| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Special Report |
1 From the Division of Abdominal Imaging and Intervention, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, White 270-E, 55 Fruit St, Boston, MA 02114 (M.K.K., M.M.M., R.S.K., E.F.H., S.S.); and GE Medical Systems, Waukesha, Wis (T.L.T.). Received July 25, 2003; revision requested October 2; final revision received December 18; accepted January 29, 2004. Address correspondence to M.K.K. (e-mail: mannudeep_k_kalra@yahoo.com).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Superior and inferior levels at routine abdominal and/or pelvic CT were defined as the dome of the diaphragm and the inferior margin of the pubic symphysis, respectively. Records of 106 consecutive examinations (male-to-female ratio, 45:61; age range, 21–86 years) performed from June 1 to June 30, 2003, were reviewed to determine the number of "extra" images. Sixty-two abdominal and/or pelvic CT examinations performed concurrently with chest or thigh CT or for trauma were not included in the 106. Abdominal and/or pelvic CT was performed with either ATCM (n = 44) or manual selection of tube current (n = 62). CT parameters recorded for each extra image included tube current, peak kilovoltage, and gantry rotation time. Mean and median tube current–time products were calculated for extra images. Extra images were analyzed for pathologic findings. Statistical analysis was performed with the Student t test.
RESULTS: Extra images were acquired above the dome of the diaphragm in 103 (97%) of 106 examinations and below the pubic symphysis in 100 (94%) of 106. A total of 1,280 extra images were acquired in 106 examinations (mean, 12 images per examination). Nineteen additional findings were observed on extra images. With ATCM, mean tube current–time product was 74.5 and 120.6 mAs for extra images acquired above the diaphragm and below the pubic symphysis, respectively; with manual selection, mean tube current–time products were 167.5 and 168.3 mAs (P < .05).
CONCLUSION: Most extra images acquired at abdominal and/or pelvic CT contributed no additional information. With ATCM, the radiation dose was reduced by a mean of 56% (median, 72%) for extra images above the diaphragm and 29% (median, 36%) for images below the pubic symphysis, compared with dose levels with manual selection.
© RSNA, 2004
Index terms: Abdomen, CT, 78.1211 • Computed tomography (CT), radiation exposure • Computed tomography (CT), technology • Computed tomography (CT), thin-section, 78.12118, 88.12118 • Pelvis, CT, 88.1211 • Special Reports
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
It seems a reasonable assumption that eliminating "extra" images—that is, images obtained beyond the desired area of interest—could help to reduce radiation exposure. Thus, the purpose of our study was to retrospectively determine the number and diagnostic yield of extra images acquired beyond the intended anatomic area of interest at abdominal and/or pelvic CT scanning and to assess the effect of ATCM on radiation dose associated with acquisition of these images.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
CT Technique
All examinations were performed with either a four-section multi–detector row CT scanner (LightSpeed QX/i; GE Medical Systems, Waukesha, Wis) or a 16-section multi–detector row scanner (LightSpeed 16; GE Medical Systems) by using either manual selection of fixed tube current (n = 62) or the ATCM technique (n = 44). All examinations that included the use of ATCM (Auto mA; GE Medical Systems) were performed with identical scanning parameters (minimum tube current, 75 mA; maximum tube current, 380 mA; noise index, 15.0 HU) as per our standard department protocol. Remaining scanning parameters included peak kilovoltage of 140, 0.5–1.0-second gantry rotation time, 5-mm reconstructed section thickness, and beam pitch of 1.5:1 with the four-section scanner (n = 28) and 0.938:1 with the 16-section scanner (n = 78). The z-axis modulation technique of ATCM allows the technologist to select a noise index, which determines the image quality (noise). The CT scanner then modulates tube current at each level along the z-axis to produce an image with the preselected noise index (1). The noise index value is approximately equal to the image noise (± standard deviation) in the central region of the image obtained from scanning a uniform phantom.
Number of Extra Images
The superior and inferior margins of the scan volume at abdominal CT were defined on the basis of standard department protocol. All images above the level of the dome of the diaphragm were categorized as supradiaphragmatic extra images, and those below the inferior margin of the pubic symphysis were deemed infrapubic extra images. A score of –1 was assigned to examinations that began at a level below the dome of the diaphragm or ended above the inferior margin of the pubic symphysis. A score of 0 was assigned to examinations that began at the dome of the diaphragm or ended at the inferior margin of the pubic symphysis. The score for examinations that began above the dome of the diaphragm and ended below the inferior margin of the pubic symphysis was equivalent to the number of supradiaphragmatic and infrapubic extra images acquired. The numbers of supradiaphragmatic and infrapubic extra images were recorded separately for each examination (M.K.K., M.M.M.). These images were assessed (M.M.M., with 9 years of experience) for signs of disease in the lung parenchyma, mediastinum, skeleton, and adjoining soft tissue (chest wall and upper thighs). All additional findings seen on extra images were recorded. Indications for each CT examination were recorded, and the distribution of extra images in patients with cancer and noncancer indications was assessed.
