lastly cooled down slowly to the ambient temperature.

The configuration of TLD reader presented as in Fig. 1.

FIGURE 1 Diagram of typical configuration of TLD reader (Attix 1986)

101 subjects who underwent PA chest X-ray examination from five different imaging centres; KPL Mahmoodiah, KK Batu Pahat, Poliklinik luar Hospital Sultanah Aminah JB, Hospital Pakar Sultanah Fatimah Muar and Hospital Sultanah Johor Bahru denote as H1, H2, H3, H4 and H5, respectively. In each examinations, three TLDs were attached on the skin in line of thoracic spine level T7 with medical adhesive tape and one TLD was used as background measurement.

ESD AND CANCER RISK ASSESSMENT USING CALDOSE_X SOFTWARE

The second technique for evaluating ESD was by using CalDose_X (Ver 5, Recife, Brazil) and the layout is shown in Fig. 2 (Kramer et al. 2008)two important quantities used in x-ray diagnosis, based on the output of the x-ray equipment. Additionally, the software uses conversion coefficients (CCs. Parameters such as the gender, age, type of projection, focus to detector distance (FDD), tube potential (kV), and tube current (mAs) were inserted into the software before the calculation was performed. The ESD can be calculated mathematically by using Eqn. (2) :

(2)

FIGURE 2 Layout of CalDose_X allows user inserting multitypes of parameter for evaluation of ESD and LAR

where the output of X-ray tube at 80 kV at 1 m was normalised to 10 mAs, kV is the current tube potential, FSD is the focus to skin distance in cm, mAs is the tube current and BSF is the backscatter factor. Furthermore, the risk of cancer assessment is also calculated by using the same software and the results are recorded to be analysed

STATISTICAL ANALYSIS

The data obtained was analysed using the Statistical Package for Social Sciences (SPSS, also known as IBM SPSS Statistics) software version 25.0. The normality of the calculated dose presented in the form of ESD and the effective dose was tested using the Shapiro-Wilk test.

Mean comparisons between variables were descriptively described and expressed as mean ± standard deviation (SD).

RESULTS AND DISCUSSION

PATIENT CHARACTERISTICS

Table 1 showed the characteristics of the 101 subjects from the five different centres. The range of age of the subjects was between 28 to 51 years. The mean weight of the patients was between 53.9 to 67.7 kg, whereas the

mean height was between 1.59 to 1.66 m. Meanwhile, the mean BMI of the patients was between 21.3 to 25.3 kg m^-2 and the mean body thickness was between 15.3 to 20.0 cm.

The patient characteristics were important information for the radiographers to determine the most optimal radiation dose for a patient while producing a radiographic image of good quality. For instance, the thickness of the patient’s body determines the amount of X-ray photon that should be projected. This effect is because the range of densities or contrast produced on a radiograph is created by the absorption characteristics of the anatomic tissue and affected by the composition and thickness of the tissue and cell compactness (Ang et al. 2017).

ESD MEASUREMENT

Table 2 showed the X-ray exposure parameters and the ESD measurement of all subjects. The mean X-ray

field size was ranged between 1299 to 1535 cm² and the focus-detector distance (FDD) was constant at 180 cm. The field size of the X-rays refers to the size of the field of X-rays being radiated to the patient. Meanwhile, the FDD is the distance between the X-ray tube to the detector of the X-rays. For PA CXR examinations, the standard FDD is 180 cm.

The mean tube potential was between 68.5 to 123.2 kV, while the mean tube current was between 1.29 to 4.13 mAs. It is observed that the mean ESD of this study was ranged from 0.11 to 0.64 mGy. To minimize the patient dose, the X-ray beam tube potential should be optimized. In relation to the PA CXR examinations, the tube potential could be set as low (60 to 100 kV) or high (95 to 150 kV). High kV technique allow for the use of lower beam quality and therefore decrease radiation dose to a patient. Nevertheless, as displayed in Table 2, only hospital H5 utilized a high kV technique while the remaining hospitals preferred a low kV technique.

