Organ-Based Dose Current Modulation and Thyroid Shields: Techniques of Radiation Dose Reduction for Neck CT

Medical Physics and Informatics Original Research Hoang et al. Radiation Dose Reduction for Neck CT Medical Physics and Informatics Original Research Jenny K. Hoang 1,2 Terry T. Yoshizumi 1 Kingshuk Roy Choudhury 1 Giao B. Nguyen 1 Greta Toncheva 1 Andreia R. Gafton 1 James D. Eastwood 1 Carolyn Lowry 1 Lynne M. Hurwitz 1 Hoang JK, Yoshizumi TT, Roy Choudhury K, et al. Keywords: CT, neck CT, organ-based dose modulation, radiation dose reduction, tube current modulation DOI:10.2214/AJR.11.7445 Received June 25, 2011; accepted after revision September 5, 2011. J.K. Hoang is a GE-AUR Fellow for 2010 2012. 1 Department of Radiology, Duke University Medical Center, Box 3808, Erwin Rd, Durham, NC 27710. Address correspondence to J. K. Hoang (jennykh@gmail.com). 2 Department of Radiation Oncology, Duke University Medical Center, Durham, NC. AJR 2012; 198:1132 1138 0361 803X/12/1985 1132 American Roentgen Ray Society Organ-Based Dose Current Modulation and Thyroid Shields: Techniques of Radiation Dose Reduction for Neck CT OBJECTIVE. The purpose of this study was to assess the difference in absorbed organ dose and image quality for MDCT neck protocols using automatic tube current modulation alone compared with organ-based dose modulation and in-plane thyroid bismuth shielding. MATERIALS AND METHODS. An anthropomorphic female phantom with metal oxide semiconductor field effect transistor (MOSFET) detectors was scanned on a 64-MDCT scanner. The protocols included a reference neck CT protocol using automatic tube current modulation and three modified protocols: organ-based dose modulation, automatic tube current modulation with thyroid shield, and organ-based dose modulation with thyroid shield. Image noise was evaluated quantitatively with the SD of the attenuation value, and subjectively by two neuroradiologists. RESULTS. Organ-based dose modulation, automatic tube current modulation with thyroid shield, and organ-based dose modulation with thyroid shield protocols reduced the thyroid dose by 28%, 33%, and 45%, respectively, compared with the use of automatic tube current modulation alone (p 0.005). Organ-based dose modulation also reduced the radiation dose to the ocular lens (33 47%) compared with the use of automatic tube current modulation (p 0.04). There was no significant difference in measured noise and subjective image quality between the protocols. CONCLUSION. Both organ-based dose modulation and thyroid shields significantly reduce the thyroid organ dose without degradation of subjective image quality compared with automatic tube current modulation. Organ-based dose modulation has the additional benefit of dose reduction to the ocular lens. O ptimization of CT parameters considers the balance between image quality needed for accurate diagnosis and radiation dose exposure. A neck CT examination extends from the lower portion of the brain to the upper portion of the chest, irradiating several radiosensitive organs, including the eyes, thyroid, and bone marrow. These organs are regarded to have greater stochastic risk for injury and future malignancy, especially with cumulative doses and in younger patients [1, 2]. Prior studies assessing dose reduction techniques for neck CT have focused on thyroid shielding and automatic tube current modulation [3 5]. These forms of dose reduction have been shown to reduce radiation dose to the thyroid by 47% and 78%, respectively, compared with fixed tube current [3, 6]. Organ-based dose modulation is a new technique of radiation dose reduction that modulates tube current and can be used for neck imaging. Current MDCT scanners use tube current modulation software to modulate the tube current in the x-, y-, and z-axes on the basis of body habitus. This is known as automatic tube current modulation or real-time tube current modulation. Organ-based dose modulation similarly modulates along the z- axis on the basis of body habitus, but in contrast to automatic tube current modulation, it modulates the beam in the x- and y-planes by reducing the tube current over a prescribed 120 radial arc over the anterior, lateral, or posterior aspect of the body (operator preference). The overall radiation dose is kept constant with organ-based dose modulation as the tube current is increased in the remaining 240 radial arc. As a result, centrally located organs have a constant radiation exposure, whereas superficially located organs have a reduced exposure within the 120 radial arc and increased exposure over the 240 arc. Organ-based dose modulation has 1132 AJR:198, May 2012

