M e d i c a l P hy s i c s a n d I n f o r m a t i c s • O r i g i n a l R e s e a r c h Leng et al. Radiation Doses in Interventional CT

Downloaded from www.ajronline.org by 37.44.207.139 on 01/19/17 from IP address 37.44.207.139. Copyright ARRS. For personal use only; all rights reserved

Medical Physics and Informatics Original Research

Radiation Dose Levels for Interventional CT Procedures Shuai Leng1 Jodie A. Christner Stephanie K. Carlson Megan Jacobsen Thomas J. Vrieze Thomas D. Atwell Cynthia H. McCollough Leng S, Christner JA, Carlson SK, et al.

Keywords: CT, interventional CT, radiation dose DOI:10.2214/AJR.10.5057 Received June 2, 2010; accepted after revision December 23, 2010. Presented at the 2009 Radiological Society of North America annual meeting, Chicago, IL. 1

All authors: Department of Radiology, Mayo Clinic, 200 First St SW, Rochester, MN 55905. Address correspondence to S. Leng ([email protected]).

WEB This is a Web exclusive article. AJR 2011; 197:W97–W103 0361–803X/11/1971–W97 © American Roentgen Ray Society

OBJECTIVE. The purpose of this study was to determine typical radiation dose levels to patients undergoing CT-guided interventional procedures. MATERIALS AND METHODS. A total of 571 patients undergoing CT interventional procedures were included in this retrospective data analysis study. Enrolled patients underwent one of five procedures: cryoablation, aspiration, biopsy, drain, or injection. With each procedure, two scan modes were used, either intermittent (no table increment) or helical mode. Skin dose was estimated from the volumetric CT dose index (CTDIvol) and phantom measurements. Effective dose was calculated by multiplying dose-length product (DLP) and conversion factor (k factor) for helical mode, and using Monte Carlo organ dose coefficients for intermittent mode. RESULTS. The mean (± SD) skin doses were 728 ± 382, 130 ± 104, 128 ± 81, 152 ± 105, and 195 ± 147 mGy, and the mean effective doses were 119.7 ± 50.3, 20.1 ± 11.0, 13.8 ± 9.2, 25.3 ± 15.4, and 9.1 ± 5.5 mSv for each of the five procedures, respectively. The maximum skin dose was 1.95 Gy. The mean effective dose across all procedure types was 24.1 mSv, with 2.3 mSv from intermittent scans and 21.8 mSv from helical scans. CONCLUSION. Substantial dose differences were observed among the five procedures. The risk of deterministic effects appears to be very low, because the maximum observed skin dose did not exceed the threshold for transient skin erythema (2 Gy). The average risk of stochastic effects was comparable to that of 1–10 abdomen and pelvis CT examinations. Although the intermittent mode can contribute substantially to skin dose, it contributes minimally to the effective dose because of the much shorter scan range used.

C

T-guided interventional procedures have been widely used for a variety of diagnostic and therapeutic purposes [1–8]. Because these procedures are less invasive and more cost effective than open surgery, the number of interventional procedures has increased [2, 9, 10]. During CT-guided interventional procedures, 2D or 3D images are acquired to generate a road map of the targets and their positions relative to the interventional instruments (e.g., needles and catheters). This valuable information provides guidance for the operator to locate the target, plan an interventional path, adjust the interventional instruments, and evaluate the efficacy of the procedure. Interventional CT is performed using a variety of scan modes, where the choice of mode depends on the imaging purpose, scanner configuration, and operator preference. CT fluoroscopy, as its name implies, uses real-time reconstructions of continuously ac-

quired data to provide immediate feedback to the operator [6]. The patient table is either stationary or moved manually by the operator. Alternatively, the operator can acquire discrete axial images, one image per step on the pedal, between table or instrumentation movements [11]. This intermittent scan mode has the advantage of lowering patient and operator dose and is used exclusively in our practice [2]. During both continuous (fluoroscopic) and intermittent scans, automatic table motion is disabled. Thus, the same section of anatomy is scanned repeatedly unless the operator manually adjusts the table position, limiting the scan range to the width of the detector along the z-axis. For some procedures, such as cryoablation, where it is essential to monitor anatomy over a z-axis range greater than the detector width of most CT systems, the patient must be moved through the scan plane. For these procedures, helical scans are repeated over