TABLE 1 Characteristics of patients/subjects in this study

Hospital Gender Total

samples Age/years Weight/kg Height/m BMI/kg m^-2 Patient thickness/cm Male Female

H1 15 17 32 41 ± 20

(16 - 72)

67.7 ± 18.0 (40.8 - 122.0)

1.64 ± 0.11 (1.40 - 1.85)

25.3 ± 7.7 (14.1 - 46.5)

20.0 ± 2.4 (15.0 - 24.0)

H2 13 12 25 28 ± 16

(16 - 68)

53.9 ± 8.7 (42.0 - 72.0)

1.60 ± 0.01 (1.42 - 1.75)

21.3 ± 3.8 (16.3 - 33.3)

18.2 ± 2.3 (15.0 - 23.0)

H3 8 11 19 47 ± 19

(13 - 75)

61.5 ± 13.8 (35.0 - 85.0)

1.64 ± 0.08 (1.45 - 1.76)

22.9 ± 4.4 (14.7 - 31.2)

15.3 ± 1.9 (12.0 - 18.0)

H4 4 12 16 51 ± 25

(6 - 91)

57.8 ± 13.0 (36.0 - 78.0)

1.59 ± 0.09 (1.30 - 1.75)

23.0 ± 4.0 (16.6 - 31.3)

17.2 ± 2.4 (10.0 - 20.0)

H5 5 4 9 44 ± 13

(25 - 69)

65.3 ± 9.4 (50.0 - 80.0)

1.66 ± 0.07 (1.60 - 1.75)

23.5 ± 2.5 (19.0 - 26.1)

19.6 ± 2.2 (15.0 - 23.0)

TABLE 2 Scanning acquisition parameters and ESD from TLD of all subjets/patients

Hospitals Scanning parameter (mean ± SD)(min – max) Entrance Surface Dose/mGy Field Size/cm² Tube Potential/(kV Tube Current/mAs

H1 *1359 ± 233

**(900 - 1086)

80.9 ± 4.2 (74.0 - 92.0)

4.13 ± 0.84 (2.40 - 6.40)

0.24 ± 0.09 (0.11 - 0.47)

H2 1300 ± 237

(960 - 1800)

74.3 ± 3.5 (67.0 - 83.0)

2.23 ± 0.28 (1.80 - 2.80)

0.29 ± 0.08 (0.18 - 0.40)

H3 1535 ± 296

(1120 - 2025)

68.5 ± 4.9 (63.0 - 80.0)

3.61 ± 0.92 (1.60 - 5.00)

0.29 ± 0.08 (0.19 – 0.49)

H4 1299 ± 269

(1050 - 1600)

71.56 ± 16.7 (60.0 - 90.0)

4.08 ± 1.69 (2.20 - 8.50)

0.38 ± 0.15 (0.11 - 0.64)

H5 1310 ± 291

(1050 - 1800)

123.2 ± 6.7 (109.0 - 133.0)

1.29 ± 0.12 (1.25 - 1.60)

0.33 ± 0.12 (0.13 - 0.58)

COMPARISON OF ESD VALUE

Fig. 3 showed the comparison between the ESD value from the TLD-100 and the CalDose_X. The mean ESD of TLD-100 ranged from 0.24 to 0.38 mGy, while the mean ESD of CalDose_X ranged from 0.11 to 0.25 mGy.

Generally, the mean ESD from CalDose_X was lower as compared to the mean ESD from TLD-100, except for hospital H1. The ESD of CalDose_X for H1 was slightly higher than the ESD of TLD-100 by 0.01 mGy.

Meanwhile, the ESD of CalDose_X was lower compared to the direct estimation in H2, H3, H4, and H5 by a factor of 2.6, 1.9, 2.0, and 1.9, respectively.

machines used in this study, however, had a filtration of more than 2.5 mm aluminum for exposure at 70 kV.

Thirdly, the parameters that could be inserted by the users into the CalDose_X software were limited. In this study, only the gender, age, type of projection, FDD, kV, and mAs could be keyed into the software, failing to include other important parameters for the calculation of ESD. The ESD and backscatter factor were calculated using the provided settings in the software (Musa et al.

2017).

Fig. 4 portrayed the comparison between the ESD obtained from the TLD-100, the national DRLs, and the international DRLs recommended by different organisations. In this comparison, the ESD values demonstrated by all hospitals were below the reference level set by the Ministry of Health Malaysia (MOH).