Radiation Dose Reduction for Neck CT been proposed for use in chest and brain imaging to reduce breast and ocular lens dose, respectively [7]. Its use during neck imaging to reduce dose to the thyroid gland has not been evaluated. The purpose of this study was to assess the difference in absorbed organ dose and image quality for MDCT neck protocols by comparing the use of organ-based dose modulation and in-plane thyroid bismuth shielding to a reference protocol with only automatic tube current modulation. Materials and Methods Anthropomorphic Phantom and Dosimetry Institutional review board approval and HIPAA compliance were not required because this was a phantom study. A commercially available anthropomorphic female phantom (model 702-D, CIRS) (Fig. 1) was used for measurement of absorbed organ doses. The phantom is equivalent to a human measuring 160 cm in height and 55 kg in weight. Twenty metal oxide semiconductor field effect transistor (MOSFET) dosimeters (model TABLE 1: MDCT Scanning Parameters A 1002RD, Best Medical) with active detector areas of 200 200 µm (total dimensions, 2.5 mm width 1.3 mm thickness 8 mm length) were placed in specific anatomic locations in the neck, chest, abdomen, and pelvis to allow determination of whole-body effective dose and absorbed organ doses. The detectors were in the following organs: breast, skin over the mid anterior neck, cerebellum, ocular lens, mandible, cervical spine, thyroid, esophagus, upper lung, mid lung, lower lung, sternum, thymus, thoracolumbar spine, heart, liver, kidney, spleen-adrenals, and stomach. Each MOSFET detector was calibrated for the appropriate beam energy, and individual calibration factors for all 20 detectors were stored in a laptop computer. Detailed calibration methods and validation of MOSFET methods have been described previously by Yoshizumi et al. [8]. The lower limit of detection of absorbed dose for the MOSFETs with the Autosense patient dose verification system (TN-RD-60, Best Medical) was 1.50 mgy. The bismuth thyroid shield was 15 cm long, 9 cm wide, and 1 mm thick (40 ml thickness or 4-ply) (AttenuRad, JRT Associates). We attached B Fig. 1 Anthropomorphic phantom. A, Photograph shows anthropomorphic phantom in supine position with leads attached to metal oxide semiconductor field effect transistor (MOSFET) dosimeters. Thyroid bismuth shield has been placed on phantom after obtaining scout image. Note 1-cm foam standoff pad attached to shield. B, Scout image of phantom shows z-axis coverage of neck CT protocols (double-headed arrow). Dashed lines indicate upper and lower boundaries. it to a 1-cm-thick foam standoff pad (Fig. 1). The standoff foam was used to reduce streak artifact over the thyroid [6, 9]. CT Protocols The MDCT neck protocols are summarized in Table 1. A dual source 64-MDCT scanner (Definition, Siemens Healthcare) was used for all examinations. The reference clinical protocol was helically acquired using real-time dose modulation, CARE Dose4D (Siemens Healthcare). CARE Dose4D combines two types of tube current modulation: axial tube current modulation based on the topogram image attenuation profile in the z-axis and angular tube current modulation based on an axial image during the examination in the x- and y-axis [10]. The parameters for this protocol were 120 kvp; effective tube current-seconds, 192 mas; rotation time, 1 second; pitch, 0.9; and beam collimation, 64 0.6 mm (single tube). There were three modified protocols. The organ-based dose modulation protocol used organ-based dose modulation (X-CARE, Siemens Healthcare, adapted thoracic version) in combination with dose modula- CT Protocol ATCM OBDM ATCM With TS OBDM With TS Tube current modulation a On On On On OBDM Off On Off On Bismuth shielding No No Yes Yes Peak beam energy (kvp) 120 120 120 120 Reference tube current (mas) 200 400 200 400 Scan length (cm) 24 24 24 24 Effective tube current (mas) 192 186 192 186 CTDI vol (mgy) 13.86 13.46 13.86 13.46 DLP (mgy cm) 302 306 302 306 Note ATCM = automatic tube current modulation, OBDM = organ-based dose modulation, TS = thyroid shield, CTDI vol = volume CT dose index, DLP = dose-length product. a CARE Dose4D (GE Healthcare) with tube current modulation in the x-, y-, and z-axis. When used in combination with OBDM, tube current modulation only along the z-axis was used. AJR:198, May 2012 1133