AJR:197, July 2011 W97

Downloaded from www.ajronline.org by 37.44.207.139 on 01/19/17 from IP address 37.44.207.139. Copyright ARRS. For personal use only; all rights reserved

Leng et al. the region of interest at each time point (e.g., initial localization scan, during the procedure, and immediately after the procedure). The use of a stationary table mode (continuous or intermittent acquisitions) limits the scan range relative to a helical scan and thus delivers lower radiation doses to the patient. Radiation effects associated with ionizing radiation can be classified as either deterministic or stochastic. For deterministic effects, a threshold exists, above which biologic effects have been seen in some fraction of the exposed population. For stochastic risks, the linear-nonthreshold model of risk is currently the most widely used, although this is still a controversial topic. According to this model, stochastic effects could occur at any dose level—that is, no threshold exists below which the risk is zero. Unlike deterministic risks, the severity of the effect is independent of dose (i.e., one either gets cancer or not, but the severity of the cancer is not affected by the dose level). Instead, the probability of a radiogenic cancer is dependent on the dose level. To assess the potential risk of deterministic effects, the mean skin (or other tissue) dose is compared with the relevant dose threshold. For stochastic risks, effective dose is used to combine the absolute dose to any given tissue with radiation sensitivity information to estimate the lifetime risk of cancer incidence or mortality [12, 13]. There have been several studies dealing with radiation doses associated with certain interventional procedures. Doses to both the operators and the patients were considered, using either phantom measurements or clinical patient data [2, 7, 11, 14–17]. However, a large variation exists among these studies regarding the dose measurement method and the procedures studied. During the last decade, CT technologies have been greatly improved, and the use of interventional CT has increased. Therefore, the purpose of this study was to determine the mean effective and skin doses for five common interventional CT procedures at a large academic medical center with state-of-the-art CT equipment. Using information from the patient record, image data, and phantom measurements, skin dose and effective dose were estimated to quantify potential deterministic and stochastic risks. Materials and Methods This retrospective data analysis was HIPAA compliant and was approved by our institutional review board with a waiver of informed consent.

W98

Consecutive patients undergoing CT-guided interventional procedures between March and July of 2008 were identified and included in this study. “CT Interventional Procedure Data Sheets” were available at each of eight CT scanners in our practice where interventional procedures were performed. Patient information (e.g., age and sex) and procedure information (e.g., procedure type, scan mode, tumor size, and location) were recorded by the radiologist and technologists responsible for the case. These data sheets were routinely collected from the scan room, and the data were recorded and analyzed by diagnostic medical physicists and research fellows using a Microsoft Excel spreadsheet. A total of eight CT scanners used at our institution for interventional procedures were included in this study: one GE Healthcare LightSpeed16, one GE Healthcare VCT, one Siemens Healthcare Sensation 16, three Siemens Healthcare Sensation 64, one Siemens Healthcare Sensation 40, and one Siemens Healthcare Definition. Our practice routinely evaluates image quality (e.g., noise, uniformity, and CT number accuracy) as a function of slice width and scanner outputs (e.g., peak kilovoltage and tube current–exposure time product). We then adjust our clinical scan protocols accordingly, such that similar levels of image quality are achieved at very similar dose levels. Thus, systematic differences between the scanners used in this study are assumed to be negligible for the clinical protocols used.

Interventional Procedures The interventional procedures performed in our practice were divided into five distinct groups according to the clinical purpose. First, for cryoablation, probes containing argon gas are inserted into a tumor or metastatic region. Controlled release of the gas cools adjacent tissue quite rapidly. The thermal conduction pathways can transfer the cold toward the critical anatomy or may fail to freeze the entire target. Thus, several freeze-thaw cycles are performed to achieve complete ablation (killing) of the tumor. Second, for aspiration, a needle is positioned within the target and fluid is drawn for histologic analysis. Third, for biopsy, a device is positioned within the target that can mechanically remove (i.e., cut out) a small section of tissue for histologic analysis. Fourth, for drainage, a tube is inserted into an internal fluid volume to remove excess fluid, such as occurs with postsurgical edema or infection. Finally, for injection, a needle is positioned and a therapeutic agent injected into the target, such as the injection of stabilizing cement into vertebral fractures. Skin dose and effective dose were estimated for each procedure, and the mean and SD were calculated for each procedure type. Because of the in-

trinsic differences between the helical scan mode (table translation) and intermittent mode (no table translation), the contributions of the modes to patient dose (skin dose and effective dose) are quite different. Therefore, dose data were also analyzed separately for each scan mode within each of the five procedures types. The dose fraction that each scan mode (helical or intermittent) contributed to the total dose for each procedure type was also calculated and compared.

