Pediatric Imaging Original Research

Pediatric Imaging Original Research Podberesky et al. Dose Estimates of Pediatric Cardiac CT Angiography Using 320-MDCT Scanner Pediatric Imaging Original Research Daniel J. Podberesky 1 Erin Angel 2 Terry T. Yoshizumi 3 Greta Toncheva 3 Shelia R. Salisbury 4 Christopher Alsip 1 Alessandra Barelli 2 John C. Egelhoff 1,5 Colin Anderson-Evans 3,6 Giao B. Nguyen 3 David Dow 7 Donald P. Frush 3 Podberesky DJ, Angel E, Yoshizumi TT, et al. Keywords: cardiac CT angiography, pediatrics, radiation dosimetry DOI:10.2214/AJR.12.8480 Received December 24, 2011; accepted after revision April 18, 2012. The Department of Radiology at Cincinnati Children s Hospital receives research support from Toshiba America Medical Systems. D. J. Podberesky is a member of the Toshiba professional speaker s bureau. E. Angel and A. Barelli are employees of Toshiba America Medical Systems. This investigation was presented at the 2011 ARRS meeting. 1 Department of Radiology, Cincinnati Children s Hospital Medical Center, 3333 Burnet Ave, MLC 5031, Cincinnati, OH 45229. Address correspondence to D. J. Podberesky (daniel.podberesky@cchmc.org). 2 Toshiba America Medical Systems, Tustin, CA. 3 Radiation Safety Division, Duke University Medical Center, Durham, NC. 4 Center for Epidemiology and Biostatistics, Cincinnati Children s Hospital Medical Center, Cincinnati, OH. 5 Present affiliation: Department of Radiology, Phoenix Children s Hospital, Phoenix, AZ. 6 Present affiliation: Quality Assurance Services, Chula Vista, CA. 7 Department of Radiology, University of Cincinnati College of Medicine, Cincinnati, OH. AJR 2012; 199:1129 1135 0361 803X/12/1995 1129 American Roentgen Ray Society Radiation Dose Estimation for Prospective and Retrospective ECG-Gated Cardiac CT Angiography in Infants and Small Children Using a 320-MDCT Volume Scanner OBJECTIVE. The purpose of this study is to determine patient dose estimates for clinical pediatric cardiac-gated CT angiography (CTA) protocols on a 320-MDCT volume scanner. MATERIALS AND METHODS. Organ doses were measured using 20 metal oxide semiconductor field effect transistor (MOSFET) dosimeters. Radiation dose was estimated for volumetrically acquired clinical pediatric prospectively and retrospectively ECG-gated cardiac CTA protocols in 5-year-old and 1-year-old anthropomorphic phantoms on a 320- MDCT scanner. Simulated heart rates of 60 beats/min (5-year-old phantom) and 120 beats/ min (1- and 5-year-old phantoms) were used. Effective doses (EDs) were calculated using average measured organ doses and International Commission on Radiological Protection 103 tissue-weighting factors. Dose-length product (DLP) was recorded for each examination and was used to develop dose conversion factors for pediatric cardiac examinations acquired with volume scan mode. DLP was also used to estimate ED according to recently published dose conversion factors for pediatric helical chest examinations. Repeated measures and paired Student t test analyses were performed. RESULTS. For the 5-year-old phantom, at 60 beats/min, EDs ranged from 1.2 msv for a prospectively gated examination to 4.5 msv for a retrospectively gated examination. For the 5-year-old phantom, at 120 beats/min, EDs ranged from 3.0 msv for a prospectively gated examination to 4.9 msv for a retrospectively gated examination. For the 1-year-old phantom, at 120 beats/min, EDs ranged from 2.7 msv for a prospectively gated examination to 4.5 msv for a retrospectively gated examination. CONCLUSION. EDs for 320-MDCT volumetrically acquired ECG-gated pediatric cardiac CTA are lower than those published for conventional 16- and 64-MDCT scanners. R ecent advancements in CT technology have allowed more rapid imaging of children with congenital and acquired heart disease, with better spatial and temporal resolution and fewer motion artifacts. These advances have increased the number of, and indications for, cardiac CT angiography (CTA) examinations in children [1]. Radiation exposure, especially with this increasing use, is of paramount concern in children, because they are at increased risk for cancer development at similar radiation exposures compared with adults [2]. A 320-MDCT volume scanner allows axial volumetric scanning up to 16 cm in the z- axis in a single 0.35-second gantry rotation with no table movement [3]. This scanner is alluring to pediatric radiologists because it offers the potential advantages of decreased scan durations, reduced motion artifacts, reduced need for sedation, and reduced con- trast agent volume requirements. In addition, the lack of z-axis overranging, lack of overlapping helical rotations, and minimal penumbral overbeaming with volumetric scanning have the advantage of reducing patient radiation exposure [4]. Effective dose (ED) is a commonly used estimate of patient dose [5]. For CT, ED can be estimated using a dose-length product (DLP) based method. However, this method may be inaccurate [2, 6 8]. The purported inaccuracy of the DLP method may be particularly true with newer CT scanner technology and in pediatric patients [9]. Additionally, organ doses cannot be easily or accurately derived from the DLP. Another method of ED estimation uses metal oxide semiconductor field effect transistors (MOSFET) in organ locations in anthropomorphic phantoms and has been shown to be accurate and relatively efficient [7]. AJR:199, November 2012 1129