Hoang et al. tion, CARE Dose4D. When using CARE Dose4D with organ-based dose modulation, only the z-axis dose modulation of the CARE Dose4D software was used in combination with the 120 reduction in tube output along the radial arc in the x- and y- planes. For this protocol, the parameters were 120 kvp; effective tube current-seconds, 186 mas; rotation time, 1 second; pitch 0.6; and beam collimation, 64 0.6 mm (single tube). For the organbased dose modulation protocol, the dose reduction arc was selected for the anterior portion of the neck to reduce the dose to the thyroid gland and ocular lens. The reference (automatic tube current modulation) protocol and organ-based dose modulation protocols were repeated using a bismuth thyroid shield placed over the lower neck to cover the thyroid gland after the scout images were obtained. These were labeled as the automatic tube current modulation with thyroid shield and organ-based dose modulation with thyroid shield protocols, respectively. The latter protocol was performed to evaluate the additional dose reduction possible with combined techniques. For all protocols, the scan length was constant at 24 cm, covering from sella turcica to the bottom of the aortic arch, and the FOV was constant at 22 cm (Fig. 1). The reconstruction algorithm was 2.5-mm thickness with 2.5-mm increments using a soft-tissue algorithm filter of B31s. The volume CT dose index (CTDI) and dose-length product were matched for each protocol to assess the difference in absorbed organ doses with a constant overall radiation exposure. The phantom was scanned three times for each of the four A Fig. 2 Noise measurements. A C, Noise measurements were obtained from three axial levels at upper neck (just below mandible) (A), thyroid gland (B), and upper mediastinum (C). ROI = region of interest. protocols to obtain the mean and SD of absorbed organ doses. Radiation Dose The mean absorbed organ doses were obtained from the MOSFET dosimeters. The mean organ doses for the modified protocols were compared with the automatic tube current modulation (ATCM) protocol (reference protocol). Comparison was made as absolute values and as percentage organ dose reduction. The equation for percentage dose reduction was [(ATCM protocol) OBDM (modified protocol)] 100 ATCM protocol where ACTM is automatic tube current modulation and OBDM is organ-based dose modulation. The effective dose (ED), an equivalent uniform dose to the entire body, was calculated from the measured organ dose (D) by applying tissueweighting factors (W T ) using publication number 103 of the International Commission on Radiologic Protection (ICRP) [11], and by assuming a radiation-weighting factor (W R ) of 1.0 for x-rays. The effective dose is computed by the following equation: ED = i W T i H i = W R i W T i D i where H i is the equivalent dose for organ i. The effective dose for the organ-based dose modulation, automatic tube current modulation with thyroid shield, and organ-based dose modulation with thyroid shield protocols were compared with the automatic tube current modulation protocol as absolute values and as percentage organ dose reduction. B Image Quality Assessment Image quality was evaluated by quantitative and subjective analysis using an Advantage workstation (GE Healthcare). For quantitative analysis, a circular 1-cm 2 region of interest (ROI) was placed in the soft tissues of the anterior and posterior neck on three selected axial images at the level of the upper neck (just below the mandible), mid neck (at the level of the thyroid gland), and lower neck-upper mediastinum as shown in Figure 2. Noise measurements were made on each of the three CT scans for each protocol by recording the SD of the attenuation value (in Hounsfield units) for the ROIs. The anterior and posterior aspects of the neck were selected because organ-based dose modulation and thyroid