| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiation Physics, Unit 94, The University of Texas M. D.
Anderson Cancer Center, Houston, TX 77030.
2 Department of Imaging Physics, Unit 56, The University of Texas M. D. Anderson
Cancer Center, 1515 Holcombe Blvd., Houston, TX 77030.
Received July 28, 2004; accepted after revision October 15, 2004.
Address correspondence to D. D. Cody.
OBJECTIVE. The purpose of this study was to focus attention on the technique factors commonly used in survey CT scans (e.g., scout, topogram, or pilot scans) to measure the radiation exposure from typical survey CT scans, to compare their exposure to that of typical chest radiographs, and to explore methods for radiation exposure reduction.
MATERIALS AND METHODS. The default survey CT scans on 21 CT scanners, representing three different vendors and 11 different models, were investigated. Exposure measurements were obtained with an ion chamber at isocenter and adjusted to be consistent with standard chest radiographic exposure measurement methods (single posterior–anterior projection). These entrance exposures were compared with those of typical chest radiographs, for which the mean for average-sized adults is 16 mR (4.1 x 10–6 C/kg).
RESULTS. The entrance exposures of the default survey CT scans ranged from 3.2 to 74.7 mR (0.8 to 19.3 x 10–6 C/kg), which is equivalent to approximately 0.2 to 4.7 chest radiographs. By changing the default scan parameters from 120 kVp to 80 kVp and the tube position from 0° (tube above table) to 180° (tube below table), the entrance exposure for the survey CT scan was reduced to less than that of one chest radiograph for all CT scanners.
CONCLUSION. For institutions at which the interpreting radiologists do not rely heavily on the appearance of the survey CT image, we recommend adjusting the technique parameters (kilovoltage and X-ray tube position) to decrease radiation exposure, especially for vulnerable patient populations such as children and young women.
The current medical climate requires careful consideration of radiographic techniques for pediatric CT examinations, adapting the CT parameters for the patient's size [1–7]. Furthermore, a single routine CT protocol is not appropriate for all adults or for all imaging tasks. Very large adults often require a substantially higher CT technique to produce clinically acceptable image quality. CT parameters also may need adjustment for specific indications (e.g., renal stone protocols). Recently, the CT community has begun to recognize that small adults would benefit from a more size-appropriate CT technique [8–17]. In addition to the X-ray tube current and scanning time/rotation, CT operators can adjust the X-ray tube voltage, pitch, or some combination of these parameters according to patient size factors and the imaging task.
The contribution of the survey CT scan (also referred to as the scout, topogram, or pilot scan) to the total radiation dose of a CT examination typically has been considered negligible. While the dose from the survey CT scan is quite small compared with the dose from the tomographic portion of the examination, the former readily can be measured and compared with typical doses from chest radiographs. The effects of radiation exposure at the level of chest radiographs is not well understood; therefore, we should follow the principle of ALARA (as low as reasonably achievable) and minimize the dose from the survey CT scan whenever possible. This issue generally has been ignored by the imaging physics community.
Our study was designed to evaluate the radiation exposure from commonly used survey projection-view CT scans for several CT manufacturers and models and to evaluate how to minimize the radiation exposure by taking advantage of adjustments to X-ray tube kilovoltage, X-ray tube current, and X-ray tube position parameters.
Materials and Methods
All CT scanners at our institution were available for inclusion in this investigation, resulting in the evaluation of 21 CT scanners representing three vendors and 11 models. We evaluated models from GE Healthcare, Philips Medical Systems, and Siemens Medical Solutions. Single-slice helical CT scanners and MDCT scanners (4-, 8-, and 16-channel systems) were investigated. Through the manufacturers' manuals, discussions with manufacturers' representatives, and the default settings of each CT scanner, we collected information on the design and defaults of each CT scanner (e.g., the number of channels, default kVp for survey CT scans, default mA for survey CT scans, default tube positions for survey CT scans, survey CT scan table speed, and survey CT scan beam filtration). We had modified the survey CT scan parameters more than 1 year ago for pediatric patients on the GE LightSpeed scanners to minimize the radiation exposure. The pediatric exposures from the GE LightSpeed scanners are from these user-selected parameters, and not the manufacturer's default parameters.
