| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Experimental Studies |
1 From the Department of Radiology, Duke University Medical Center, Box 3808, Durham, NC 27710 (L.M.H., R.E.R., T.T.Y., P.C.G., C.L.); and Radiation Safety Division, Duke University Health System, Durham, NC (R.E.R., T.T.Y., G.T., G.N.). Received December 1, 2006; revision requested January 26, 2007; revision received February 8; accepted March 22; final version accepted May 24. Supported in part by GE Healthcare, Milwaukee, Wis. Address correspondence to L.M.H. (e-mail: hurwi001{at}mc.duke.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATIONS FOR PATIENT CAREReferences |
|---|
Materials and Methods: Cardiac and pulmonary 64-detector CT angiography was performed by using current clinical protocols to evaluate the pulmonary veins (electrocardiographically [ECG] gated, 64 sections at 0.625-mm collimation, 120 kVp, 300 mA, 0.35-second tube rotation), native coronary arteries (ECG gated; 64 sections at 0.625 mm; 120 kVp; maximum current, 500–750 mA; minimum, 100–350 mA; 0.35-second tube rotation) and pulmonary embolus (64 sections at 1.25 mm, 140 kVp, 645 mA, 0.5-second tube rotation). Absorbed organ doses were measured by using an anthropomorphic female phantom and metal oxide semiconductor field effect transistor detectors. ED was calculated from measured organ doses and the dose-length product.
Results: ED for current adult cardiac and pulmonary 64-detector CT angiography protocols were 12.4–31.8 mSv. Overall, skin, breast, and esophagus and heart had the highest recorded absorbed organ doses. Relative risk for breast cancer incidence for girls and women was 1.004–1.042 for a single examination. Relative risk for lung cancer incidence for men and women was 1.005–1.076 from a single examination.
Conclusion: EDs and organ doses from 64-detector CT are higher than those previously reported for adult cardiac and pulmonary CT angiography protocols. Risk for breast and lung cancer induction from these studies is greatest for the younger patient population.
© RSNA, 2007
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATIONS FOR PATIENT CAREReferences |
|---|
Estimated ED from radiologic procedures is typically used to describe patient exposure, since actual radiation doses cannot be assessed. Superficially placed detectors on the skin surface only measure entrance dose, not absorbed organ dose. Estimated doses obtained with a variety of indirect methods are typically used. The most readily available technique uses values displayed on the CT console at the time of examination. This method employs the CT dose index volume, or CTDIvol, and the DLP as part of an equation that results in an estimate of patient exposure, the ED to a patient for any given examination. The variables used in the equation are problematic in this application, however, and their use has been shown to underestimate radiation exposure for 16-detector CT scanner protocols in adults (6–8). The same may be true for 64-detector CT scanners, which are now in use for evaluating the coronary and pulmonary vessels. An alternative technique involves the use of anthropomorphic phantoms with internally placed dosimeters to generate more accurate measurements of organ doses and EDs.
Radiation-induced cancer is a problem that has been addressed by the National Research Council (9), which has shown that the risk of cancer being caused by radiation is dependent upon the person's age at time of radiation exposure, total absorbed radiation to the body, total absorbed radiation to specific organs, and type of radiation. However, much of the information is derived from studies of individuals exposed to relatively large amounts of radiation, such as the Hiroshima, Japan, atomic bomb survivors. What is less clear is the relationship between levels of radiation exposure from diagnostic imaging and cancer risk. When evaluating the potential cancer-producing effect of cardiopulmonary imaging with 64-detector CT protocols, the organs of most concern are the lung and breast because they are in the direct path of the radiation beam, and the relationship between radiation exposure and cancer incidence in these organs is better understood than that for other organs (9).