shields could result in increased image noise in the region of the anterior neck. For qualitative assessment, images were reviewed independently by two fellowship-trained neuroradiologists with 8 and 18 years of experience. The images were reviewed in soft-tissue window settings (window level, 40 HU; window width, 300 HU). Image quality was assessed on the same three selected axial levels used for quantitative assessment and were chosen to be in a region separated from the MOSFET detectors to exclude any change in image quality due to the presence of the detectors. Readers compared image quality for noise and streak artifact for the modified protocols with images from the automatic tube current modulation protocol and graded images on a 5-point scale. Assessment was performed by placing two images side by side on the same screen, with the anterior neck and chest masked for all images to C 1134 AJR:198, May 2012

Radiation Dose Reduction for Neck CT remove identification of the presence or absence of the bismuth shields. The automatic tube current modulation protocol was identified, but the readers were blinded to the modified protocol. On a 5-point scale, in comparison with the automatic tube current modulation protocol, a score of 5 represented much better image quality with markedly less noise and streak artifact, a score of 4 represented slightly better image quality with slightly less noise and streak artifact, a score of 3 represented no significant difference in image quality with similar noise and artifact, a score of 2 represented slightly inferior image quality with slightly more noise and streak artifact, and a score of 1 represented markedly inferior image quality with markedly more noise and streak artifact. Statistical Analysis Data were entered into a Microsoft Excel, version 12.2.8, spreadsheet. Statistical analyses were performed using Enterprise (version 4.2, SAS Institute) and the R program (R Development Core Team). By scanning the phantom with each protocol three times, we were able to obtain dose data for calculation of the mean organ dose and mean effective dose measurements for each protocol. The mean organ doses for the organ-based dose modulation, organ-based dose modulation with thyroid shield, and automatic tube current modulation with thyroid shield protocols were compared with the organ doses for the reference automatic tube current modulation protocol using a two-factor analysis of variance with all possible pairwise interactions. The cervical spine was used as a reference organ for treatment-type interaction contrasts. The noise values for the modified protocols were also compared with the reference automatic tube current modulation protocol using a three-factor analysis of variance with all possible interactions. The three factors were the type of protocol, side (anterior or posterior), and location (above the thyroid shield, thyroid, mediastinum). A two-tailed p value of less than 0.05 was considered statistically significant. Results Radiation Dose Comparison For the automatic tube current modulation protocol, the organ with the highest dose was the thyroid gland followed by the ocular lens, the bone marrow of the sternum, and the skin over the anterior neck (Table 2). These are all anteriorly located organs. The thyroid gland dose was reduced by 29% with organ-based dose modulation, by 33% with automatic tube current modulation with thyroid shield, and by 45% with organ-based dose modulation with thyroid shield (p 0.005) (Table 2). Using automatic tube current modulation as the base protocol, significant reductions in dose were observed for only one organ (thyroid, p < 0.001) when using the automatic TABLE 2: Mean Organ Dose and Effective Dose (ED) Measured by MOSFET Dosimeters Organ ATCM (mgy) Dose (mgy) tube current modulation with thyroid shield protocol. For the organ-based dose modulation protocol, significant reductions were observed for four organs (both ocular lenses, skin, and thyroid). For the organ-based dose modulation with thyroid shield protocol, significant reductions were observed for the same four organs as organ-based dose modulation plus the sternum (p = 0.0004). There was no statistically significant change in the dose to cervical spine bone marrow with organ-based dose modulation compared