Podberesky et al. Fig. 1 5-year-old anthropomorphic phantom. A, Photograph of phantom with metal oxide semiconductor field effect transistor (MOSFET) detectors in place. B, CT scanogram image shows wires leading to in-place MOSFET detectors. C, Representative axial CT image shows three MOSFET detectors (arrows) in place. Several recent reports have found diagnostic-quality images with 320-MDCT volume cardiac CTA in both adults [10] and children, even at very high heart rates [11]. In addition, A recent reports have been published on the estimated ED associated with adult cardiac CTA imaging on a 320-MDCT volume scanner [12 15]. However, to our knowledge, TABLE 1: Locations of Organ Metal Oxide Semiconductor Field Effect Transistor (MOSFET) Detectors in 1- and 5-Year-Old Phantoms Dosimeter No. Organ or Structure Slice No., Phantom B Location No., Phantom 1 Year Old 5 Year Old 1 Year Old 5 Year Old 1 Bone marrow, mandible 6 6 17 17 2 Thyroid 7 9 21 24 3 Thymus 9 11 39 50 4 Bone marrow, ribs 9 11 30 50 5 Bone marrow, spine 9 13 38 75 6 Breast, left 10 12 48 58 7 Lungs, middle 10 11 46 49 8 Heart 11 14 H4 H4 9 Esophagus 12 14 73 85 10 Lungs, lower 12 14 76 82 11 Liver 13 15 96 90 12 Pancreas and stomach 13 16 90 113 13 Spleen 13 16 94 111 14 Kidney and adrenal gland, right 14 16 101 117 15 Liver 15 17 112 120 16 Intestine 15 18 118 141 17 Bone marrow, pelvis 17 22 124 152 18 Urinary bladder 17 22 128 158 19 Gonads 20 25 147 176 20 Skin surface Variable Variable NA NA Note NA = not applicable. there have been only two published reports on this topic in the pediatric population, but those investigators used the potentially inaccurate DLP method of ED estimation [11, 16]. What is lacking from the literature is more accurate dosimetry data, whether in phantoms or in patients, for clinical pediatric cardiac CTA protocols performed on a 320-MDCT scanner. Therefore, the purposes of this investigation are to determine organ doses and EDs for clinically used ECG-gated cardiac CTA protocols for a 1- and 5-year-old phantom on a 320-MDCT volume scanner with MOSFET technology, to compare the MOSFET ED to the DLP ED using a set of recently available helical pediatric chest examination dose conversion factors [9, 17], and to develop cardiacspecific pediatric dose conversion factors for volume scan modes. Materials and Methods Institutional review board approval of this study was not required, because no patients or animals were scanned and no clinical images were reviewed. Phantom and Detector Placement Tissue-equivalent anthropomorphic phantoms of a 1- and 5-year-old child (models 704-D and 705-D, respectively, CIRS) were used (Fig. 1). Organ dose measurements were obtained using 20 MOSFET dosimeters (model TN-502RD-H, Best Medical Canada) placed in defined anatomic locations throughout the phantom (Table 1). Each detector was calibrated at 100 kvp. Detailed calibration methods and validation of MOSFET organ dose measurement methods have been described elsewhere [7]. C 1130 AJR:199, November 2012

Dose Estimates of Pediatric Cardiac CT Angiography Using 320-MDCT Scanner Fig. 2 Diagrammatic representation of ECG-gated scan acquisition modes tested. A, Prospective acquisition at heart rate of 60 beats/min with exposure phase window of 70 80% of R-R of one cardiac cycle. B, Two-segment prospective acquisition at heart rate of 120 beats/min with exposure phase window of 30 40% of R-R of two cardiac cycles. C, Retrospective acquisition at heart rate of 60 beats/min with ECG-gated dose modulation. With exception of 70 80% phase window, tube current is decreased throughout cardiac cycle. D, Retrospective acquisition at heart rate of 120 beats/min with continuous, uniform tube current throughout cardiac cycle. Scanning Parameters All CT examinations were performed on a single 320-MDCT volume scanner (Aquilion ONE, Toshiba Medical Systems), henceforth referred to as a 320-MDCT scanner. The phantoms were scanned supine according to the clinical ECG-gated cardiac CTA protocols at our institution (Fig. 2 and Table 2). All scans were performed at a tube voltage of 100 kvp and a 10-cm scan length (total collimated beam width of 10 cm). Our experience, which has been supported by other investigators using the 320-MDCT scanner, is that a tube voltage