Skin Dose Estimation Skin dose and effective dose for each patient were estimated from recorded volumetric CT dose index (CTDIvol) and dose-length product (DLP) [18, 19], which were obtained directly from the scanner console and recorded on the data sheet, as well as previously acquired phantom measurements that calibrated skin dose (using an anthropomorphic adult phantom) to scanner output (i.e., CTDIvol) for each scanner model included in this study. CTDIvol has been widely used to objectively quantify the radiation output of a CT system, and is calculated as follows: CTDIvol = CTDIw / pitch (1),



in which CTDIw is the weighted CTDI [20], calculated as follows:

CTDIw =

1 2 CTDIcenter + CTDIperipheral (2), 3 3

and CTDIcenter and CTDIperipheral are the CTDI100 values at the center and periphery, respectively, of the standard 16-cm (head) or 32-cm (body) acrylic CTDI phantoms [18, 19]. Bauhs et al. [21] found that skin dose is approximately 100% of the peripheral CTDI100 in helical scan mode and about 50% of the peripheral CTDI100 in intermittent mode using an adult anthropomorphic phantom, calculated as follows:

skin dose =

helical mode CTDIperipheral (3). 0.5 × CTDIperipheral intermittent mode

The lower percentage in the intermittent mode is the result of the fundamental differences between CTDI100 and dose to a specific point when the patient is not moved through the scan plane [21]. CTDI100 is a dose index measuring dose integrated along the longitudinal direction with a 10-cm-long ionization chamber for a single axial scan. It is normalized to the total nominal image thickness. CTDI100 can estimate the multiple scan average dose for scans with table translation, either step-and-shoot or helical mode [20]. However, for axial scans without table motion, CTDI100 overestimates radiation dose, because it includes the tails of the dose profile along the longitudinal direction, which should not be included in the skin

AJR:197, July 2011

dose measurement for this scan mode because there is no table translation. In body CT scans, the CTDI100 measured at the center of the 32-cm-diameter cylindric phantom is about one half of the CTDI100 measured at the phantom periphery [19]. Combining this relationship, equation 1 and equation 2, together with the observations by Bauhs et al. [21] (i.e., equation 3), and other phantom measurements made in our practice, skin dose can be estimated for a given scan mode and scanner output (CTDIvol), as follows: skin dose =

1.2 × CTDIvol helical mode (4). 0.6 × CTDIvol intermittent mode

Similar results were reported in another study in which skin dose in intermittent mode was found to be 49–65% of the measured CTDIvol for phantom sizes ranging from 25 to 50 cm at a beam collimation of 12 × 0.6 mm (the collimation used in most of our intermittent mode scans) [22]. By use of these relationships and the CTDIvol reported for each scan, skin dose was estimated for each patient and procedure.

Effective Dose Estimation The effective dose for each procedure can be estimated using organ dose coefficients derived from Monte Carlo simulations [23, 24] and International Commission on Radiological Protection–specified tissue weighting factors [12, 25, 26]. Alternatively, a simplified method using DLP and a DLP-to– effective dose conversion factor (k factor), which depends only on the body region scanned, can be used [27–29]. For helical scans, k factors used to calculate effective dose from DLP were available [27, 28]. Here, we used a published k factor of 0.015 because most of the procedures were performed in the abdomen and pelvis region, and the difference in k factors among the chest, abdomen, and pelvis regions is relatively small (0.015 for abdomen and pelvis and 0.014 for chest) [28, 29]. The DLP-to–effective dose calculation model, which uses region-specific k factors, is valid only for helical or sequential scan modes, where the table is incremented through the scan plane during the study. No such k factors have been defined for the stationary table intermittent mode. Hence, effective dose was estimated for intermittent mode scans using a computationally more complete model, whereby a scan is defined over a finite section of anatomy and the dose is estimated to all relevant organs. The typical scan parameters (e.g., peak kilovoltage, tube current–exposure time product, and beam collimation) used in intermittent mode were entered into the ImPACT CT Dose Calculator [30], which uses the organ dose coefficients calculated by the UK’s National Radiological Protection Board [23]. Thus, the dose to any defined organ was able to be calculated for any CTDIvol value, with the