with the automatic tube current modulation protocol. However, use of organ-based dose modulation resulted in an increased dose to some posterior structures: the upper lungs (p = 0.01) and the brain (p = 0.02). The calculated effective dose did not change when scanning the neck with the organ-based dose modulation protocol compared with automatic tube current modulation. The automatic tube current modulation with thyroid shield and organ-based dose modulation with thyroid shield protocols resulted in 15% and 21% reduction in the calculated effective dose, respectively. Image Quality Assessment Table 3 shows the noise measurements for the different protocols. By analysis of variance analysis, 91% of the total variability in noise OBDM ATCM With TS OBDM With TS p Reduction (%) Dose (mgy) p Reduction (%) Dose (mgy) p Reduction (%) Anterior structures Anterior skin (overlying the chin) 18.10 (1.51) 9.72 (1.99) 0.002 46.32 15.20 (0.17) 0.14 16.02 10.70 (0.56) 0.0005 40.88 Thyroid 27.63 (1.70) 19.77 (3.00) 0.005 28.47 18.40 (2.44) < 0.0001 33.41 15.20 (2.05) < 0.0001 44.99 Right ocular lens 20.40 (3.80) 10.82 (2.91) 0.0003 46.94 17.93 (2.59) 0.20 12.09 13.20 (0.56) 0.0006 35.29 Left ocular lens 18.30 (2.96) 12.20 (1.39) 0.04 33.33 16.43 (1.75) 0.31 10.20 11.27 (1.42) 0.0006 38.43 Bone marrow mandible 16.77 (4.10) 15.97 (0.68) 0.66 4.77 17.67 (1.80) 0.78 5.37 17.20 (1.76) 0.93 2.56 Bone marrow sternum 20.03 (1.29) 14.23 (2.35) 0.06 28.95 16.60 (2.86) 0.08 17.14 12.53 (2.43) 0.0004 37.44 Posterior structures Brain (cerebellum) 15.50 (2.41) 18.67 (1.21) 0.02 20.43 16.53 (1.75) 0.73 6.67 18.33 (3.63) 0.23 18.28 Lung upper 13.17 (1.96) 16.97 (0.50) 0.01 28.86 12.54 (2.52) 0.67 4.73 16.00 (1.01) 0.23 21.52 Central structures Bone marrow cervical spine 18.07 (4.17) 16.33 (1.80) 0.25 9.63 18.37 (3.10) 0.84 1.66 18.30 (1.15) 0.88 1.27 Esophagus 16.20 (4.00) 17.40 (2.08) 0.17 7.41 14.23 (3.53) 0.29 12.16 15.70 (0.69) 0.73 3.09 Thymus 15.60 (2.19) 15.73 (0.91) 0.38 0.83 15.37 (1.92) 0.80 1.47 13.43 (1.90) 0.26 13.91 Effective dose 3.69 (0.14) 3.51 (0.26) 4.88 3.13 (0.16) 15.18 2.93 (0.15) 20.60 Note Reported p values correspond to the significance of estimated effects in the two-way analysis of variance model. The corresponding analysis of variance table and estimated effect values are shown in Table 4. Data in parentheses are SD. Only the 10 organs with the highest radiation doses (mgy) are shown; all other organs measured had doses of less than 8 mgy. Changes with dose reduction techniques of organ-based dose modulation (OBDM) and thyroid shield (TS) are shown as percentage dose reduction with positive values indicating dose reduction compared with the automated tube current modulation (ATCM) protocol. AJR:198, May 2012 1135

Hoang et al. was attributed to the anterior or posterior location (6%), level in the neck (57%), or the interaction between these two factors (28%). By contrast, only 5% of the variability was due to the type of protocol or its interaction with these factors. There was no statistically significant difference in overall noise between the modified protocols compared with the automatic tube current modulation reference protocol (p, 0.28 0.77). The detailed three-way analysis of variance table and estimated effect values are shown in Table 4. By subjective assessment, both neuroradiologists graded all images as grade 3 on the 5-point scale, indicating no appreciable difference in image quality with the use of organ-based dose modulation or thyroid shields or both compared with the automatic tube current modulation protocol (Fig. 3). Discussion Organ-based dose modulation is a new radiation dose reduction technique for MDCT scanners that was originally intended for thoracic and brain imaging for dose reduction to the breast and ocular lens, respectively [7]. To the best of our knowledge, organ-based dose modulation has not been evaluated for radiation dose reduction for MDCT neck examinations. This study shows that both organ-based dose modulation and thyroid shielding result in significant dose reductions to the thyroid A C D Fig. 3 Images at thyroid level for automatic tube current modulation (A), organ-based dose modulation (B), automatic tube current modulation with thyroid shield (C), and organ-based dose modulation with thyroid shield (D) protocols. Two readers found no difference in subjective image quality. B gland compared with automatic tube current modulation. Additionally, organ-based dose modulation (in contrast to the use of thyroid bismuth shields) reduces dose to the ocular lens without impairing subjective image quality or increasing overall image noise. To date, dose reduction for MDCT neck imaging has been reported with the use of automatic tube current modulation and thyroid bismuth shields. Organ-based dose modulation is a variant of tube current modulation, but there is a paucity of literature evaluating the effect of organ-based dose modulation. The only other work in the literature is by Vollmar and Kalender [7], who evaluated a preclinical version of this software (known as partial scanning) during which the tube current is turned off completely during part of the image acquisition. Those authors compared the partial scanning technique for thoracic imaging to the full scanning technique using bismuth breast shields and showed similar reductions in breast dose for the partial scanning technique (48%) and breast bismuth shielding (47%) [7]. Our study for neck CT shows similar dose savings to the thyroid gland for organ-based dose modulation and bismuth shields: 28% for organ-based dose modulation compared with 33% for the automatic tube current modulation with thyroid shield protocol. An advantage of organ-based dose modulation compared with thyroid shields is additional dose reduction to other anterior structures, such as the ocular lens and sternum. In adult and pediatric patients, the dose reduction to the bone marrow and the ocular lens is important because of the risk of radiationinduced cataracts and leukemia, respectively [12 14]. Furthermore, from a practical perspective, organ-based dose modulation may be easier to implement than thyroid shields in a busy radiology department because the use of thyroid shields requires the technologist to enter the scanning room to place the shield over the patient after the scout images are acquired [15]. Combining both of these TABLE 3: Quantitative Mean Noise Measurements in the Anterior and Posterior Neck Mean Noise at Neck Site ATCM OBDM ATCM With TS OBDM With TS Upper neck Anterior 4.65 (0.32) 5.08 (0.40) 4.86 (0.21) 4.99 (0.48) Posterior 5.91 (0.44) 6.57 (0.48) 5.56 (0.91) 6.09 (0.55) Mid neck Anterior 4.48 (0.23) 4.66 (0.49) 5.68 (0.77) 6.00 (0.51) Posterior 5.24 (0.18) 5.14 (0.35) 5.73 (0.25) 5.46 (0.24) Lower neck Anterior 11.94 (1.37) 11.94 (1.87) 14.80 (1.47) 12.24 (1.01) Posterior 7.43 (0.63) 6.40 (0.31) 7.39 (0.50) 7.13 (0.26) Mean of three levels Anterior 7.02 (0.43) 7.22 (0.71) 8.45 (0.66) 7.75 (0.48) Posterior 6.19 (0.18) 6.04 (0.13) 6.23 (0.31) 6.23 (0.26) Note Data are mean (SD). ATCM = automatic tube current modulation, OBDM = organ-based dose modulation, TS = thyroid shield, CTDI vol = volume CT dose index, DLP = dose-length product. 1136 AJR:198, May 2012

Radiation Dose Reduction for Neck CT TABLE 4: Three-Way Analysis of Variance of Image Noise Parameter Degrees of Freedom Sum of Squares dose reduction techniques (organ-based dose modulation and thyroid shields) may be beneficial in pediatric patients because of their greater stochastic risk from radiation exposure [16]. Our results show that the use of thyroid shields with organ-based dose modulation can almost halve the thyroid organ dose compared with automatic tube current modulation, as well as lower the absorbed dose to the lens of the eye and the bone marrow. An important consideration for CT dose reduction techniques is the effect on image quality. In our study, readers found no difference in subjective image quality assessment for organ-based dose modulation compared with the automatic tube current modulation protocol. Furthermore, organ-based dose modulation did not increase the image noise, even when noise was evaluated separately for the anterior and posterior neck. These findings agree with the study by Vollmar and Kalender [7] of partial scanning in the thorax and other studies for neck CT examination using automatic tube current modulation [3 5, 7]. For protocols that used a bismuth shield, there were no significant differences in subjective evaluation of image quality and only slight increases in image noise (Fig. 3). Prior