of 100 kvp (as opposed to 80 kvp) is an acceptable selection when performing pediatric ECG-gated cardiac CTA examinations, except for the smallest neonates and infants (< 10 kg) [18]. Although up to 16 cm of z-axis scan length can be covered in a single volume acquisition, only 10 cm were required to cover the heart on both the 1- and 5-year-old phantoms. For the 1-year-old phantom, a small FOV (240 mm) was used, and for the 5-year-old phantom, a medium FOV (320 mm) was used. Protocols were performed at simulated heart rates of 60 and 120 beats/min on the 5-year-old phantom and only 120 beats/min on the 1-year-old phantom using a commercially available heart rate simulator (BSM-2351A, Nihon Kohden). The normal heart rate range for a 1-yearold is 90 150 beats/min, and that for a 5-year-old is 65 135 beats/min [19]. We chose to use the upper and lower ends of the normal heart rate spectrum in the 5-year-old phantom to show the dose range that can be expected for children in this age range. The lower rate might also reflect the use of β-blockers at this age by some imagers. A C Two cardiac scan acquisition modes were tested: retrospectively and prospectively ECG-gated acquisition (Fig. 2). The retrospectively gated acquisition mode with ECG tube current modulation uses a variable tube current, with relatively higher tube current during diastole and reduced tube current during the other cardiac phases. In this investigation, because of the routine low tube current used for pediatrics, the variation in tube current was only 10%. In other words, the maximum tube current was only 10% higher than the minimum tube current. To accommodate the necessary temporal resolution for the higher heart rate of 120 beats/min, the retrospective mode automatically selects a continuous tube current with no modulation. The retrospective scan mode can be useful for functional analysis because images can be reconstructed at all phases of the R-R. For prospective acquisition, imaging is performed during the quiescent phase of the cardiac cycle only and the tube current is turned off during the remainder of the cycle. For patients with heart rates less than 70 beats/min, images of the entire heart are acquired from a single heartbeat. To accommodate the necessary temporal resolution for the higher heart rate of 120 beats/min, the prospective mode automatically selects a multisegmented two-beat acquisition in which segments of data from adjacent R-R s are combined. The scan modes used in this study are summarized in Table 2. Measurement Tabulation For each protocol on each phantom, organ dose was determined and ED was subsequently calculated. Volume CT dose index (CTDI vol ) and DLP were recorded for each protocol as well. To calculate ED using MOSFET technology, three replications for each protocol were performed to measure the mean (± SD) absorbed organ doses. To increase photon statistics for each MOSFET reading and to minimize source path geometry variations between scans, each replication included three consecutive acquisitions, each with a different tube start location (0, 120, and 240 ). The resulting MOSFET reading was divided by three. The organ doses and EDs were computed separately for each of the three replication sets of three scans. For organs with multiple MOSFET locations, the average absorbed dose was calculated. Each average organ dose was weighted using tissue-weighting factors according to International Commission on Radiological Protection (ICRP) 103 guidelines [20], and the ED was obtained by summing the dose equivalent for each organ. The CT scanogram was not included in ED determinations because it does not contribute substantially to the total ED of a CT scan [21]. In addition, on the basis of scanner-displayed CTDI vol, we estimate that the bolus tracking series comprises less than 10% of the total ED of a cardiac CTA, and, therefore, it was not included in ED determinations. The ED was also estimated by the DLP method using the displayed DLP from the console for each protocol using the following formula: ED(mSv)= k (msv mgy 1 cm 1 ) DLP(mGy cm) where k is a body part specific dose conversion factor [9, 17]. Because the scan length used in this investigation was only 10 cm, the CTDI vol (and thus DLP) was determined in accordance with Interna- B D AJR:199, November 2012 1131