entire CTDIvol being delivered over the same fixed number of tissue slabs. This software tool additionally calculates DLP. Having estimated effective dose and DLP, we calculated a k factor for the intermittent mode ki, defined as the ratio of effective dose to DLP. For these calculations, scan parameters were selected to be similar to those used clinically: 120 kVp, 10-mm beam width, and 100 mAs. Because both effective dose and DLP are linearly proportional to tube current–exposure time product, the influence of tube current–exposure time product was canceled out for the k factor calculation. Therefore, 100 mAs was used only for convenience. This procedure was repeated at varying locations in the chest, abdomen, and pelvis, where the intermittent procedures were typically performed, and were repeated for the different scanners used in our study. A series of conversion factors were obtained at each location, and the average conversion factor (ki = 0.018; range, 0.010–0.026) was used in the effective dose calculation for intermittent mode.

Results A total of 571 patients underwent CTguided interventional procedures during the data collection period (March to July 2008), resulting in 42 cryoablations, 50 aspirations, 329 biopsies, 103 drainages, and 47 injections. Three hundred male and 271 female subjects were included, with an average age of 60 ± 17 years (range, 11–95 years).

The mean and SD of CTDIvol values for each procedure were calculated separately for helical mode and intermittent mode (Table 1 and Fig. 1). The mean total CTDIvol values varied considerably among the five different procedures, as did the ratio of CTDIvol values between the intermittent and helical scan mode. For a given procedure, doses varied substantially among patients, resulting in a large SD, especially in intermittent mode. The mean and SD of DLP for each procedure were also calculated for each scanning mode (intermittent and helical) and for the total procedure (Table 2). Skin dose was calculated from CTDIvol using the relationship presented in equation 4. The mean skin doses were 728 ± 382, 130 ± 104, 128 ± 81, 152 ± 105, and 195 ± 147 mGy (Table 3 and Fig. 2) for the cryoablation, aspiration, biopsy, drain, and injection procedures, respectively. The maximum skin dose observed was 1.95 Gy, and only 1.4% of patients received a skin dose exceeding 1000 mGy. These values are below the dose threshold range for transient (nonpermanent) skin effects (2000–5000 mGy) [31]. Cryoablation had the highest skin dose among the five procedure types. The effective dose was calculated using the DLP values recorded from the patient examination and the k factors for intermittent mode (ki = 0.018) and helical mode (k factor =

TABLE 1: Volumetric CT Dose Index (CTDIvol) for Each Scan Mode and Procedure Type CTDIvol (mGy) Intermittent Mode

Helical Mode

Cryoablation (n = 42)

Procedure

183 ± 338

515 ± 217

Aspiration (n = 50)



89 ± 141

65 ± 41

Biopsy (n = 329)



102 ± 105

56 ± 36

Drain (n = 103)



95 ± 124

79 ± 45

Injection (n = 47)



273 ± 222

26 ± 23

Note—Data are mean ± SD. 800

Intermittent mode Helical mode

700 600 CTDIvol (mGy)

Downloaded from www.ajronline.org by 37.44.207.139 on 01/19/17 from IP address 37.44.207.139. Copyright ARRS. For personal use only; all rights reserved

Radiation Doses in Interventional CT

Fig. 1—Volumetric CT dose index (CTDIvol) for each type of procedure. Error bars show range of mean ± SD.

500 400 300 200 100 0

Cryoablation Aspiration

Biopsy

Drain

Injection

AJR:197, July 2011 W99

Leng et al. TABLE 2: Dose-Length Product (DLP) for Each Procedure DLP (mGy × cm) Mode

Aspiration

Biopsy

Drain

Injection

132 ± 244

97 ± 135

110 ± 111

108 ± 125

198 ± 158

Helical mode

7814 ± 3360

1221 ± 696

792 ± 582

1554 ± 984

367 ± 283

Total

7946 ± 3351

1318 ± 724

902 ± 606

1662 ± 1019

565 ± 348

Intermittent/total (%)

2

7

12

7

35

Helical/total (%)

98

93

88

93

65

Note—Except where noted, data are mean ± SD.

