| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Full paper |
1 Department of Radiology, University Hospital Maastricht, P Debijelaan 25, 6229 HX Maastricht 2 Interfaculty Reactor Institute, Delft University of Technology, Mekelweg 15, 2629 JB Delft, The Netherlands
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
In the case of double contrast barium enema (DCBE) examinations, the contribution to the collective dose is relatively high, about 15% [1, 2], suggesting a large potential benefit of optimization of the radiological procedure for this type of examination. It is not surprising, therefore, that nearly all possible technical measures to reduce patient dose have been investigated or applied, including the use of a high X-ray tube voltage [1, 3], modification of the voltagecurrent characteristic used during fluoroscopy towards higher voltages or lower currents [35], limiting the exposure rate during fluoroscopy [6], increasing the filtration of the X-ray beam [1, 7], increasing the sensitivity of the image intensifier chain [6, 8], removal of the anti-scatter grid [9] and the use of screenfilm combinations of a higher speed class [3]. Comparisons between digital and non-digital systems have also been made [1013]. A considerable part of the information on exposure conditions and patient dose in DCBE examinations comes from these studies. Regarding the medical aspects of the procedure, it has been shown that fluoroscopy time [10, 14], number of images [10, 14], the use of protocols and internal audits [15] have a large impact on patient dose. In addition, the replacement of anteroposterior (AP) projections by posteroanterior (PA) projections also affects patient dose [16]. Many steps that reduce patient dose will usually also reduce occupational dose. Additional reduction in occupational dose can be achieved by proper shielding against scattered X-rays.
Dosimetric evaluation of a particular set of technical parameters, or some particular protocol for DCBE examinations, is complicated because a series of projections is usually required to visualize the whole colon. The situation is complicated further by a considerable variability in the way the procedure is performed medically, with a corresponding variability in X-ray exposure. This variability finds its origin both in differences between patients and in the personal preferences of the radiologist. Several investigators have tried to solve the resulting dosimetric problem by closely monitoring a number of DCBE examinations to derive the major projections and their characteristics [1, 2, 3, 17]. This approach, however, is time consuming, which limits the number of examinations that can be evaluated. And even then, approximations cannot be avoided. In the present study we propose a different method, which is easier to perform and more suitable for monitoring larger series of examinations. The principle is that the measured total dosearea product (DAP) is distributed over a set of hypothetical projections that cover the relevant part of the abdomen. This distribution is based on a key derived from entrance doses measured with thermoluminescent dosemeters (TLDs) on the skin of the patient at the centre of each projection. To facilitate the calculation of effective dose from the various DAP values obtained in this way, new DAP to effective dose conversion coefficients (ECCs) were derived using Monte Carlo radiation transport calculations.
To our knowledge, no occupational doses have been published for DBCE thus far. As DCBE examinations are performed frequently, it was decided also to quantify entrance doses as well as the effective dose to the physician performing the examination.
| Materials and methods |
|---|
|
|
|---|
Patients and colon examination technique
From October 1999 to March 2000, patient and X-ray exposure data were acquired for 150 patients referred for DCBE examination of the colon. Patient parameters such as age, height and weight were recorded. Among the 150 examinations, 50 were studied in greater detail, as described below.
DCBE examinations of the colon are performed in the following way. All patients undergo a preparation to clean the colon. After introduction of a canula into the rectum with the patient in left lateral decubitus position, the barium sulphate suspension is introduced to fill the colon up to the flexura lienalis. Air is insufflated and the patient is rotated and moved between the erect and the Trendelenburg position to achieve proper coating of the entire colon wall. A spasmolytic drug (Glucagon) is intravenously given to improve the relaxation of the bowel musculature. Fluoroscopy is used to monitor the coating and to select projections for detail and overview images. Spot films from the rectosigmoid in prone, left and right posterior oblique, and left lateral position are obtained. Whenever necessary, an additional right lateral image is made. Different spot images of the colon and caecum in the supine and oblique position are made, as well as images of both flexures in the upright position. Two large overview radiographs are obtained using the ceiling-mounted tube and conventional film with the patient in the prone and supine position. Personnel remain behind protective barriers when these overviews are taken. During the examination, the physician will collimate the X-ray fields as much as possible and the primary beam will normally not expose the male testes. In our university hospital, nearly all DCBE examinations are performed by a resident supervised by a board certified radiologist.