Podberesky et al. TABLE 2: ECG-Gated Cardiac CT Angiography (CTA) Scan Protocols Phantom Years Scan Protocol tional Electrotechnical Commission standard 60601 2-44 Ed.3. We chose to use the recently published conversion factors by Deak et al. [9], which are derived as a function of ICRP 103 tissue-weighting factors, tube voltage (in this case 100 kvp), body region, and age. For the 1-year-old phantom, a chest conversion factor of 0.048 msv mgy 1 cm 1 was used, and for the 5-year-old phantom, 0.032 msv mgy 1 cm 1 was used [9]. The pediatric chest conversion factors by Deak et al. are reported for a body (32-cm) CTDI phantom. The 320- MDCT scanner displays DLP for a head (16-cm) phantom for small-fov scans, which were used for the 1-year-old phantom. To calculate ED from the DLP conversion factors, one must reference the same phantom size. Therefore, when estimating ED from DLP for the 1-year-old phantom, a division factor of 1.9 (determined on the CT scanner used for this investigation) was included in the calculation to normalize to a 32-cm phantom. Statistical Analysis Repeated measures analyses were performed to compare the ED among the protocols using the mixed procedure with the Kenward-Roger correction. For each protocol, a paired Student t test was used to compare the ED calculated from the MOS- FET organ dose measurements to the estimated ED calculated using DLP. All analyses were performed using SAS (version 9.2, SAS Institute) software using two-sided tests and were specified a priori; p values 0.05 or less were considered statistically significant. Reported p values are not adjusted for multiple testing. Results are reported as means (± SD). Results The measured absorbed organ doses are summarized in Table 3. The MOSFET- and DLP-estimated EDs for the various protocols on each phantom are provided in Table 4. Heart Rate (beats/min) Peak Tube Current (ma) Exposure Phase Window Organ doses varied depending on the specific acquisition type. The highest organ doses from all protocols (in descending maximum recorded dose order) were in the bone surface (10.64 37.57 mgy), breast (4.04 15.18 mgy), skin (4.14 14.48 mgy), thymus (3.43 12.86 mgy), esophagus (2.68 11.27 mgy), heart (2.51 11.11 mgy), and lungs (2.55 11.00 mgy). The ED was lower for prospectively compared with retrospectively acquired cardiac CTA protocols. In the 1-yearold phantom, the ED for the prospectively acquired scan was approximately 40% lower than the ED for the retrospectively acquired scan (p < 0.001). In the 5-year-old phantom, at 60 beats/min, the ED for the prospectively acquired scan was approximately 73% lower than the ED for the retrospectively acquired scan (p < 0.001). In the 5-year-old phantom, at 120 beats/min, the ED for the prospectively acquired scan was approximately 39% lower than the ED for the retrospectively acquired scan (p < 0.0024). For the 1-year-old phantom, the MOS- FET ED was 69% higher than the DLP ED for the prospective scan and 196% higher for the retrospective scan (after accounting for the phantom size differences between that reported on the console [16 cm] vs the used conversion factors [32 cm]). For the 5-yearold phantom, the MOSFET ED was 235% and 250% higher than the DLP ED for the prospective scans at 60 and 120 beats/min, respectively, and 300% and 306% higher for the retrospective scans at 60 and 120 beats/ min, respectively. Adjusted cardiac-specific conversion factors based on the MOSFET ED and console-displayed DLP ED compared with published conversion factors for pediatric chest CT are provided in Table 4. No. of R-R Intervals Gantry Rotation Time (s) 1 Prospectively gated CTA 120 80 30 40% of R-R 2 0.4 Small 1 Retrospectively gated CTA/cardiac functional 120 80 Continuous 1 0.4 Small assessment 5 Prospectively gated CTA 60 110 70 80% of R-R 1 0.35 Medium 5 Retrospectively gated CTA/cardiac functional assessment 60 110 70 80% of R-R FOV 1 0.35 Medium 5 Prospectively gated CTA 120 110 30 40% of R-R 2 0.4 Medium 5 Retrospectively gated CTA/cardiac functional assessment 120 110 Continuous 1 0.4 Medium Note Scan parameters were automatically selected for pediatric patients on the basis of patient age and heart rate. All scans were performed at 100 kvp and with a 10-cm scan length. A small FOV is 240 mm, and a medium FOV is 320 mm. Discussion The 320-MDCT volume scanner has theoretic advantages of increased temporal resolution, decreased scan duration, and decreased motion artifacts compared with conventional 16-MDCT and non dual-source 64-MDCT scanners. Volume acquisitions can potentially decrease radiation exposure during cardiac imaging, compared with helical acquisitions, because of the lack of overranging, minimal overbeaming, and avoidance of low-pitch helical scanning with volume acquisition [4]. Diagnostic-quality imaging for cardiac CTA using this volume CT scanner has been shown recently in adults and children [10, 11]. As use of this volume scanner technology potentially becomes more widespread, it is important to have more accurate dosimetry information than that currently available in the medical literature. Our study showed a statistically significant reduction in radiation dose using a prospectively gated technique compared with the retrospective technique. Recently, Al-Mousily et al. [16] reported a mean estimated ED of 0.8 msv using the DLP method with an ICRP 60 pediatric chest conversion factor (0.039 msv mgy 1 cm 1 ) on eight prospectively ECG-gated cardiac CTA examinations performed on predominantly young children (age range, 0.1 55 months) with heart rates above 80 beats/min on a 320-MDCT scanner with an 80-kVp tube voltage, and scan ranges varying between 8 and 16 cm. Al-Mousily et al. manually selected a single-heartbeat technique ( Target mode) for prospective acquisitions, as opposed to the automatically selected two-heartbeat prospective technique used for our measurements at 120 beats/min [16]. This distinction, along with 1132 AJR:199, November 2012