|
|
|
We measured the radiation exposure-in-air from three different survey CT scan techniques on every CT scanner. (The measurement technique will be thoroughly described in the following paragraph.) We measured each technique (kVp and mA combination) at the two tube positions. The first technique we measured was the default X-ray tube voltage and default X-ray tube current at anteroposterior (0°, or A/P) and posteroanterior (180°, or P/A) X-ray tube positions. If the default child and adult techniques were not identical, then we collected data for both. Second, we examined a technique of 120 kVp and the minimum X-ray tube current at 0° and 180° X-ray tube positions. If the 120 kVp setting was not available, we chose the 130 kVp setting instead. Third, we measured a technique of 80 kVp and the minimum X-ray tube current at 0° and 180° X-ray tube positions. To provide a fair comparison between the 120 kVp and 80 kVp data, the minimum X-ray tube current chosen was the minimum current that could be used at both voltage levels. For example, if the lowest X-ray tube current available with 80 kVp was 40 mA but the minimum available with 120 kVp was 10 mA, 40 mA was used for both measurements. For all techniques we used the parameters associated with a survey CT scan of the body. We believe that our results should also apply to survey CT scans of the head, because we found no difference in beam filtration between head and body survey CT scans.
|
To compare the measurements taken at the CT gantry isocenter to the entrance exposure measurements for a chest radiograph, inverse-square corrections were applied to the collected exposure readings. Standard adult chest radiographic exposure measurements are obtained by placing the ion chamber 22.5 cm in front of the imaging detector [18]. For our measurements, the source-to-isocenter distance was known for each CT scanner and the patient was assumed to be centered at isocenter. Assuming a thickness of 22.5 cm (A/P) for the adult chest, the readings were corrected to the anterior surface for 0° tube positions and to the posterior surface for 180° tube positions (Fig. 3). We selected 14 cm to represent the chest wall thickness of a child, based on the thickness of our "5-year-old pediatric" anthropomorphic phantom (ATOM Dosimetric Phantoms, Computer Imaging Reference Systems). The pediatric exposure estimations were corrected to the anterior and posterior surfaces of a 14-cm-thick pediatric chest.
In the most recent Nationwide Evaluation of X-ray Trends (NEXT) study, radiation exposure to an average-sized adult from a typical chest radiograph ranged from less than 5 mR (1.3 x 10–6 C/kg) to above 31 mR (8.0 x 10–6 C/kg), with a mean value of 16 mR (4.1 x 10–6 C/kg) [18]. This is slightly lower than the mean exposure reported in the previous NEXT study in 1984 of 20 mR (5.2 x 10–6 C/kg) [19]. For our study, we assumed a midrange value of 16 mR (4.1 x 10–6 C/kg) per adult chest radiograph.
Factors other than X-ray tube voltage, X-ray tube current, and X-ray tube position can affect the survey CT scan radiation exposure. We also investigated the survey CT scan beam width, the survey CT scan beam filtration, and the survey CT scan table speed for each CT platform. We chose to operate the CT scanners in clinical mode rather than service mode to be certain our measurements corresponded to typical clinical CT scanner usage. In clinical mode the table moves during the scout scan, so the physical beam width cannot be measured by placing film on the table. Instead, the physical beam width at isocenter in survey CT scan mode for each scanner model was measured using a foam block suspended in the bore of the CT scanner. A piece of Kodak X-Omat V or X-Omat TL ready-pack film (Eastman Kodak Company) was placed on it horizontally at the isocenter (Fig. 4). A survey CT scan was then taken with the table positioned away from the film, and the beam width was measured using a 7x magnifier with graticule (a network of lines, each marking 0.1-mm spacing, in the focal plane of the eyepiece of the optical magnifier). The table speed and beam filtration were determined by examination of the CT scanner model specifications and by discussions with manufacturer's representatives when necessary. This information was used to select exposure time(s) for appropriate optical density for the beam width measurements.
Results
The projection scan default values are displayed in Table 1. For 8 of 11 models, the default X-ray tube position was set at the 0° position for adult survey CT scans, that is to say, typically A/P projections and not P/A projections were part of routine adult CT protocols for supine patients. For 6 of 11 models, the projection scan default values were the same for adult and pediatric patients. With these settings, the radiation exposure ranged from 3.9 mR to 74.7 mR (1.0 x 10–6 C/kg to 19.3 x 10–6 C/kg) for an average adult and 3.2 mR to 47.6 mR (0.8 x 10–6 C/kg to 12.3 x 10–6 C/kg) for an average pediatric patient (Fig. 5). Therefore, we estimated that the radiation exposure from a default survey CT scan was equivalent to that from 0.2 to 4.7 adult chest radiographs, assuming 16 mR (4.1 x 10–6 C/kg) per chest radiograph. The projection view technique (combination of X-ray tube voltage, current, and position) was found, as expected, to be consistent for scanners of the same make and model. Survey CT scan beam widths ranged from 2 mm to 7.5 mm (Table 2). Table speeds were 75 mm/sec or 100 mm/sec. The GE and Siemens models used a beam-shaping filter typically used for body tomographic scans during the survey CT scans. The three Philips platforms used only inherent filtration (no additional filters) for their survey CT scans. We expected that wider beam widths and slower table speeds would increase the measured radiation exposure. However, we did not observe a clear relationship between beam width, table speed, and radiation exposure.