Dosimetric measurements on patients
In 150 DCBE examinations, DAP values were obtained with a flat transparent ionization chamber (Diamentor M3; PTW, Freiburg, Germany). The chamber was mounted directly on the diaphragm of the X-ray tube in the primary beam of the Pantoskop. A calibration was performed before putting the system into use.
Additional measurements were performed on the skin of 50 patients using TLDs consisting of LiF chips of 3 x 3 x 0.9 mm3 (TLD 100; Harshaw, Solon, USA). The TLDs were calibrated in air kerma. 12 regions were considered in the abdomen, i.e. four quadrants in anterior view, four in posterior view, two in left view and two in right view (Figure 1
; Table 1
). Sets of three TLD chips were placed in the centre of each of these regions before the start of the DCBE examination and were removed afterwards. The DAP of the ceiling X-ray tube was estimated for each separate exposure by measuring the entrance dose with an additional set of TLDs and using the (automatically selected) field size as derived from the exposed 35 cm x 43 cm film, with correction for magnification. The TLD reading was converted into air kerma free in air by correcting for backscatter using a factor of 1/1.3 [18]. This suboptimal method to measure DAP for the overcouch tube was applied because another flat ionization chamber for DAP measurements was not available at the time of the experiments.
|
|
|
|
In the next step, effective dose E was estimated according to
|
|
Monte Carlo calculations
The main incentive for the present approach was the wish for a simpler method for estimating DAPs for the various projections applied in DCBE. A second consideration was that we also wanted other ECCs than those most frequently used thus far [19]. The reason is that several DCBE projections in NRPB Report 262 [19] (partly) include the testes, while this is normally not the case, at least in our hospital. The projections in the new simulations, therefore, do not include the male testes.
The present ECC calculations were performed at the Medical Physics Department of the Interfaculty Reactor Institute using the standard radiation transport code MNCP version 4B [20] and the mathematical phantoms ADAM and EVA to represent male and female adult patients [21, 22]. Input X-ray spectra were generated with the IPEM 78 spectrum processor [23]. The projections that were simulated are given in Table 1
and were derived by inspecting several 35 cm x 43 cm overview films showing the position of the colon.
Occupational dosimetry
The physician performing the DCBE, who is either a resident or a radiologist, wears a 0.35 mm Pb equivalent apron. Some physicians wear a thyroid collar, but most do not.
The dosimetry protocol for measuring occupational exposure is in principle identical to that described in earlier studies [24, 25]. In the current approach, the physician wears sets of TLDs on the forehead (glabella), neck (thyroid), upper arms ( x 2), lower legs ( x 2) and right hand.
The transmission of the apron was measured under clinical circumstances by using TLDs at identical positions at the inside and outside of the protective clothing, complemented with some transmission measurements in the primary beam at different tube potentials. Aprons with TLDs were worn during a 4-week period to obtain readable doses for the shielded TLDs. Also during this experiment, the relationships between the unshielded dose at the neck and that at the thorax and abdomen were determined. In the individual studies, these relationships were used to calculate the thorax and abdomen entrance dose from the dose measured at the neck. Effective dose was calculated as described previously [25]. Entrance doses were calculated as tissue absorbed dose, according to "TLD-result in terms of air kerma" multiplied by (µen/
)tissue/(µen/
)air, i.e. the ratio of the mass energy absorption coefficients of tissue and air. For diagnostic X-ray energies, this factor equals approximately 1.06 [26].