Tube Current–Time Product Associated with Extra Images
Tube current (in milliamperes) and gantry rotation time (in seconds) were noted for each extra image acquired with manual selection of tube current, and the range of tube current–time products was determined (R.S.K.). Tube current–time product (in milliampere-seconds) was determined by multiplying tube current by gantry rotation time. For each examination performed with ATCM, mean tube current was calculated separately for supradiaphragmatic and infrapubic extra images by dividing the sum of individual tube current values for extra images by the number of extra images. Thus, tube current–time product for ATCM techniques was calculated by multiplying the mean tube current by the gantry rotation time. The CT dose index (CTDI) and dose-length product (DLP) were calculated (T.L.T., M.K.K.) as described in the dosimetry section of the manual for the four-section scanner (9). These calculations were identical to the estimates of CTDI volume (CTDIVOL) and DLP displayed on the user interface of the scanners. The standard dose values for the 32-cm-diameter body phantom (CTDI100) were adjusted by using the published technique adjustment factors that were employed for each examination (9). The CTDIVOL dose was calculated by adding one-third of the central CTDI100 value to two-thirds of the peripheral CTDI100 value and dividing the sum by the pitch value. The DLP was calculated by multiplying the CTDIVOL dose (in grays) by the length of the scan volume (in centimeters).
Statistical Analysis
To assess whether there was an association between the numbers of extra images acquired in each patient and the patient’s age, a Spearman correlation test was performed by one of the authors (M.K.K.) with statistical software (VassarStats; Vassar College, Poughkeepsie, NY), which is available online at faculty.vassar.edu/lowry/ank4.html. The numbers of extra images acquired in patients with and without malignancy were compared by using spreadsheet software (Excel; Microsoft, Redmond, Wash), and data were analyzed with the Student t test. Mean and median tube current–time products were determined separately for ATCM and for manual selection by using the spreadsheet software. Average DLP was calculated for examinations, as well as for extra images. Differences between mean tube current–time product and DLP with manual selection of tube current versus ATCM for supradiaphragmatic and infrapubic extra images were determined by another author (E.F.H.) with the Hotelling T statistic for multivariate analysis of variance, by using software (SAS/STAT; SAS, Cary, NC). A P value of less than .05 was considered to indicate a significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Additional Images
Extra images from 19 (18%) of 106 examinations revealed additional findings, including bilateral or unilateral pleural effusions or thickening (12 examinations), emphysematous bullae (n = 2), esophageal varices (n = 1), two subcentimeter pulmonary nodules in the right lower lobe (n = 1) (Fig 1), and unilateral or bilateral inguinal hernia containing intraperitoneal fat (n = 3). Extra images from the other 87 examinations revealed no additional abnormalities (Fig 2). In three examinations, additional findings were seen only on supradiaphragmatic extra images. These findings included two pulmonary nodules in the right lung (Fig 1), an emphysematous bulla in the right lower lobe, and superior extent of esophageal varices. Thirteen other supradiaphragmatic findings were seen also on images obtained in the area of interest (Fig 3a). All abnormalities found on infrapubic extra images were visible also on images confined to the defined area of interest above the pubic symphysis (Fig 3b). There was no significant correlation between patient age and number of extra images acquired during routine CT examinations (correlation coefficient, 0.05; P > .5).
|
|
|
|
|
Tube Current–Time Product Associated with Extra Images
With manual selection of fixed tube current, mean and median tube currents for extra images were 221.5 and 220.0 mA, respectively. The tube current–time product range for examinations performed with manual selection of fixed tube current (200–290 mA) was 105–290 mAs. The mean and median tube current–time products for supradiaphragmatic extra images were 167.5 and 163.7 mAs, respectively. For infrapubic extra images, the mean and median tube current–time products were 168.3 and 176.0 mAs, respectively. The tube current–time product ranges for supradiaphragmatic and infrapubic extra images in examinations performed with z-axis modulation by using ATCM were 37.5–296.6 mAs (based on tube current of 75–379 mA) and 47.5–294.3 mAs (based on tube current of 93–380 mA), respectively. For supradiaphragmatic extra images acquired with ATCM, the mean and median tube current–time products were 74.5 and 45.9 mAs, respectively. For infrapubic extra images acquired with ATCM, the mean and median tube current–time products were 120.6 and 112.9 mAs, respectively. There was a significant reduction of mean tube current–time product for extra images with the ATCM technique compared with conventional manual selection of tube current at both the supradiaphragmatic and infrapubic levels (P < .001).