|
|
|
The measured entrance exposure values observed when varying X-ray tube voltage and current parameters for each scanner included in the investigation are shown in Table 3 and Figure 6. For two models, only 130 kVp was available for survey CT scans. For a third model, only a 120 kVp and 50 mA combination could be used for survey CT scans. By using a 180° tube position, projection view exposure was reduced by a mean of 20% relative to the standard 0° technique. By using this method, the exposure was reduced by a mean of 6.2 mR (1.6 x 10–6 C/kg, range = 3.4 mR to 13.0 mR [0.9 to 3.3 x 10–6 C/kg]) at 120 kVp and by a mean of 2.2 mR (0.6 x 10–6 C/kg, range = 1.6 mR to 3.1 mR [0.4 to 0.8 x 10–6 C/kg]) at 80 kVp. When the X-ray tube voltage was reduced to the minimum possible value (from 120 kVp to 80 kVp for most scanners), the exposure for both the 0° and the 180° techniques was reduced by a mean of 64% (reduced by approximately 1.1 chest radiographs). When both the minimum X-ray tube voltage and the 180° geometry were used, the exposure due to the projection view was reduced by a mean total of 71% (or 20.7 mR [5.3 x 10–6 C/kg], reduced by approximately 1.3 chest radiographs) among all scanners and models. Switching from a 120 kVp, 0° technique to an 80 kVp, 180° technique resulted in entrance exposure levels of 5.8 mR (1.5 x 10–6 C/kg) to 12.2 mR (3.2 x 10–6 C/kg), or 0.4 to 0.8 chest radiographs.
|
|
Discussion
We found that the default survey CT scan protocols were fairly uniform for each vendor. The Philips scanners used a default 180° tube position but had high X-ray tube voltage and/or current values. All GE Healthcare scanners used a default 0° tube position and 120 kVp. The single-slice GE helical scanners (HiSpeed) had a large range of user-selectable X-ray tube currents from 10 mA to 80 mA. The multislice helical GE LightSpeed scanners had the same default parameters for adult patients, using 120 kVp, 10 mA, and 0° tube positions. The pediatric default parameters on the GE LightSpeed scanners had been user-selected during the previous year. We had only one Siemens model (two scanners) available; it also had a 0° projection as its default setting. The Siemens Somatom Sensation 16 provided different default scanning protocols for adults and children. The definition of child and pediatric varies from institution to institution. At our clinic, pediatric patients are patients ages 15 years or younger, unless they are obviously almost fully developed at a younger age. In this article, pediatric patients are those patients 15 years or younger whose display field-of-view (DFOV) was 35 cm or less.
We predicted that increasing the beam width and decreasing the table speed would increase the radiation exposure from survey CT scans. However, we did not observe that variations in beam width or table speed had a clear effect on exposure, likely owing to the many complicating factors (X-ray tube voltage, current, and position). Beam filtration also had no discernable effect between scanners, probably because a similar filter was used for all scanners. The GE LightSpeed series had a greater table speed and beam width than the GE HiSpeed FX/i series but the same default survey CT settings; however, the GE HiSpeed FX/i scanner had much lower radiation exposure measurements than the GE LightSpeed scanners. The faster table speed of the GE LightSpeed (100 mm/sec vs 75 mm/sec) might not fully compensate for its wider beam width (7.5 mm vs 2 mm).