TABLE 3: Skin Dose for Each Procedure Skin Dose (mGy) Mode Intermittent mode

Cryoablation

Aspiration

Biopsy

Drain

Injection

110 ± 203

53 ± 85

61 ± 63

57 ± 75

164 ± 133

Helical mode

618 ± 260

77 ± 49

67 ± 43

95 ± 54

31 ± 27

Total

728 ± 382

130 ± 104

128 ± 81

152 ± 105

195 ± 147

Intermittent/total (%)

15

41

48

38

84

Helical/total (%)

85

59

52

62

16

Note—Except where noted, data are mean ± SD.

0.015). The mean effective doses were 119.7 ± 50.3, 20.1 ± 11.0, 13.8 ± 9.2, 25.3 ± 15.4, and 9.1 ± 5.5 mSv for the cryoablation, aspiration, biopsy, drain, and injection procedures, respectively (Table 4 and Fig. 3). The mean effective dose across all procedure types was 24.1 ± 32.4 mSv (range, 0.8–251.4 mSv), with respective mean contributions of 2.3 mSv (9%) from intermittent mode and 21.8 mSv (91%) from helical mode. To show the relative contributions of intermittent mode and helical mode scans to the total skin dose and the total effective dose, the dose fraction from each mode was calculated for each type of procedure (Fig. 4). For skin doses, the contribution from intermittent mode was comparable to that from helical mode. However, the intermittent mode contributed only a small fraction of the effective dose compared with that from helical mode. This is mainly because of the shorter scan range used in intermittent mode compared with longer scan ranges used in helical mode. Discussion Interventional CT procedures provide minimally invasive and cost-effective diagnosis and treatment methods for various clinical purposes. CT plays a critical role during these procedures to provide accurate image guidance for the operators. Because repeated scans over the same anatomy are usually required, the radiation levels associated with these CT scans might be higher than

W100

Fig. 2—Average skin dose for each type of procedure and each scan mode. Skin Dose (mGy)

Downloaded from www.ajronline.org by 37.44.207.139 on 01/19/17 from IP address 37.44.207.139. Copyright ARRS. For personal use only; all rights reserved

Intermittent mode

Cryoablation

800

Intermittent mode

700

Helical mode Total

600 500 400 300 200 100 0

Cryoablation Aspiration

those associated with routine CT scans. In this study, the risks of deterministic effects (skin erythema or epilation) appear to be very low, because observed skin doses were not found to exceed the threshold for transient skin erythema for even the highest dose procedures. The mean total effective dose (an indicator of risk for stochastic effects) for each procedure ranged from 9.1 mSv (injection) to 120 mSv (ablation). For some biopsy and injection procedures, the effective dose was similar to that of a routine abdomen and pelvis scan (8–11 mSv). The effective dose was higher for aspiration and drain procedures. The highest effective dose was delivered during cryoablation procedures, yielding approximately 10–15 times higher effective doses than a routine abdomen and pelvis CT scan. Substantial variations were observed for both skin doses and effective doses among different procedures, as well

Biopsy

Drain

Injection

as from patient to patient. This reflects the highly individualized nature of interventional CT where multiple factors influence the radiation dose during any given interventional procedure, including, among other things, the clinical task; the number, size, and location of targets; the anatomy surrounding the target and adjacency to other critical structures; patient body habitus; procedure complexity; and operator skill. In most interventional procedures, both intermittent mode and helical mode were used. Because of the intrinsic differences between these two modes, the dose contributions from these two scan modes were very different. Comparing the fractional dose contributions from each mode revealed that, although intermittent mode contributed substantially to the skin dose, its contribution to effective dose was usually small compared with that of the helical mode. This was because of the

AJR:197, July 2011

Radiation Doses in Interventional CT TABLE 4: Effective Dose for Each Procedure Effective Dose (mSv) Mode

Aspiration

Biopsy

Drain

Injection

2.4 ± 4.4

1.7 ± 2.4

2.0 ± 2.0

1.9 ± 2.2

3.6 ± 2.9

Helical mode

117.2 ± 50.4

18.3 ± 10.4

11.9 ± 8.7

23.3 ± 14.8

5.5 ± 4.2

Total

119.7 ± 50.3

20.1 ± 11.0

13.8 ± 9.2

25.3 ± 15.4

9.1 ± 5.5

Intermittent/total (%)

2

9

14

8

39

Helical/total (%)

98

91

86

92

61

Note—Except where noted, data are mean ± SD.