The mean ESD of H1, H2, H3, H4, and H5 were lower than that of MOH by a factor of 3.8, 3.1, 3.1, 2.4, and 2.7, respectively. IAEA BSS which showed a mean ESD of 0.40 mGy was higher than the mean ESD of this study by a factor of 1.7, 1.4, 1.4, 1.1, and 1.2 for H1, H2, H3, H4, and H5, respectively. The mean ESD of UK NRPB, EC EUR96, and UK IPSM reports were the same which was 0.30 mGy. The ESD of two hospitals, H4 and H5, exceeded this value by a factor of 1.3 and 1.1, respectively. Among the national and international DRLs, the lowest mean ESD was recommended by the American Association of Physicists in Medicine (US- AAPM) which was 0.25 mGy. H1 was the only hospital that succeeded to achieve the lowest mean ESD of 0.24 mGy and this value was lower than that of the US AAPM by a factor of 1.04. Generally, the mean ESD from all hospitals in this research were lower as compared to the national DRL, whereas only one hospital demonstrated a mean ESD which was lower than one of the international DRLs.

Table 3 shows the mean absorbed dose of five organs and tissues and the risk of cancer incidence and mortality for all five hospitals calculated by CalDose_X Version 5.0 software. The absorbed dose or simply referred to as the organ dose is the mean energy imparted to matter of mass by ionising radiation.

In general, the mean absorbed dose of the adrenals, kidneys, lungs, oesophagus, and heart walls were 0.047 mGy, 0.042 mGy, 0.050 mGy, 0.041 mGy, and 0.029 mGy, respectively. Compared to the ESDs calculated by the software in Fig. 1, the mean organ or tissue absorbed doses were lower. This finding could be explained by the ESD which refers to the absorbed dose in the air including the contribution from the backscatter factor.

Therefore, only a certain amount of photons could penetrate through the body to allow visualization of an organ.

Additionally, the risk of cancer incidence from the examination was 0.150 cases per 100,000 people and FIGURE 3 Comparison between ESD from TLD-100 and

CalDose_X software

The percentage deviation ranged from 25.5% to 61.3%. Specifically, the percentage deviations were 25.5%, 61.3%, 47.0%, 50.9%, and 48.6% for H1, H2, H3, H4, and H5, respectively. In general, the percentage deviation of ESD was higher than the acceptable value which was 30%. Only H1 had a percentage deviation of 25.5%, which was slightly lower than the acceptable value. The highest percentage deviation was H2 (61.3%) due to its lowest ESD of CalDose_X.

The higher percentage deviation demonstrated in this research could be due to several factors contributed by the CalDose_X software. Firstly, every X-ray tube emits its own X-ray emission spectrum, whereby different exposure and kV settings on the control panel give rise to a different spectrum (Porto et al. 2014). Nevertheless, CalDose_X simulates and calculates the ESD using only one type of spectrum as standardisation that set at 70 kV. The software also limited to only one X-ray tube filter that is 2.5 mm aluminum. It is known that X-ray tubes have different beam quality, therefore requiring a different filter for the X-ray projection. Most of the X-ray

the risk of cancer mortality was 0.124 cases per 100,000 people. The incidence rate reveals the estimated number of people diagnosed with cancer per 100,000 people, whereas the cancer mortality rate presents the estimated number of deaths from cancer per 100,000 people. The legalized annual dose limit for the public is 1 mSv per year . Based on the results of ESDs, despite the wide variations among the hospitals, it can be concluded that no hospital exceeded the reference levels. The cancer risk assessment also established that the incidence and mortality of cancer associated with the PA CXR examinations were very low.

CONCLUSION

In this study, the ESDs obtained from TLD-100 and CalDose_X were compared to the national DRL, The mean value of ESD was lower by a factor of 2.9 with national DRL but comparable and within the range of other established international DRLs. The risk of cancer incidence and mortality rates associated with the use of PA CXR examinations were very low, establishing the safety of this procedure for all patients. Furthermore, the

high kV technique should be implemented in all imaging centres to reduce the harmful effect of radiation dose exposure while producing a good quality radiographic images.

REFERENCES

Abdul Karim, M. K., Hashim, S., Bakar, K. A., Muhammad, H., Sabarudin, A., Ang, W. C. &

Bahruddin, N. A. 2016. Establishment of Multi- Slice Computed Tomography (MSCT) Reference Level in Johor, Malaysia. J. Phys. Conf. Ser. 694:

012033.

Abdul Karim, M. K., Hashim, S., Sabarudin, A., Bradley, D. A. & Bahruddin, N. A. 2016. Evaluating Organ Dose and Radiation Risk of Routine CT Examinations in Johor Malaysia. Sains Mal. 45:

567-573.

Abdul Karim, M. K., Rahim, N. A., Matsubara, K., Hashim, S., Mhareb, M. H. A. & Musa, Y. 2019.

The Effectiveness of Bismuth Breast Shielding with Protocol Optimization in CT Thorax Examination.