The radiation exposure associated with the survey CT scans generally has been considered negligible in the CT physics literature or has been ignored altogether. While this exposure is small compared with that from the tomographic data acquisitions, the radiation exposures from the default survey CT scans in our investigation typically were equivalent to 0.2–4.7 chest radiographs (Fig. 5). Within the same make and model, radiation exposure values were always within ± 2.1 mR (0.5 x 10–6 C/kg) of the mean radiation exposure. There was a wide range of radiation exposures between the different CT platforms at their default settings, with the child default radiation exposure ranging from 3.2 to 47.6 mR (0.8 to 12.3 x 10–6 C/kg) and the adult default radiation exposure ranging from 3.9 to 74.7 mR (1.0 to 19.3 x 10–6 C/kg). It is important for all institutions to check the default settings for survey images on each CT platform to make certain they are using as low a technique as possible for their applications.
Figure 6 compares the various CT platforms for the same kVp and tube position settings. The mA displayed is the minimum mA allowed at that kVp for that particular CT platform. Even though there is a range of radiation exposures between the different CT platforms, those exposures can be reduced substantially by dropping the kVp and/or rotating the tube to 180°.
The lowest exposures were achieved using 80 kVp, minimum X-ray tube current, and a 180° tube position. If these settings can be used for survey CT scans, the associated radiation exposure could be reduced to that of less than one chest radiograph. We recommend that the 180° tube position be implemented in place of the 0° tube position for all survey CT scans. This change should not affect image quality, since the radiation passes through both the table and the patient for 0° and 180° tube positions. However, the patient's radiation exposure would be reduced if the radiation beam strikes the table first, allowing it to absorb the lowest-energy X-rays. By the same logic, the radiation exposure of the breast in particular will be decreased.
When the X-ray tube voltage or current is altered the image quality will be affected. We suggest that sites determine the minimum acceptable image quality for their survey CT images. They then can determine the appropriate combination of X-ray tube voltage and current needed to achieve that level of image quality while minimizing the radiation exposure of the patient. If the survey CT scan is routinely interpreted as a projection radiograph as part of the overall CT examination, it will be very important to maintain acceptable image quality. In addition, if the survey CT image is used to provide patient geometric information for a variable mA program, the quality of the survey CT scan may need to be maintained in order for the tomographic image quality to remain reliable. However, if the survey CT scan is used only to determine the limits of the tomographic examination by locating bony or other high-contrast landmarks, high-quality survey CT images may not be needed. In these cases we suggest that sites consider decreasing their X-ray tube voltage and current so that the relative radiation exposure from the survey CT scan is as low as possible while maintaining adequate visualization of anatomic landmarks used when establishing scan limits. We recommend that CT clinics in which pediatric examinations are performed reduce their X-ray tube voltage and current to the minimum possible values (80 kVp and 10–50 mA for our scanners). In addition, if both P/A (180°) and lateral survey CT scans are not necessary for defining a tomographic acquisition, then only one survey CT scan should be acquired. Both A/P and lateral survey CT scan views are commonly included as a default setting in some scan protocols even though both views are not routinely used.
Size-specific determination of X-ray tube voltage and current also should be considered, particularly for pediatric cases. We implemented these changes (80 kVp, 10 mA, 180° tube position) in our pediatric CT protocols at this institution on our GE LightSpeed platform scanners approximately 1 year ago and have had no complaints from the radiologists or technologists concerning the quality of the survey CT images. Small adults also could benefit from these changes.
In summary, we have established a method for evaluating the radiation exposure from survey CT scans by measuring the exposure at isocenter and then correcting to an entrance exposure position for comparison to chest radiograph exposures. We measured the entrance exposures associated with survey CT scans using the default parameters for 21 CT scanners (11 CT models) at our institution and found that the exposure levels were equivalent to 0.2–4.7 chest radiographs. By changing the acquisition parameters of the survey CT scan, we reduced the exposure to less than one chest radiograph. Implementing these changes, particularly in children and young women, will help ensure that our patients receive "as low as reasonably achievable" doses from their CT examinations.
Acknowledgments
We thank John Zullo and Richard Wu for their assistance with the Philips scanners.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  N. Schneebaum, H. Blau, R. Soferman, H. Mussaffi, L. Ben-Sira, M. Schwarz, and Y. Sivan Use and Yield of Chest Computed Tomography in the Diagnostic Evaluation of Pediatric Lung Disease Pediatrics, August 1, 2009; 124(2): 472 - 479. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.B. Nauer, F. Kellner-Weldon, G. Von Allmen, D. Schaller, and J. Gralla Effective Doses from Scan Projection Radiographs of the Head: Impact of Different Scanning Practices and Comparison with Conventional Radiography AJNR Am. J. Neuroradiol., January 1, 2009; 30(1): 155 - 159. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