120 Effective Dose (mSv)

Downloaded from www.ajronline.org by 37.44.207.139 on 01/19/17 from IP address 37.44.207.139. Copyright ARRS. For personal use only; all rights reserved

Intermittent mode

Cryoablation

Fig. 3—Average effective dose for each type of procedure and each scan mode.

Intermittent mode Helical mode

100

Total

80 60 40 20 0

Cryoablation Aspiration

Biopsy

Drain

limited scan range (about 1 cm) used in the intermittent mode compared with the longer scan ranges (about 10–30 cm) used in the helical mode. As with any other CT examination, the clinical appropriateness of CT scans in interventional procedures should be considered according to benefit versus risk. InforAblation

Aspiration

Injection

mation obtained from CT images is critical for the success of the procedure, providing operators with direct visualization of patient anatomy and instruments to ensure proper positioning. Accuracy and safety are critical in interventional procedures because errors may results in severe consequences, such as an inaccurate diagnosis (e.g., biopsy or aspiBiopsy

ration that missed the pathologic abnormality), incomplete therapy, or damage to adjacent critical tissues and organs. Therefore, the benefits from CT guidance outweigh the potential radiation risk for medically justified procedures. Furthermore, even for lengthy procedures, the skin dose falls below the threshold where transient skin injury has been observed. However, it is essential to understand the dose levels that could potentially be delivered and to monitor the skin dose to a region by keeping a close eye on the displayed CTDIvol value. Practices should carefully evaluate the dose levels and image quality being used for various procedure types. As reported elsewhere [32], the image quality requirements for observing the ice ball in cryoablation do not require the same dose as for a routine diagnostic examination. According to work in our laboratory and after a review of the data presented here by our Drain

Injection

Drain

Injection

Skin Dose Ablation

Aspiration

Biopsy

Effective Dose Intermittent mode

Helical mode

Fig. 4—Skin dose and effective dose fractions for intermittent mode and helical mode for each type of procedure.

AJR:197, July 2011 W101

Downloaded from www.ajronline.org by 37.44.207.139 on 01/19/17 from IP address 37.44.207.139. Copyright ARRS. For personal use only; all rights reserved

Leng et al. interventional CT staff, we have reduced the doses in our cryoablation procedures by a factor of 2 since the time that this study was performed [32]. Ongoing work in our laboratory promises to further reduce image noise; hence, we anticipate another reduction in dose by a factor of 2 (or more) [33]. There are several limitations of this study. First, skin dose for a procedure was estimated according to measurements made using only one size of anthropomorphic phantom [21]. Patients at the extremes, either very large or very small, might have somewhat different skin doses. Most of our patients were adults (99%), however, and in a separate phantom study, the dependence of surface dose on phantom size over a range of patient lateral width from 25 to 50 cm, was found to be only about ± 10% [22]. For intermittent mode scans, the operator can manually move the table small amounts to best position the target. Our calculations assumed that the table remained stationary and, thus, all skins dose were applied to the same region. Thus, there may have been overestimations of skin dose in our study if patient table positioning was performed with any regularity. The radiologists who performed the majority of the cases studied here indicated that any manual positioning was very small. Thus, our skin dose values serve as a reasonable “worst case” estimate of skin dose, even though it could hypothetically be somewhat lower. The average conversion factor (ki = 0.018) was used in the effective dose calculation for intermittent mode. Depending on scan location, this conversion factor can change from 0.010 to 0.026. The variety depends on whether critical organs are included in the scan range, which is difficult to accurately determine for each specific patient, especially for small organs such as the thyroid and gonads. However, the effective dose from the intermittent mode is only a small fraction of the total effective dose because of the limited scan range. Therefore, using the average conversion factor is adequate to estimate effective dose and to show that its contribution is small compared with helical mode. Finally, this study included data from interventional procedures performed at only one center. The dose levels at other medical facilities might be different because of different devices, protocols, or operators’ experience. For example, the continuous x-ray on “CT fluoro” mode is frequently used in some institutes, but it is not used in our practice. Future directions