Mean CTDIVOL was 22.9 and 11.4 mGy for examinations performed with fixed tube current and ATCM, respectively. Mean DLP was 1,038.0 and 519.0 mGy · cm for examinations performed with fixed tube current and ATCM, respectively. Mean DLP associated with extra images acquired in examinations performed with fixed tube current was 136.4 mGy · cm (mean for supradiaphragmatic images, 63.1 mGy · cm; mean for infrapubic images, 73.3 mGy · cm). Mean DLP associated with extra images acquired in examinations performed with ATCM was 86.6 mGy · cm (mean for supradiaphragmatic images, 29.6 mGy · cm; mean for infrapubic images, 57.0 mGy · cm). Thus, compared with fixed tube current selection, the ATCM technique resulted in a 36% reduction in the radiation dose associated with the acquisition of extra images (53% reduction for supradiaphragmatic images, 22% reduction for infrapubic images) (P < .001).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Based on the large variations in beam attenuation in different body regions and the fact that image noise is determined by quantum noise in the incident beam projections, ATCM techniques allow automatic adjustment of tube current to maintain constant image quality at an optimal radiation dose. In ATCM with the z-axis modulation technique, the tube current is automatically adjusted according to regional body anatomy to maintain a constant user-specified quantum image noise level and to improve radiation dose efficiency. A noise index is provided to allow users to select the amount of noise that will be acceptable on the reconstructed images. The noise index value is approximately equal to the value of image noise (standard deviation) in the central region of an image obtained by scanning a uniform phantom. With a single radiographic scout image, the system calculates the tube current that will be needed to obtain images with the selected noise index. The scout projection data include information about patient density, size, and shape. Thus, the system is designed to ensure that all images have a constant noise level regardless of differences in patient size and anatomy. ATCM determines the tube current for each anatomic region on the basis of the scout projection data acquired in the patient, as well as a set of empirically determined noise prediction coefficients for the reference technique. The reference technique consists of scanning of the uniform phantom to obtain an arbitrary 2.5-mm-thick section at the selected peak kilovoltage and tube current–time product of 100 mAs, followed by transverse reconstruction with a standard reconstruction algorithm. On the basis of our initial experience with ATCM, we have selected a noise index of 15 HU for routine abdominal CT examinations at our institution.
Although previous scientific publications have recommended the restriction of scanning to the region of interest, to the best of our knowledge, no formal investigation has been conducted to assess the extent of the CT examination beyond the area of interest (1,2). The results of our study reveal that a substantial number of extra images are acquired beyond the borders of the area of interest at abdominal and/or pelvic CT. An average routine abdominal and/or pelvic CT examination with the aforementioned scanning protocol typically results in 80–100 transverse images in a single-phase study. The 1,280 extra images acquired in our 106 patients were therefore equivalent to the total acquisitions from 12 to 16 additional examinations, and if these extra images had not been acquired, the mean cumulative radiation dose would have been about 10% lower. Extra acquisitions accounted for a 16.7% increase in mean DLP with ATCM (86.6 vs 519.0 mGy · cm) and a 13.1% increase with fixed tube current (136.4 vs 1,038.0 mGy · cm). Of even greater concern with these extra acquisitions is the increased radiation exposure specifically to body parts that are more vulnerable to radiation-associated risks (eg, breasts in women and gonads in men).
Our study results show that a substantial number of extra images are acquired at routine abdominal CT both for benign and for malignant indications, irrespective of patient age and sex. It is noteworthy that additional findings, not seen on images acquired in the defined region of interest, were observed on extra images in only three (3%) of 106 examinations and that only one finding (two pulmonary nodules in the right lung of a patient with colon cancer) could have had a substantial effect on management. Thus, findings on extra images affected diagnosis in only one (1%) of 106 cases.