J. Xray. Sci. Technol. 27: 139-147.

FIGURE 4 Comparison between ESD in this study and DRLs recommended by different international organisations (Green:Hospitals, Pink: Malaysian DRL)

TABLE 3 The organ and tissue absorbed dose and risk of cancer assessment Organ/Tissue Absorbed Dose/mGy *Risk of Cancer

Incidence *Risk of Cancer Mortality Adrenals Kidneys Lungs Oesophagus Heart wall

0.047 0.042 0.05 0.041 0.029 0.15 0.124

*cases per 100,000 populations

Ang, W. C., Hashim, S., Abdul Karim, M. K., Ashiqin, N. & Salehhon, N. 2017. Adaptive Iterative Dose Reduction (AIDR) 3D in Low Dose CT Abdomen- Pelvic : Effects on Image Quality and Radiation Exposure. J. Phys. Conf. Ser. 851(1): 012006.

Attix, F. H. 1986. Introduction to Radiological Physics and Radiation Dosimetry. Weinheim: WILEY-VCH Verlag GmbH & Co. KGaA.

Freire, L., Calado, A., Cardoso, J. V., Santos, L. M. &

Alves, J. G. 2008. Comparison of LiF (TLD-100 and TLD-100H) Detectors for Extremity Monitoring.

Radiat. Meas. 43: 646-650.

Hart, D., Hillier, M. C. & Wall, B. F. 2009. National Reference Doses for Common Radiographic, Fluoroscopic and Dental X-Ray Examinations in the UK. Br. J. Radiol. 82: 1-12.

Harun, H. H., Abdul Karim, M. K., Abbas, Z., Abdul Rahman, M. A., Sabarudin, A. & Ng, K. H. 2020.

Association of Radiation Doses and Cancer Risks from CT Pulmonary Angiography Examinations in Relation to Body Diameter. Diagnostics 10(9): 681.

Harun, H. H., Abdul Karim, M. K., Abbas, Z., Sabarudin, A., Muniandy, S. C., Razak, H. R. A. & Ng, K.H., 2020. The Influence of Iterative Reconstruction Level on Image Quality and Radiation Dose in CT Pulmonary Angiography Examinations.

Radiat. Phys. Chem. https:// doi.org/10.1016/j.

radphyschem.2020.108989.

International Atomic Energy Agency. 2007. Dosimetry in Diagnostic Radiology: An International Code of Practice, Technical Reports Series No. 457. Vienna:

IAEA.

Kramer, R., Khoury, H. J. & Vieira, J. W. 2008.

Caldose_X - A Software Tool for the Assessment of Organ and Tissue Absorbed Doses, Effective Dose and Cancer Risks in Diagnostic Radiology. Phys.

Med. Biol. 53(22): 6437-6459.

Ministry of Health Malaysia. 2013. Guidelines in Malaysia Diagnostic Reference Levels in Medical Imaging (Radiology). MOH Malaysia: Putrajaya.

Muhammad, N. A. A., Abdul Karim, M. K., Hassan, H. A. A., Kamarudin, M. A. A., Wong, J. H. D.,

& Ibrahim, M. J. J. 2019. Estimation of Effective Dose and Organ Cancer Risk from Paediatric Computed Tomography Thorax – Abdomen - Pelvis Examinations. Radiat. Phys. Chem. 165: 108438.

Musa, Y., Hashim, S., Abdul Karim, M. K., Bakar, K.

A., Ang, W. C. & Salehhon, N. 2017. Response of Optically Stimulated Luminescence Dosimeters (Oslds) Subjected to X- Ray Photons in Diagnostic Range. J. Phys. Conf. Ser. 851: 012001.

Porto, L., Lunelli, N., Paschuk, S., Oliveira, A., Ferreira, J. L., Schelin, H., Miguel, C., Denyak, V., Kmiecik, C., Tilly, J. & Khoury, H. 2014.

Evaluation of Entrance Surface Air Kerma in Pediatric Chest Radiography. Radiat. Phys. Chem.

104: 252-259.

Shahbazi-Gahrouei, D. 2006. Entrance Surface Dose Measurements for Routine X-Ray Examinations in Chaharmahal and Bakhtiari Hospitals. Iran. J.

Radiat. Res. 4(1): 29-33.

UNSCEAR. 2011. Report of the United Nations Scientific Committee on the Effects of Atomic Radiation 2010. United Nations Publication: Vienna.