studies have criticized thyroid bismuth shields for producing significant streak artifact; this effect is most pronounced when the shields are folded or undulate across the patient s body [3, 17]. Our study did not show any increase in streak artifacts from thyroid shields, and we attribute this to the use of a 1-cm-thick foam standoff pad with the manufacturer s bismuth shield. A foam pad has also been used with breast shields and acts to elevate the shield from the patient, thereby displacing streak artifact from the anterior thorax [18, 19]. A potential disadvantage of organ-based dose modulation is the increase in organ dose Mean of Squares F p Variability (%) Protocol 3 6.40 2.13 3.98 0.01 1 Side 1 37.27 37.27 69.43 0 6 Location 2 328.21 164.11 305.76 0 57 Protocol:side 3 4.73 1.58 2.94 0.04 1 Protocol:location 6 11.59 1.93 3.60 0.005 2 Side:location 2 161.86 80.93 150.79 0 28 Protocol:side:location 6 4.27 0.71 1.32 0.26 1 Residuals 48 25.76 0.54 4 Total 580.09 to posterior neck structures because the output of the primary beam may be increased over the posterior 240 arc. Vollmar and Kalender [7] found that partial scanning in the thorax increased the dose to the posteriorly located spinal bone marrow by 15 20%. Although the bone marrow located anteriorly may have a decreased dose, the dose to the bone marrow in total may be increased because a large proportion of the bone marrow lies in the spine. Our study found no increase in dose to the cervical spine, possibly because the detector was placed in the vertebral body, making it more centrally located. The absorbed dose did increase in other posteriorly located organs, such as the upper lung (29%) and cerebellum (20%). Additional sampling of bone marrow along the entire spine may allow better evaluation of this effect. The effective dose remained similar with organ-based dose modulation compared with the automatic tube current modulation protocol, reflecting the balance of change in organ dose with change in tube current along the arc of the patient (i.e., tube current decreases in the anterior aspect and increases in the posterolateral aspect) with a constant central CTDI. The protocols using thyroid shields resulted in moderate decreases in effective dose. There are several limitations to our study. First, only one phantom size was used. Evaluation of this technique should also be studied in larger phantoms or pediatric phantoms because the effect of organ-based dose modulation on radiation dose may differ significantly depending on body size. The posterior beam tube current may vary by different magnitudes in large versus small patients to maintain constant image quality. Second, although we did not find differences in subjective image quality, there are limitations to evaluating image quality in phantoms because iodinated contrast media cannot be administered and the small neck structures cannot be individually evaluated. Third, the current version of organ-based dose modulation is prescribed as a 120 angle and was selected because it was the preset angle for organ-based dose modulation used for thoracic imaging. In imaging the chest, the objective for dose reduction was similar to the neck: to reduce dose to the anterior structures. Future research may be able to determine the ideal radial arc of dose reduction for neck structures. Furthermore, we used only a single scanner for our experiments because at the time of the study organ-based dose modulation was only available on one of our CT scanners. We recognize that there are other techniques to reduce dose, such as lowering the peak kilovoltage or reference tube current, but we elected to keep these factors constant to isolate the effect of organ-based dose modulation on dose reduction. Finally, individual organ doses obtained in our study are point-location specific and may not represent the overall absorbed dose for the entire organs; this is due to the point-dose measurement method used with the MOSFET detectors, i.e., an inherent limitation in the maximal number of detectors that can be placed in the organs. In conclusion, both organ-based dose modulation and thyroid shields reduce the thyroid organ dose without degradation of image quality, but organ-based dose modulation has additional benefits of dose reduction to the ocular