Thus, the purpose of our study was to measure prospectively and directly in an anthropomorphic female phantom both organ dose and ED for adult cardiac and pulmonary CT angiography by using current clinical protocols for 64-detector CT and to estimate the lifetime attributable risk (LAR) of breast and lung cancer incidence on the basis of the ED and organ doses.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATIONS FOR PATIENT CAREReferences |
|---|
Phantom
A female adult anthropomorphic phantom (model 702-D; CIRS, Norfolk, Va) that has been validated for human organ dosimetry measurements was used (Fig 1a). The specific dimensions of the phantom are as follows: weight, 55 kg; height, 160 cm; thorax, 20 x 25 cm; abdomen, 18 x 22 cm; and pelvis, 19 x 34 cm. The phantom includes bone, lung, and soft-tissue compositions and is subdivided into 38 contiguous 2.5-cm-thick sections. Each section contains several 5-mm-diameter holes through which detectors are placed for organ dose measurements (Fig 1b). The phantom comes with assignable anatomic locations (Table 1) for major organs, and each hole location is optimized for precise dosimetry of a specific internal organ that can be referenced to the manufacturer's user manual (10). Tissue-equivalent plugs fill the holes when they are not being used. The breasts (base diameter 10.8 cm, height 4.3 cm) are attachable to the main body, as shown in Figure 1b. The breasts are made of a 50:50 glandular-to-adipose formula. Specific holes are located in the breasts 1 cm below the skin surface for detector placement.
|
|
|
|
Organ Dose Measurements
Dose measurements were performed by one medical physicist (T.T.Y.) and two members of the health physics staff (G.T., G.N.). Each of the three had 5 years of experience in radiation dosimetry determination with metal oxide semiconductor field effect transistor (MOSFET) detectors. Twenty MOSFET detectors were calibrated at respective clinical energies of 120 and 140 kVp, and two sets of 20 detector calibration factors for each scan energy were created and stored in a laptop computer by using AutoSence PCSoftware (model TN-RD-49; Thomson-Nielsen, Ottawa, Ontario, Canada). Calibration of detectors at respective scan energies is required to compensate for the energy dependence of the MOSFET detectors. For calibration, we added a 0.2-mm-thick copper filter to a conventional x-ray tube to match the half-value layer of a GE Healthcare scanner (11); the measured beam quality with 0.2-mm copper sheets was 7.27 mm aluminum half-value layer at 120 kVp. Individual MOSFET detectors were calibrated in air side by side with an ion chamber (model 10 x 5-6; Radcal, Monrovia, Calif) at respective scan energies (120 or 140 kVp). Absorbed dose in soft tissue was computed by using an f factor (roentgens-to-rads conversion factor) of 0.95 for 120 kVp and 0.95 for 140 kVp. The f factor for the breast was 0.88 for 120 kVp and 0.90 for 140 kVp. A chamber correction factor was applied by using a calibration value obtained from the University of Wisconsin Radiation Calibration Laboratory (Madison, Wis). No temperature and pressure corrections were necessary, because the monitor (model 9015; Radcal) automatically compensates for temperature and pressure.
ED Determination
ED was computed with two methods by a medical physicist (T.T.Y.) and members of the health physics staff (G.N., G.T.).
Method 1: phantom method.—The ED was calculated as the sum of the measured absorbed organ doses, including that of bone marrow (12), multiplied by individual tissue weighting factors published by the International Commission on Radiological Protection (13). Uncertainty for the ED was estimated by using quadrature summing.
Method 2: DLP method.—The ED was estimated as the product of the DLP and the normalized ED conversion factor (EDLP). The DLP was recorded directly from the console display at the time of the scan; this ensures that the DLP method to determine ED and the organ dose method are of the same scan z-axis coverage. The EDLP of 0.017 mSv·mGy–1·cm–1 was used for the chest (14) to determine the ED.
Determination of LAR, Relative Risk, and Excess Relative Risk for Radiation-induced Breast and Lung Cancer
Estimates for the LAR of the incidence of radiation-induced cancer of the breast and lung for girls and women and cancer of the lung for boys and men were calculated for a single examination for each of the four multidetector CT protocols by an experienced nuclear medicine physician (R.E.R.) with 11 years of radiation protection and risk communication experience. Values for LAR at 0.1 Gy for various solid tumors have been tabulated as a function of age at the time of exposure by the National Research Council (15). These values were used as the basis for risk estimation.