| Results |
|---|
|
|
|---|
|
|
The results of the Monte Carlo simulations are presented in Table 4
under the heading ECC. Effective dose E to the patients was calculated using these new conversion coefficients. The results of these calculations are shown in Table 3
for male and female patients separately (last column). The effective dose averaged over male and female patients was 6.4±2.1 mSv. The contribution of each of the 12 projections to the average effective dose is shown in Table 4
. Mean total DAP in the 50 cases investigated in detail was 44±22 Gy cm2.
|
Table 5
gives dosimetric data for the physician (resident or radiologist) performing the DCBE examination. The results presented are entrance doses, expressed in tissue. It is noted that TLD readings were very small, as may be seen from a comparison with the standard deviation of unexposed chips used for background determination during the time of the experiments: the average standard deviation of 19 sets of TLDs, each comprising 6 chips, was 5±2 µGy.
|
Effective dose E to the worker was estimated for a few scenarios using entrance dose to organ dose conversion factors derived in a previous study [25]. Wearing a 0.35 mm Pb equivalent apron, which is the common situation, results in an effective dose of 0.52 µSv. When a 0.5 mm Pb equivalent thyroid collar is worn in addition to the 0.35 mm Pb equivalent apron, which happens infrequently, E is 0.32 µSv. Applying no protection at all, which does not occur in practice, would have resulted in E=3.9 µSv. A protection factor can be calculated from these figures: wearing a 0.35 mm Pb apron reduces effective dose by a factor of 8 compared with wearing no apron, while an apron plus a 0.5 mm Pb thyroid collar reduces E by a factor of 12. The contribution to the physician's effective dose by the various body parts that were distinguished in our occupational dosimetry [24, 25] is shown in Table 6
.
|
| Discussion |
|---|
|
|
|---|
If TLD facilities are available, the present method of patient dosimetry in DCBE is reasonably convenient. However, like in any other approach using a limited number of fixed projections where in practice an infinite amount of different projections can be used, no great accuracy can be claimed. Nevertheless, it is believed that the proposed method accounts for the major anisotropy in radiation sensitivity in the abdomen, e.g. for the differences existing in AP, PA and lateral sensitivities. Oblique projections are approximated with AP/PA or lateral projections. A rather rough comparison between ECCs can be made with the average of all barium enema related projections published by the NRPB [17, 19] and the average of the present values. In doing this, the rectum AP and left posterior oblique projections from the NRPB were excluded because they incorporate high contributions to the effective dose from the testes. The average of the remaining 11 NRPB projections is 0.15 mSv/(Gy cm2) for 80 kV X-rays filtered with 4 mm Al, which is identical to the average of all our male and female projections (80 kV and 90 kV, 3.5 mm Al).
Several authors have published DAP figures for DCBE examinations. These data have been summarised in Table 7
. From this table it is seen that the data for our hospital are among the higher values reported in the literature, although our total DAP value of approximately 56 Gy cm2 is still below the reference level of 60 Gy cm2 recommended in a document by the Institute of Physical Sciences in Medicine [27]. DAP during detailed monitoring was lower than during the less conspicuous surveillance in which only DAP was recorded by radiographers. Awareness of being monitored is a strong incentive to minimize the use of radiation, as has been observed by several other investigators (see e.g. [15, 29]). Our values for DAP being high has a number of causes. First, our fluoroscopy time is relatively long, which is probably related to the fact that residents perform nearly all DCBE examinations. Second, relatively low tube voltages are used, with little additional filtration of the X-ray beam. The entrance dose rate, with the anti-scatter grid in front of the image intensifier, was 0.25 µGy s-1 for the field size most often used. This value is relatively low, as can be seen from a comparison with the values reported by Geleijns et al (on average 0.44 µGy s-1), where it has to be noted that their dose rates were measured without an anti-scatter grid [1].