W102

for this work would involve multiple medical centers, performing a national interventional CT dose survey. In this manner, diagnostic reference levels could be estimated for specific procedure types. These data would allow practices to compare their dose levels to national averages, providing an opportunity to make doses more uniform by addressing outlier high-dose data points. With the continuing advances of CT technologies, dose reduction methods for general CT scans and specific procedures have been introduced [34, 35]. As with all CT scans, some general rules should be applied to control radiation dose, such as limiting the scan range and tube current to that which is minimally required for the clinical task. Other dose management methods also exist that can specifically benefit interventional procedures. For example, a warning message is usually triggered once the scanning time exceeds a certain threshold (e.g., 100 s). In some scanners, the interventional scan (helical or stationary table) prescription needs to be reloaded after a certain number of scans to alert operators of the accumulated dose. As the dose begins to approach such warning levels, staff members are encouraged to stop and think about their options before continuing. The user could also be encouraged to switch from continuous scan mode to intermittent scan mode, thereby reducing patient (and operator) dose. In summary, we have computed the effective doses and skin doses for five unique types of CT-guided interventional procedures at a large academic medical center. The results showed that deterministic effects (i.e., skin injuries) are extremely unlikely for most procedures. Only for the highest dose procedure, cryoablation, did a small number of patients received skin doses that began to approach the threshold for transient skin effects. The equations provided here to estimate skin dose from the CTDIvol value displayed on the scanner console allow users to more closely monitor skin doses, especially for very complex procedures. Substantial variation existed for effective dose values among different procedures. The typical effective dose values were approximately 1–2 times that of a routine abdomen and pelvis examination. However, for cryoablations, the need to repeat multiple helical scans to visualize a longer scan range can lead to effective doses up to 10 times higher than that for a routine abdomen and pelvis scan. Finally, for the same procedure type, both skin doses

and effective doses varied substantially from patient to patient because of the highly individualized nature of each case. References 1. Atwell TD, Farrell MA, Callstrom MR, et al. Percutaneous cryoablation of 40 solid renal tumors with US guidance and CT monitoring: initial experience. Radiology 2007; 243:276–283 2. Carlson SK, Bender CE, Classic KL, et al. Benefits and safety of CT fluoroscopy in interventional radiologic procedures. Radiology 2001; 219:515–520 3. Daly B, Templeton PA. Real-time CT fluoroscopy: evolution of an interventional tool. Radiology 1999; 211:309–315 4. Gangi A, Dietemann JL, Mortazavi R, Pfleger D, Kauff C, Roy C. CT-guided interventional procedures for pain management in the lumbosacral spine. RadioGraphics 1998; 18:621–633 5. Gianfelice D, Lepanto L, Perreault P, ChartrandLefebvre C, Milette PC. Value of CT fluoroscopy for percutaneous biopsy procedures. J Vasc Interv Radiol 2000; 11:879–884 6. Katada K, Kato R, Anno H, et al. Guidance with real-time CT fluoroscopy: early clinical experience. Radiology 1996; 200:851–856 7. Paulson EK, Sheafor DH, Enterline DS, McAdams HP, Yoshizumi TT. CT fluoroscopy–guided interventional procedures: techniques and radiation dose to radiologists. Radiology 2001; 220: 161–167 8. Poe LB, Petro GR, Matta I. Percutaneous CTguided aspiration of deep neck abscesses. AJNR 1996; 17:1359–1363 9. Silverman SG, Deuson TE, Kane N, et al. Percutaneous abdominal biopsy: cost-identification analysis. Radiology 1998; 206:429–435 10. Fraser-Hill MA, Renfrew DL, Hilsenrath PE. Percutaneous needle biopsy of musculoskeletal lesions. Part 2. Cost-effectiveness. AJR 1992; 158: 813–818 11. Silverman SG, Tuncali K, Adams DF, Nawfel RD, Zou KH, Judy PF. CT fluoroscopy-guided abdominal interventions: techniques, results, and radiation exposure. Radiology 1999; 212:673–681 12. International Commission on Radiological Protection Recommendations of the International Commission on Radiological Protection. Oxford, UK: Pergamon Press, 1977:ICRP publication no. 26 13. Jacobi W. The concept of effective dose: a proposal for the combination of organ doses. Radiat Environ Biophys 1975; 12:101–109 14. Nawfel RD, Judy PF, Silverman SG, Hooton S, Tuncali K, Adams DF. Patient and personnel exposure during CT fluoroscopy-guided interventional procedures. Radiology 2000; 216:180–184 15. Teeuwisse WM, Geleijns J, Broerse JJ, Obermann WR, van Persijn van Meerten EL. Patient and staff dose during CT guided biopsy, drainage and

AJR:197, July 2011

Downloaded from www.ajronline.org by 37.44.207.139 on 01/19/17 from IP address 37.44.207.139. Copyright ARRS. For personal use only; all rights reserved