LAR was calculated for selected multidetector CT protocols performed for ages 15, 25, and 55 years. An exposure age of 15 years was chosen to be representative of older children in a population of pediatric patients who undergo selective coronary angiography (16). An exposure age of 25 years was chosen to be representative of patients being evaluated for peripartum pulmonary embolism. An exposure age of 55 years was chosen to be representative of patients undergoing multidetector CT coronary angiography, as reflected by the age distribution of patients undergoing CT for calcium scoring (17).
Values for LAR at 0.1 Gy for men and women at ages 25 and 55 years were calculated by using linear interpolation of the tabulated values at ages 20–30 and 50–60 years, respectively. Estimates of LAR for radiation doses to breast and lung for the various multidetector CT protocols were computed by multiplying the interpolated LAR values by the ratio of the protocol's absorbed organ dose to 0.1 Gy. Breast doses used in these calculations were the average of the doses for the right and left breasts.
Estimates of lifetime relative risk (RR) of radiation-induced cancer were computed on the basis of baseline (unexposed) values for lifetime risk (LR) per 100 000 population (18), as follows: RR = (LAR + LR)/LR.
Estimates of the joint probability (Pe-joint) of inducing either breast cancer or lung cancer in radiation-exposed female patients were calculated as follows: Pe-joint = Pe-lung + Pe-breast – (Pe-lung x Pe-breast), where Pe-lung is the LAR for lung divided by 100 000 and Pe-breast is the LAR for breast divided by 100 000. Joint LAR (LARjoint) and joint relative risk (RRjoint) were calculated as follows: LARjoint = Pjoint x 100 000 and RRjoint = LARjoint/LRjoint. Joint lifetime risk, LRjoint, was computed by using joint probabilities calculated as above for the unexposed population. Excess relative risk (ERR), the amount of additional relative risk above baseline, was calculated as a percentage over baseline, as follows: ERR = (RRjoint – 1.0) x 100.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATIONS FOR PATIENT CAREReferences |
|---|
|
|
|
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATIONS FOR PATIENT CAREReferences |
|---|
ED Measurements
Previously published EDs for cardiac and pulmonary CT angiography (2–5) are lower than those noted in our study of 64-detector CT clinical protocols. As expected, the ECG-gated cardiac CT angiographic protocol with the maximum tube current (coronary CT angiography protocol 2) had the overall highest ED, and the pulmonary vein protocol with the lowest overall milliampere setting and the shortest x-axis coverage had the lowest ED. This finding was similar for both methods used to determine ED. The difference between our results and previously published results may be related to differences in protocol parameters (eg, z-axis coverage or tube current setting) and a difference in determination method for ED.
Previous studies that reported ED for cardiac and pulmonary CT angiography used the DLPs that are displayed on the CT console at the time of data acquisition to estimate the ED. In other investigations, these values have been shown to underestimate ED by 20%–30% for adult 16-detector CT protocols (6–8). Our results demonstrate that for current adult clinical cardiac and pulmonary vascular 64-detector CT protocols, use of the DLP to estimate the ED will underestimate the overall radiation exposure. Thus, we believe that while CT dose index and DLP may be used as reference values to modify and develop CT angiography protocols, their use is not appropriate to assess the risk of radiation-induced cancer.
Lifetime Risks of Radiation from Cardiac and Pulmonary 64-Detector CT Angiography
Our estimates of lifetime excess relative risk for lung and breast cancer after a single multidetector CT examination of the chest are generally below 1.0% for individuals aged 55 years or older, who constitute the majority of patients undergoing evaluation of coronary artery disease. This risk is small for any one individual in this age group. For any individual patient, the use of this technology to diagnose coronary artery disease should be balanced against the risk of this test and the risks of other diagnostic tests (eg, stroke and vascular dissection due to coronary artery catheterization), and thoughtful use of other diagnostic tests should be considered. Our estimates of cancer risk from a single coronary CT angiographic examination illustrate that although the relative risk is small in older patients, inappropriate use of this technology may result in a large number of additional cancer cases if many people in this age group are screened for coronary artery disease with this imaging modality.