Dose Estimates of Pediatric Cardiac CT Angiography Using 320-MDCT Scanner TABLE 3: Organ Radiation Dose Measured by Metal Oxide Semiconductor Field Effect Transistor Detectors During ECG-Gated Cardiac CT Angiography in 1- and 5-Year-Old Anthropomorphic Phantoms Organ Location Prospective 1-Year-Old Phantom Retrospective Prospective (60 beats/min) Retrospective (60 beats/min) their lower tube voltage setting, may partly account for the difference between their DLP ED of 0.8 msv and our DLP ED of 1.3 and 1.4 msv for 120 beats/min prospective scans in the 1-year-old and 5-year-old phantoms, respectively, when using the same conversion factor [16, 17]. For both the study by Al-Mousily et al. [16] and our investigation, cardiac CTA doses were comparatively lower than published doses on helical non dual-source MDCT. Ou et al. [22] reported a mean DLP ED of 4.5 msv on 126 prospectively ECG-gated cardiac CTA examinations performed on 5- to 6-year-old patients with heart rates below 80 beats/min on a 64-MDCT scanner at 80 kvp. Herzog et al. [23], using 64-MDCT clinical scan parameters and a commercially available dose-estimation software package, reported a mean estimated ED on nongated pediatric cardiac CTA examinations using automated exposure controls of 2.5 msv. Using the DLP method, an adult chest conversion factor of 0.016 msv mgy 1 cm 1, and a 120-kVp technique on a 64-MDCT, Horiguchi et al. [24] reported an estimated ED of 3.0 msv for a prospectively ECG-gated cardiac CTA. If one were to apply the published 120-kVp conversion factors by Deak et al. [9] for a 1- and 5-year-old chest CT of 0.047 and 0.031 msv mgy 1 cm 1 on the scans used by Horiguchi et al. [24] (assuming the same imaging parameters), the resultant estimated EDs would be 8.1 and 5.3 msv, respectively. To our knowledge, there is only one published report of radiation dose associated with volumetrically acquired retrospectively gated cardiac CT in children. Using the DLP method with a range of ICRP 60 pediatric chest conversion factors based on patient age (0.013 0.039 msv mgy 1 cm 1 ), Greenberg et al. [11], using 80 kvp and 250 ma (87.5 mas) and a mean volume length of 11 cm, reported an ED of 8.0 msv in primarily infants on the 320-MDCT volume scanner. This ED is nearly double our MOSFET ED for a retrospectively ECG-gated cardiac CTA in a 1-year-old of 4.5 msv. In addition to the inherent DLP ED estimation issues discussed already, we think that the higher tube current used by Greenberg et al. partly explains this disparity. Huang et al. [1], using a 5-year-old 5-Year-Old Phantom Prospective Retrospective Bone marrow 0.95 ± 0.05 1.71 ± 0.09 0.72 ± 0.01 1.57 ± 0.05 2.44 ± 0.07 2.53 ± 0.06 Lung 6.19 ± 0.09 10.38 ± 0.26 2.55 ± 0.03 6.88 ± 0.26 9.69 ± 0.38 11.00 ± 0.40 Liver 0.74 ± 0.07 1.21 ± 0.12 0.05 ± 0.09 0.75 ± 0.09 1.24 ± 0.26 1.08 ± 0.08 Adrenal glands 0.64 ± 0.07 0.90 ± 0.11 0.03 ± 0.06 0.03 ± 0.23 0.86 ± 0.66 0.65 ± 0.39 Stomach 1.57 ± 0.21 2.61 ± 0.14 0.07 ± 0.06 0.89 ± 0.22 1.10 ± 0.42 1.20 ± 0.22 Bone surface 19.87 ± 0.99 32.61± 0.67 10.64 ± 0.67 23.05 ± 1.36 34.55 ± 0.87 37.57 ± 0.41 Skin 7.26 ± 0.29 12.02 ± 0.17 4.14 ± 0.33 9.23 ± 0.24 13.53 ± 0.56 14.48 ± 0.33 Thyroid 0.64 ± 0.00 1.07 ± 0.11 0.47 ± 0.12 0.90 ± 0.24 1.60 ± 0.30 1.60 ± 0.13 Left breast 7.00 ± 0.10 11.93 ± 0.15 4.04 ± 0.19 9.00 ± 0.60 14.2 ± 0.19 15.18 ± 0.59 Esophagus 5.86 ± 0.30 9.43 ± 0.67 2.68 ± 0.40 6.93 ± 0.37 9.77 ± 0.73 11.27 ± 0.50 Heart 7.13 ± 0.52 11.11 ± 0.44 2.51 ± 0.13 6.64 ± 0.30 9.31 ± 0.48 10.63 ± 0.42 Thymus 7.03 ± 0.26 11.50 ± 0.21 3.43 ± 0.18 7.86 ± 0.32 12.06 ± 0.32 12.86 ± 0.23 Spleen 1.15 ± 0.12 1.95 ± 0.12 0.17 ± 0.21 0.50 ± 0.00 0.81 ± 0.10 0.84 ± 0.06 Right kidney 0.64 ± 0.07 0.90 ± 0.11 0.03 ± 0.06 0.28 ± 0.24 0.86 ± 0.66 0.65 ± 0.39 Pancreas 1.57 ± 0.21 2.61 ± 0.14 0.07 ± 0.06 0.89 ± 0.22 1.10 ± 0.42 1.20 ± 0.22 Intestine 0.32 ± 0.11 0.54 ± 0.00 0 a 0.08 ± 0.07 0.04 ± 0.07 0 a Ovaries 0.04 ± 0.07 0.19 ± 0.07 0 a 0.04 ± 0.07 0 a 0 a Colon 0.32 ± 0.11 0.54 ± 0.00 0 a 0.08 ± 0.07 0.04 ± 0.07 0 a Bladder 0.39 ± 0.27 0.19 ± 0.07 0 a 0.04 ± 0.07 0 a 0 a Note Data are mean ± SD organ radiation dose (mgy). a Zero denotes that no dose was detected for any replication. phantom and thermoluminescent dosimeters with ICRP 103 tissue-weighting