Among the factors that may contribute to the acquisition of extra images are possible inaccuracies in scanners in the localizer radiograph annotations for level; specific instructions by the referring physician to include supradiaphragmatic or infrapubic regions; technologists’ intention to include or cover lesions seen on scout images; errors related to technologists’ training; and uncooperative, breathless, or unconscious patients in whom technologists do not want to miss the margins. Per our departmental policy, however, technologists are instructed to include all requests from referring physicians in the patient’s record. At the supradiaphragmatic level, the acquisition of extra images may be justified to ensure that the entire liver and spleen are included in one phase of contrast enhancement. However, in a small number of cases in our study, as many as 36 extra images were acquired, a fact that may suggest a lack of attention in the selection of scan volume. Infrapubic extension of routine abdominal and/or pelvic CT examinations without appropriate clinical request or reason, especially given the risks of radiation to the gonads, cannot be as easily justified. Likewise, although the acquisition of extra images might be acceptable in uncooperative or breathless patients, we found that it adds no diagnostic information to that provided by images in the area of interest for routine abdominal and/or pelvic CT and should be restricted when not indicated or requested for specific clinical reasons. Indeed, it is important for radiologists and technologists to understand that the extension of image acquisition beyond the region of interest is associated with an additional radiation dose and that they share the responsibility of ensuring that scanning is restricted to the region of interest.
The results of previous studies have shown substantial radiation dose reduction with online angular modulation (CARE Dose; Siemens Medical Solutions, Forchheim, Germany) (10–12). Perhaps the most substantial finding of our study is the effect of z-axis ATCM in reducing the radiation associated with these extra images. Overall reductions in mean and median tube current–time product for extra images acquired with ATCM technique compared with those acquired with manual selection of tube current were 58% and 47%, respectively. For supradiaphragmatic extra images, the use of ATCM was associated with reductions of 56% and 72% in mean and median tube current–time product, respectively, compared with the values associated with manual selection of tube current. For infrapubic images, ATCM resulted in reductions of 29% and 36% in mean and median tube current–time product, respectively, compared with the values associated with manual selection of tube current. The range of tube current–time products for supradiaphragmatic and infrapubic extra images was larger with ATCM than with manual selection of fixed tube current. The use of ATCM at acquisitions in the supradiaphragmatic region offers greater opportunity for dose reduction than in the infrapubic region, because there is a greater change in attenuation caused by the lungs. In concordance with findings in previous studies that demonstrated the efficiency of ATCM with the angular modulation technique, our findings prove that z-axis modulation is effective for reducing radiation dose by reducing tube current–time product. Various manufacturers supply scanners with features for automatic tube current modulation that are substantially comparable with those described here.
Supradiaphragmatic extra images in 36 (82%) of 44 examinations performed with ATCM were obtained with tube current–time products of less than 100 mAs, whereas in three (7%) of 44 examinations performed with ATCM, mean tube current–time products were greater than those associated with manual selection. On the other hand, infrapubic extra images in 18 (41%) of 44 examinations performed with ATCM were obtained with a tube current–time product of less than 100 mAs, whereas in nine (20%) of 44 examinations with ATCM, mean tube current–time products were greater than those associated with manual selection. Thus, ATCM seems to be more effective in reducing radiation dose in supradiaphragmatic regions than in infrapubic regions, and its use may result in a greater radiation dose to the pelvis or infrapubic region, as noted in our study.
Our study had several limitations. The study cohort comprised examinations performed with different scanners (four- and 16-section multi–detector row CT scanners). The number of examinations performed with 16-section multi–detector row CT scanners with manual selection of tube current (n = 20) seemed sufficient for comparison with 16-section multi–detector row CT examinations performed with ATCM (n = 44). It is important to note that the differences between tube current–time products with use of ATCM on 16-section scanners versus manual selection on four- and 16-section scanners were significant (P < .05). We did not, however, perform a power analysis to determine the number of examinations necessary for the study, because of the absence of any prior study assessing our hypothesis. One other limitation is that the rate of extra image acquisition for different technologists was not analyzed.
In conclusion, the acquisition of a substantial number of extra images at routine abdominal and/or pelvic CT is associated with additional radiation exposure to the patient. ATCM along the z-axis aids in reducing the radiation dose associated with extra images acquired with abdominal CT.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Author contributions: Guarantors of integrity of entire study, S.S., M.K.K.; study concepts and design, M.K.K.; literature research, M.K.K.; clinical studies, M.K.K., M.M.M.; data acquisition, M.K.K., M.M.M., R.S.K.; data analysis/interpretation, M.K.K., E.F.H.; statistical analysis, E.F.H.; manuscript preparation, M.K.K., M.M.M.; manuscript definition of intellectual content, editing, and revision/review, M.K.K.; manuscript final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. Just da Costa e Silva and G. Alves Pontes da Silva Eliminating Unenhanced CT When Evaluating Abdominal Neoplasms in Children Am. J. Roentgenol., November 1, 2007; 189(5): 1211 - 1214. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. van der Molen and J. Geleijns Overranging in Multisection CT: Quantification and Relative Contribution to Dose--Comparison of Four 16-Section CT Scanners Radiology, December 1, 2006; 242(1): 208 - 216. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