On the other hand, the lifetime excess relative risk for breast or lung cancer in girls and young women (aged 15 and 25 years, respectively) undergoing a single ECG-gated CT angiographic examination is higher, ranging from 1.7% to 5.5% for a single examination. The implications for cancer induction in this age group are more substantial. For example, young peripartum women are often evaluated with CT for symptoms of pulmonary embolism. This could potentially lead to an increased incidence of breast cancer in this population. As noted in a small study by Chen et al (22), women who were treated with external-beam radiation therapy for Hodgkin lymphoma while pregnant or within 1 month of pregnancy had an incidence of breast cancer of 16.0%, compared with an incidence of 2.3% in women of about the same age who were not pregnant or were more than 1 month postpartum when treated.
For all age groups, the risk of radiation-induced cancer would be increased in an additive manner if multiple follow-up multidetector CT studies were to be performed. This cumulative effect has been established by comparing the breast cancer risk in Japanese atomic bomb survivors with the risk in women who underwent multiple therapeutic exposures (23).
Our estimates of radiation risk, which are based on recent compilations by the Biological Effects of Ionizing Radiation (BEIR) committee, should be interpreted with caution for several reasons. First, The BEIR VII report (9) estimates of LAR for breast and lung cancer were derived primarily from Japanese populations, in whom the baseline risks for lung and breast cancer are different from those of the U.S. population. The BEIR committee has applied a "risk transport factor" to account for this, but the possibility remains that the radiation risks are different. Second, all such analyses are subject to uncertainties in the parameters of the models used to estimate risk. These include statistical sampling errors, uncertainties in radiation dosimetry, the possibility of incorrect diagnosis, and the presence of confounding variables such as smoking and socioeconomic factors. Third, the shape of the dose-response curve at doses below 100 mSv is not well characterized. Use of an incorrect dose-response curve would result in overestimation or underestimation of risk. In an attempt to reconcile differences in the extrapolation of human data obtained at medium and high doses into the low dose range (0–100 mSv) with the results of studies of cancer induction in animals, the BEIR committee has applied an approximate "dose and dose rate effectiveness factor" to its current risk estimates for all solid tumors. However, the possibility of the use of an inappropriate dose-response curve for site-specific cancers remains. Fourth, the effectiveness of the 120- and 140-kVp x-rays used in multidetector CT in inducing cancer may be different from that of the higher-energy gamma-rays (mean energy of about 3 MeV) that were the source of exposure for the atomic bomb survivors. The BEIR VII committee recognized this point and made no formal recommendation to account for it; however, they suggested that the actual risk due to x-rays in the diagnostic range may be two to three times higher than their results indicate. These and other limitations are discussed fully in the BEIR committee's report (9).
Our study had limitations. First, only one body type was used. Patients with more subcutaneous fat in whom the tube current settings are maximized may actually have lower organ doses due to the absorption of more of the photons in the subcutaneous tissue (24). We did not include measurements for patients undergoing bypass graft evaluation, in whom the z-axis length is increased, which thus increases the total radiation dose. Some CT angiography clinical protocols may use a higher minimum tube current than that used in our study (ie, greater than the 20%–46% of the maximum current that we used), and this too will increase potential exposure. Radiation doses for scout, unenhanced, and contrast material timing bolus images and for a dedicated calcium scoring protocol (which may be performed in conjunction with the ECG-gated coronary CT angiographic study) were not included in the organ dose and ED measurements for our study. Again, the actual doses for any patient undergoing a clinical study to assess native coronary artery stenosis, PE, or pulmonary vein stenosis will be somewhat higher than what our calculations suggest, but the increase in dose from these sequences (which use substantially lower tube current settings) would represent a very small percentage of the overall exposure.
Because this was a phantom study, our results provide an estimate of organ doses and EDs for current cardiac and pulmonary 64-detector CT angiographic clinical examinations. The actual doses for any individual will vary from patient to patient, depending on tube current setting, heart rate, z-axis coverage, and patient body habitus.