factors on a 64-MDCT with retrospective ECG-gating at 100 kvp, reported EDs of 16.5, 12.2, 12.0, and 11.8 msv at simulated heart rates of 40, 60, 70, and 90 beats/min, respectively. Our ED measurements are approximately 60% lower than those of Huang et al., even at a heart rate of 120 beats/min. We think that this difference is mostly attributable to the lack of overranging, minimal overbeaming, and avoiding the need for low-pitch acquisition on a 320-MDCT volume scanner compared with a helical acquisition on a conventional 64-MDCT scanner without adaptive collimation. We are also currently assessing dosimetry data for ECG-gated pediatric cardiac CTA on our 64-MDCT scanner for comparison purposes. Several investigators have reported lower estimated ED on both prospectively and retrospectively ECG-gated cardiac CTA in children and adults compared with this investigation using dual-source CT scanners [25, 26]. Disadvantages of dual-source CT scanner technology compared with 320-MDCT AJR:199, November 2012 1133

Podberesky et al. TABLE 4: Effective Dose (ED) Estimates by Metal Oxide Semiconductor Field Effect Transistor (MOSFET) and Dose-Length Product (DLP) Methods for Pediatric ECG-Gated Cardiac CT Angiography in 1-Year-Old and 5-Year-Old Anthropomorphic Phantoms Phantom (Heart Rate), Scan MOSFET ED (msv) CTDI vol (mgy) DLP (mgy cm) a DLP ED (msv) b 1-year-old phantom volume cardiac CTA include stairstep artifacts on image reformations, slower image acquisition increasing the potential for acquisition during an arrhythmia and patient motion, and the inability to capture the entire heart volume in a state of iodinated contrast at a single point in time [27]. In the present study, MOSFET ED estimates were 2.4 3.1 times greater than DLP ED estimates for the 5-year-old phantom in both the prospective and retrospective scans at both tested heart rates and were 1.7 2 times greater for the 1-year-old phantom. Our calculated cardiac conversion factors, which reference the 16-cm CTDI phantom, for the 1-yearold phantom were 0.85 times less for the prospective scan, to nearly equivalent for the retrospective scan, compared with the chest conversion factors determined by Deak et al. [9], which reference the 32-cm CTDI phantom, and 1.6 1.9 times greater than the published pediatric chest conversion factors based on ICRP 60, which similarly reflect the 16-cm CTDI phantom [17]. Our calculated cardiac conversion factors for the 5-year-old phantom were 2.3 3.0 times higher than chest conversion factors determined by Deak et al. [9] for the 5-year-old phantom for both prospective and retrospective scans, and 4.2 5.4 times greater than the published pediatric chest conversion factors based on ICRP 60 [17]. This disparity has also been reported by other investigators comparing both MOSFET- and Published Chest Conversion Factor (msv mgy 1 cm 1 ) [9] Calculated Conversion Factor (msv mgy 1 cm 1 ) Prospective gating 2.7 ± 0.04 c 6.5 64.8 1.6 a,d 0.048 0.042 Retrospective gating 4.5 ± 0.05 9.0 90.4 2.3 a 0.048 0.050 5-year-old phantom (60 beats/min) Prospective gating 1.2 ± 0.03 e 1.6 16.1 0.51 d 0.032 0.075 Retrospective gating 4.5 ± 0.08 4.7 47.2 1.5 d 0.032 0.095 5-year-old phantom Prospective gating 3.0 ± 0.11 f 3.6 36.2 1.2 d 0.032 0.083 Retrospective gating 4.9 ± 0.09 5.1 50.5 1.6 d 0.032 0.097 a The displayed volume CT dose index (CTDI vol ) and DLP for the 5-year-old phantom, based on a medium (320-mm) FOV, are reported for a body (32-cm) phantom. The displayed CTDI vol and DLP for the 1-year-old phantom, based on a small (240-mm) FOV, are reported for a head (16-cm) phantom. The pediatric chest conversion factors used are based on the 32-cm phantom. Therefore, ED by DLP estimates for the 1-year-old phantom are normalized by dividing by a factor of 1.9 (factor was derived on scanner for relevant protocol). b ED = k DLP, where k is a region-specific (in this case, chest) dose conversion factor. c p < 0.001, compared with 1-year-old phantom retrospective acquisition. d p < 0.01, compared with ED by MOSFET method. e p < 0.001, compared with 5-year-old phantom (60 beats/min) retrospective acquisition. f p < 0.0024, compared with 5-year-old phantom retrospective acquisition. Monte Carlo simulation based dose estimates to DLP-based dose estimations in adult cardiac and body CT [2, 5, 16, 28]. The discrepancy between MOSFET ED and DLP ED can likely be explained by several factors. Notably, the Deak et al. pediatric chest conversion factors are based on one model of scanner and were developed for a helical chest scan as opposed to a volume cardiac examination. In addition, the Deak et al. pediatric chest conversion factors are based on a scanner that reports all doses normalized to a 32-cm phantom, as opposed to the 16-cm phantom used for the 1-year-old phantom in this study. We normalized this difference