|
|
It is also interesting to look at the relationship between effective dose and DAP because it would be very practical if one could estimate effective dose from a relatively easily accessible quantity such as the total DAP. In this study the average ratio E/DAP was 0.15 mSv/(Gy cm2). Geleijns et al [1] reported for two units, both applying high X-ray tube voltages and 0.1 mm or 0.2 mm additional Cufiltration, values of 0.27 mSv/(Gy cm2) and 0.29 mSv/(Gy cm2). Hart and Wall [17] reported similar values. Lampinen et al [2] derived expressions for this ratio with patient weight as an independent variable. Using these relationships, the sex average for a male and female patient of, respectively, 70 kg and 58 kg (the weight of the mathematical phantoms ADAM and EVA [22]) is calculated as 0.29 mSv/(Gy cm2). However, considerably lower ratios are calculated from the data of Geleijns et al [1] for the units using either lower voltages or less filtration (no Cu) than the two systems that the values of 0.27 mSv/(Gy cm2) and 0.29 mSv/(Gy cm2) were given for. Using data given in their Table 5
and Figure 2 [1] for seven units, E/DAP ratios between 0.12 mSv/(Gy cm2) and 0.17 mSv/(Gy cm2) are found, similar to our value. In current clinical practice, a single ratio for E/DAP is clearly insufficient to estimate effective dose from total DAP. As can be concluded from the work of Hart et al [19] and Geleijns et al [1], one has at least to account for radiation quality (kV and filtering). In addition, the testes being inside or outside the primary beam has a large impact on this ratio. In our experience, the testes are nearly always outside the primary beam, as is also suggested by the large difference in effective dose between male and female patients in the studies by Geleijns et al [1] and Lampinen and Rannikko [2]. The potential effect of other differences in clinical protocols applied in daily practice is less clear. The effect of patient weight on E/DAP has been investigated by Lampinen and Rannikko [2].
The findings from the present study suggest that optimization of technical and clinical factors may still lead to a substantial reduction in patient dose in DCBE examinations in our hospital. In fact, fluoroscopy and radiography are now being performed at higher X-ray tube voltages and the beam filtration has been increased. The effects of these modifications on image quality and patient exposure are currently being investigated. It is expected that improving radiation consciousness of radiological residents by additional education will bring down fluoroscopy times.
| Conclusion |
|---|
|
|
|---|
A new and relatively simple method was applied to distribute measured total DAP over a number of projections that modelled exposure in DCBE. This distribution was based upon entrance doses measured with TLDs instead of on observations by a human observer. Variations in the conversion factors from DAP to patient effective dose by a factor of about two were found for DCBE examinations, mainly owing to differences in radiation quality and the testes being inside or outside the primary beam. In the present study, dose conversion factors for a complete examination for women are about twice as large as those for male patients. Total DAP and fluoroscopy time in our hospital were among the higher values reported in the literature. This finding initiated some corrective actions that are currently under evaluation.
Received for publication August 22, 2000. Revision received November 13, 2000. Accepted for publication December 21, 2000.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. A. Mettler Jr, W. Huda, T. T. Yoshizumi, and M. Mahesh Effective Doses in Radiology and Diagnostic Nuclear Medicine: A Catalog Radiology, July 1, 2008; 248(1): 254 - 263. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. M. McCartney Radiation Dose During Barium Enema Radiol. Technol., September 1, 2007; 79(1): 91 - 93. [Full Text] [PDF] |
||||
![]() |
J. Barentsz, S. Takahashi, W. Oyen, R. Mus, P. De Mulder, R. Reznek, M. Oudkerk, and W. Mali Commonly Used Imaging Techniques for Diagnosis and Staging J. Clin. Oncol., July 10, 2006; 24(20): 3234 - 3244. [Abstract] [Full Text] [PDF] |
||||
![]() |
REFERENCES J. ICRU, December 1, 2005; 5(2): 103 - 113. [Full Text] [PDF] |
||||
![]() |
S. A. Taylor, S. Halligan, C. I. Bartram, P. R. Morgan, I. C. Talbot, N. Fry, B. P. Saunders, K. Khosraviani, and W. Atkin Multi-Detector Row CT Colonography: Effect of Collimation, Pitch, and Orientation on Polyp Detection in a Human Colectomy Specimen Radiology, October 1, 2003; 229(1): 109 - 118. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| BJR | DMFR | IMAGING | ALL BIR JOURNALS |