Radiation Doses in Interventional CT coagulation. Br J Radiol 2001; 74:720–726 16. Joemai RM, Zweers D, Obermann WR, Geleijns J. Assessment of patient and occupational dose in established and new applications of MDCT fluoroscopy. AJR 2009; 192:881–886 17. Tsalafoutas IA, Tsapaki V, Triantopoulou C, Gorantonaki A, Papailiou J. CT-guided interventional procedures without CT fluoroscopy assistance: patient effective dose and absorbed dose considerations. AJR 2007; 188:1479–1484 18. International Electrotechnical Commission. Medical electrical equipment. Part 2-44. Particular requirements for the safety of x-ray equipment for computed tomography, ed. 2.1 Geneva, Switzerland: International Electrotechnical Commission (IEC) Central Office, 2002:IEC publication no. 60601-2-44 19. McNitt-Gray MF. AAPM/RSNA Physics Tutorial for Residents: topics in CT—radiation dose in CT. RadioGraphics 2002; 22:1541–1553 20. Shope TB, Gagne RM, Johnson GC. A method for describing the doses delivered by transmission x-ray computed tomography. Med Phys 1981; 8:488–495 21. Bauhs JA, Vrieze TJ, Primak AN, Bruesewitz MR, McCollough CH. CT dosimetry: comparison of measurement techniques and devices. RadioGraphics 2008; 28:245–253 22. Leng S, Vrieze TJ, Lu Y, McCollough CH. A direct skin dose calculation method in CT scans

without table motion: influence of patient size and beam collimation. Med Phys 2010; 37:3110 23. Jones DG, Shrimpton PC. Survey of CT practice in the UK. Part 3: Normalised organ doses calculated using Monte Carlo techniques. Oxford, UK: National Radiological Protection Board, 1991 24. Zankl M, Panzer W, Drexler G. The calculation of dose from external photon exposures using reference human phantoms and Monte Carlo methods. Part VI: Organ doses from computed tomographic examinations. Neuherberg, Germany: GSF—Forschungszentrum fur Umwelt und Gesundtheit, Institut fur Strahlenschutz, 1991 25. International Commission on Radiological Protection. 1990 recommendations of the International Commission on Radiological Protection. Ann ICRP 1991; 21:publication no. 60 26. International Commission on Radiological Protection. 2007 recommendations of the International Commission on Radiological Protection. Ann ICRP 2007; 37:publication no. 103 27. Jessen KA, Shrimpton PC, Geleijns J, Panzer W, Tosi G. Dosimetry for optimisation of patient protection in computed tomography. Appl Radiat Isot 1999; 50:165–172 28. Shrimpton P. Assessment of patient dose in CT. In: EUR. European guidelines for multislice computed tomography funded by the European Com-

mission 2004: contract number FIGMCT200020078-CTTIP. Luxembourg, Luxembourg: European Commission, 2004:Appendix C 29. Shrimpton PC, Hillier MC, Lewis MA, Dunn M. Doses from computed tomography (CT) examinations in the UK: 2003 review. Chilton, UK: National Radiological Protection Board, 2005:report NRPB-W67 30. ImPACT Group. CT patient dosimetry calculator. Version 0.99x. ImPACT Website. www.impactscan. org/ctdosimetry.htm. Published January 20, 2006. Accessed April 10, 2009 31. Balter S, Hopewell JW, Miller DL, Wagner LK, Zelefsky MJ. Fluoroscopically guided interventional procedures: a review of radiation effects on patients’ skin and hair. Radiology 2010; 254:326–341 32. Leng S, Atwell T, Yu L, et al. Radiation dose reduction for CT-guided renal tumor cryoablation (abstr). AJR 2010; 194(suppl):A51, abstr 126 33. Leng S, Liu X, Yu L, Atwell TD, McCollough CH. Radiation dose reduction in CT guided cryoablation using HYPR-LR. Med Phys 2009; 36:2433 34. McCollough CH, Primak AN, Braun N, Kofler J, Yu L, Christner J. Strategies for reducing radiation dose in CT. Radiol Clin North Am 2009; 47:27–40 35. Yu LF, Liu X, Leng S, et al. Radiation dose reduction in computed tomography: techniques and future perspective. Imaging Med 2009; 1:65–84

AJR:197, July 2011 W103