by dividing our 1-year-old DLP ED values by 1.9, a value that was determined on the CT scanner used for this investigation. This emphasizes the importance of ensuring that the CTDI phantom size reported on the scanner matches the phantom size used to develop the dose conversion factors. We think it is crucial that radiologists be familiar with the meaning of their CT scanner s dose display values, DLP conversion factors, and inherent limitations when using this information to estimate CT ED. There is often confusion when using the DLP method to estimate ED, because DLP reflects the total amount of radiation imparted on an acrylic phantom by the scanner, not the dose absorbed by the patient [5], and because of the variety of conversion factors that can be found in the literature [9, 17]. Our calculated conversion factors support that published conversion factors for pediatric chest examinations may be inaccurate for estimating dose for pediatric cardiac examinations, specifically resulting in low ED estimations in general, as in this investigation. Further dose assessment for cardiac CTA is necessary for more-accurate dose estimations in children. We hope that this investigation encourages this type of individualized protocolspecific dose analysis. Our study had several limitations. We only tested our protocols in 1- and 5-year-old phantoms. However, these two phantom sizes best approximate the majority of our clinical pediatric cardiac CTA examinations. The phantoms also do not account for body shape differences at those ages and the effect on ED of those potential differences. Additionally, we only tested our protocols on simulated heart rates of 60 and 120 beats/min. We think that these two heart rates appropriately cover the range of normal heart rates in this age range, with or without the use of β-blockers [19]. We chose to exclude the scanogram and bolus tracking from the ED determination of these protocols, which results in a very small underestimation of the total dose associated with a complete cardiac CTA examination. Although we did not assess image quality or diagnostic capability in this investigation, the parameters studied are those used for our clinical ECG-gated pediatric cardiac CTA examinations and were specifical- 1134 AJR:199, November 2012

Dose Estimates of Pediatric Cardiac CT Angiography Using 320-MDCT Scanner ly designed to mimic the acceptable image quality obtained in helical scan mode on our 64-MDCT scanner for our clinical practice. These parameters are similar to those used by Greenberg et al. [11] in their recent investigation showing diagnostic-quality imaging in clinical pediatric cardiac CTA on the 320-MDCT volume scanner even at heart rates above 120 beats/min. We also recognize the limitations of using published pediatric chest conversion factors for DLP ED estimations for cardiac CTA [5]. Pediatric-specific cardiac conversion factors are not available in the literature. We recognize the inherent limitations and potential error in using ED determination in medical imaging for the purposes of risk-benefit estimates, though we think it is an acceptable and familiar quantitative measurement for medical imaging dose estimations [29 31]. In conclusion, as CT scanner technology and applications continue to evolve, it is important that radiologists understand radiation doses associated with commonly performed CT examinations in children. Knowing the ED associated with a CT scan allows the radiologist, clinicians, patient, and parents to make a better informed risk-benefit decision. For pediatric ECG-gated cardiac CTA examinations on a 320-MDCT scanner, using MOSFET dosimetry, we found that the ED range was 1.2 4.9 msv depending on prospective versus retrospective gating, phantom size (1- vs 5-year-old), and heart rate (60 vs 120 beats/min). These doses are lower than those previously reported for pediatric cardiac CTA on 16-MDCT and conventional 64-MDCT scanners. Acknowledgments We thank Lane Donnelly, Brian Coley, Alan Brody, Marilyn Goske, David Larson, and Lisa Lemen for their support of this investigation. References 1. Huang B, Law MW, Mak HK, Kwok SP, Khong P. Pediatric 64-MDCT coronary angiography with ECG-modulated tube current: radiation dose and cancer risk. AJR 2009; 193:539 544 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. Choi SI, George RT, Schuleri KH, et al. Recent developments in wide-detector cardiac computed tomography. Int J Cardiovasc Imaging 2009; 25:23 29 4. Kroft LJ, Roelofs JJ, Geleijns J. Scan time and patient dose for thoracic imaging in neonates and small children using axial volumetric 320-detector row CT compared to helical 64-, 32- and 16-detector row CT acquisitions. Pediatr Radiol 2010; 40:294 300 5. Huda W, Tipnis S, Sterzik A, Shoepf UJ. Computing effective dose in cardiac CT. Phys Med Biol 2010; 55:3675 3684 6. Hollingsworth CL, Yoshizumi TT, Frush DP, et al. Pediatric cardiac-gated CT angiography: assessment of radiation dose. AJR 2007; 189:12 18 7. 