Practical applications: Our data provide the most current estimation of total ED and specific organ doses for current clinical cardiac and pulmonary 64-detector CT angiography protocols. The data can be used by the practicing radiologist and referring physician to assess the potential risk for cancer induction. Calculation of ED by using the DLP will underestimate radiation exposure for adult cardiac and pulmonary 64-detector CT angiography protocols and does not address the issue of specific organ dose. From our data, we conclude that widespread indiscriminate use or unproved application of cardiac or pulmonary CT angiography might lead to a large number of induced breast and lung cancers. Use of cardiac and pulmonary CT angiography protocols, particularly in young women and children, should be carefully considered in conjunction with clinical indications, benefits and risks, and alternative imaging modalities.
|
ADVANCE IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATIONS FOR PATIENT CAREReferences |
|---|
|
IMPLICATIONS FOR PATIENT CARE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATIONS FOR PATIENT CAREReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: DLP = dose-length product • ECG = electrocardiography • ED = effective dose • LAR = lifetime attributable risk • PE = pulmonary embolus
See Materials and Methods for pertinent disclosures
Author contributions: Guarantor of integrity of entire study, L.M.H.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, L.M.H., R.E.R., T.T.Y., P.C.G.; experimental studies, L.M.H., T.T.Y., G.T., G.N., C.L.; and manuscript editing, L.M.H., R.E.R., T.T.Y., P.C.G.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCE IN KNOWLEDGEIMPLICATIONS FOR PATIENT CAREReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. McDermott, S. C. Barry, E. E. Judge, S. Collins, P. A. de Jong, H. A. W. M. Tiddens, E. F. McKone, C. C. Gallagher, and J. D. Dodd Tracheomalacia in Adults with Cystic Fibrosis: Determination of Prevalence and Severity with Dynamic Cine CT Radiology, June 9, 2009; (2009) 2522081956. [Abstract] [Full Text]
|
|
|
|
 |
|
 W. P. Shuman, K. R. Branch, J. M. May, L. M. Mitsumori, J. N. Strote, B. H. Warren, T. J. Dubinsky, D. W. Lockhart, and J. H. Caldwell Whole-Chest 64-MDCT of Emergency Department Patients with Nonspecific Chest Pain: Radiation Dose and Coronary Artery Image Quality with Prospective ECG Triggering Versus Retrospective ECG Gating Am. J. Roentgenol., June 1, 2009; 192(6): 1662 - 1667. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. K. Pahade, D. Litmanovich, I. Pedrosa, J. Romero, A. A. Bankier, and P. M. Boiselle Quality Initiatives: Imaging Pregnant Patients with Suspected Pulmonary Embolism: What the Radiologist Needs to Know RadioGraphics, May 1, 2009; 29(3): 639 - 654. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Mayo and J. A. Leipsic Radiation Dose in Cardiac CT Am. J. Roentgenol., March 1, 2009; 192(3): 646 - 653. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Goodman, H. D. Sostman, P. D. Stein, and P. K. Woodard CT Venography: A Necessary Adjunct to CT Pulmonary Angiography or a Waste of Time, Money, and Radiation? Radiology, February 1, 2009; 250(2): 327 - 330. [Full Text] [PDF]
|
|
|
|
|
|
P. C. Shrimpton, B. F. Wall, T. T. Yoshizumi, L. M. Hurwitz, and P. C. Goodman Effective Dose and Dose-Length Product in CT Radiology, February 1, 2009; 250(2): 604 - 605. [Full Text] [PDF]
|
|
|
|
|
|
L. M. Hurwitz, T. T. Yoshizumi, P. C. Goodman, R. C. Nelson, G. Toncheva, G. B. Nguyen, C. Lowry, and C. Anderson-Evans Radiation Dose Savings for Adult Pulmonary Embolus 64-MDCT Using Bismuth Breast Shields, Lower Peak Kilovoltage, and Automatic Tube Current Modulation Am. J. Roentgenol., January 1, 2009; 192(1): 244 - 253. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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. Huda, T. T. Yoshizumi, L. M. Hurwitz, and P. C. Goodman Computing Effective Doses from Dose-Length Product in CT Radiology, July 1, 2008; 248(1): 321 - 322. [Full Text] [PDF]
|
|
|
|
 |
|
 M. D. Siegel Evaluation of Suspected Pulmonary Embolism During Pregnancy JAMA, April 9, 2008; 299(14): 1665 - 1665. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