lens. Combining thyroid shields with organ-based dose modulation further reduces thyroid organ dose and effective dose and could be considered for pediatric patients who are more sensitive to the long-term risks of radiation exposure. References 1. Brenner DJ, Hall EJ. Computed tomography: an increasing source of radiation exposure. N Engl J Med 2007; 357:2277 2284 2. Brenner D, Elliston C, Hall E, Berdon W. Estimated risks of radiation-induced fatal cancer from pediatric CT. AJR 2001; 176:289 296 3. Leswick DA, Hunt MM, Webster ST, Fladeland DA. Thyroid shields versus z-axis automatic tube current modulation for dose reduction at neck CT. Radiology 2008; 249:572 580 4. Lee EJ, Lee SK, Agid R, Howard P, Bae JM, ter- Brugge K. Comparison of image quality and radiation dose between fixed tube current and combined automatic tube current modulation in craniocervical CT angiography. AJNR 2009; 30:1754 1759 AJR:198, May 2012 1137

Hoang et al. 5. Russell MT, Fink JR, Rebeles F, Kanal K, Ramos M, Anzai Y. Balancing radiation dose and image quality: clinical applications of neck volume CT. AJNR 2008; 29:727 731 6. Hohl C, Wildberger JE, Suss C, et al. Radiation dose reduction to breast and thyroid during MDCT: effectiveness of an in-plane bismuth shield. Acta Radiol 2006; 47:562 567 7. Vollmar SV, Kalender WA. Reduction of dose to the female breast in thoracic CT: a comparison of standard-protocol, bismuth-shielded, partial and tube-current-modulated CT examinations. Eur Radiol 2008; 18:1674 1682 8. Yoshizumi TT, Goodman PC, Frush DP, et al. Validation of metal oxide semiconductor field effect transistor technology for organ dose assessment during CT: comparison with thermoluminescent dosimetry. AJR 2007; 188:1332 1336 9. Lee YH, Park ET, Cho PK, et al. Comparative analysis of radiation dose and image quality between thyroid shielding and unshielding during ct examination of the neck. AJR 2011; 196:611 615 10. Lee CH, Goo JM, Ye HJ, et al. Radiation dose modulation techniques in the multidetector CT era: from basics to practice. RadioGraphics 2008; 28:1451 1459 11. International Commission on Radiological Protection. 2007 recommendations of the International Commission on Radiological Protection (publication no. 103). Ann ICRP 2007; 37:2 4 12. Berrington de González A, Mahesh M, Kim KP, et al. Projected cancer risks from computed tomographic scans performed in the United States in 2007. Arch Intern Med 2009; 169:2071 2077 13. National Research Council; Committee to Assess Health Risks from Exposure to Low Level of Ionizing Radiation. Health risks from exposure to low levels of ionizing radiation: BEIR VII phase 2. Washington, DC: National Academies Press, 2006 14. Hein E, Rogalla P, Klingebiel R, Hamm B. Lowdose CT of the paranasal sinuses with eye lens protection: effect on image quality and radiation dose. Eur Radiol 2002; 12:1693 1696 15. Coursey C, Frush DP, Yoshizumi T, Toncheva G, Nguyen G, Greenberg SB. Pediatric chest MDCT using tube current modulation: effect on radiation dose with breast shielding. AJR 2008; 190:244; [web]w54 W61 16. Mazonakis M, Tzedakis A, Damilakis J, Gourtsoyiannis N. Thyroid dose from common head and neck CT examinations in children: is there an excess risk for thyroid cancer induction? Eur Radiol 2007; 17:1352 1357 17. Gunn ML, Kanal KM, Kolokythas O, Anzai Y. Radiation dose to the thyroid gland and breast from multidetector computed tomography of the cervical spine: does bismuth shielding with and without a cervical collar reduce dose? J Comput Assist Tomogr 2009; 33:987 990 18. Hurwitz LM, Yoshizumi TT, Goodman PC, et al. Radiation dose savings for adult pulmonary embolus 64-MDCT using bismuth breast shields, lower peak kilovoltage, and automatic tube current modulation. AJR 2009; 192:244 253 19. Fricke BL, Donnelly LF, Frush DP, et al. In-plane bismuth breast shields for pediatric CT: effects on radiation dose and image quality using experimental and clinical data. AJR 2003; 180:407 411 FOR YOUR INFORMATION The AJR has made getting the articles you really want really easy with an online tool, Really Simple Syndication, available at www.ajronline.org. It s simple. Click the RSS button located in the menu on the right side of the page. You ll be on your way to syndicating your AJR content in no time. 1138 AJR:198, May 2012