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 8. American Association of Physicists in Medicine. Report 204: size specific dose estimates (SSDE) in pediatric and adult body CT examinations. College Park, MD: American Association of Physicists in Medicine, 2011 9. Deak PD, Smal Y, Kalendar WA. Multisection CT protocols: sex- and age-specific conversion factors used to determine effective dose from dose length product. Radiology 2010; 257:158 166 10. Lee AB, Nandurkar D, Schneider-Kolsky ME, et al. Coronary imaging quality of 320-detector MDCT in patients with heart rates above 65 beats per minute: preliminary experience. AJR 2011; 196:1313; [web]w729 W735 11. Greenberg SB, Bhutta S, Braswell L, Chan F. Computed tomography angiography in children with cardiovascular disease: low dose techniques and image quality. Int J Cardiovasc Imaging 2012; 28:163 170 12. Hoe J, Toh KH. First experience with 320-row multidetector CT coronary angiography scanning with prospective electrocardiogram gating to reduce radiation dose. J Cardiovasc Comput Tomogr 2009; 3:257 261 13. Zimmermann E, Dewey M. Whole-heart 320-row computed tomography: reduction of radiation dose via prior coronary calcium scanning. Rofo 2011; 183:54 59 14. Dewey M, Zimmermann E, Deissenrieder MD, et al. Noninvasive coronary angiography by 320- row computed tomography with lower radiation exposure and maintained diagnostic accuracy: comparison of results with cardiac catheterization in a head-to-head pilot investigation. Circulation 2009; 120:867 875 15. Einstein AJ, Elliston C, Arai AA, et al. Radiation dose from single-heartbeat coronary CT angiography performed with a 320-detector row volume scanner. Radiology 2010; 254:698 706 16. Al-Mousily F, Shifrin RY, Fricker FJ, Feranec N, Quinn NS, Chandran A. Use of 320-detector computed tomographic angiography for infants and young children with congenital heart disease. Pediatr Cardiol 2011; 32:426 432 17. American Association of Physicists in Medicine. Report 96: the measurement, reporting, and management of radiation dose in CT report of AAPM Task Group 23 of the Diagnostic Imaging Council CT Committee. College Park, MD: American Association of Physicists in Medicine, 2008 18. Siripornpitak S, Pornkul R, Khowsathit P, Layangool T, Promphan W, Pongpanich B. Cardiac CT angiography in children with congenital heart disease. Eur J Radiol 2011 Dec 21; [Epub ahead of print] 19. Custer JW, Rau RE, eds. Johns Hopkins: the Harriett Lane handbook, 18th ed. Philadelphia, PA: Mosby Elsevier, 2008 20. [No authors listed]. The 2007 recommendations of the International Commission on Radiological Protection: ICRP publication 103. Ann ICRP 2007; 37:1 332 21. O Daniel JC, Stevens DM, Cody DD. Reducing radiation exposure from survey CT scans. AJR 2005; 185:509 515 22. Ou P, Celermajer DS, Marini D, et al. Safety and accuracy of 64-slice computer coronary angiography in children after the arterial switch operation for transposition of the great arteries. JACC Cardiovasc Imaging 2008; 1:331 339 23. Herzog C, Muvhihill DM, Nguyen SA, et al. Pediatric cardiovascular CT angiography: radiation dose reduction using automatic tube current modulation. AJR 2008; 190:1232 1240 24. Horiguchi J, Kiguchi M, Fujioka C, et al. Radiation dose, image quality, stenosis measurement, and CT densitometry using ECG-triggered coronary 64-MDCT angiography: a phantom study. AJR 2008; 190:315 320 25. Ben Saad M, Rohnean A, Sigal-Cinqualbre A, Adler G, Paul JF. Evaluation of image quality and radiation dose of thoracic and coronary dualsource CT in 110 infants with congenital heart disease. Pediatr Radiol 2009; 39:668 676 26. Achenbach S, Marwan M, Ropers D, et al. Coronary computed tomography angiography with a consistent dose below 1 msv using prospectively electrocardiogram-triggered high-pitch spiral acquisition. Eur Heart J 2010; 31:340 346 27. Otero HJ, Steigner ML, Rybicki FJ. The post- 64 era of coronary CT angiography: understanding new technology from physical principles. Radiol Clin North Am 2009; 47:79 90 28. Hurwitz LM, Yoshizumi TT, Goodman PC, et al. Effective dose determination using an anthropomorphic phantom and metal oxide semiconductor field effect transistor technology for clinical adult body multidetector array computed tomography protocols. J Comput Assist Tomogr 2007; 31:544 549 29. Martin CJ. Effective dose: how should it be applied to medical exposures? Br J Radiol 2007; 80:639 647 30. Brenner DJ. Effective dose: a flawed concept that could and should be replaced. Br J Radiol 2008; 81:521 523 31. McCollough CH, Christner JA, Kofler JM. How predictive is effective dose as a predictor of radiation risk? AJR 2010; 194:890 896 AJR:199, November 2012 1135