Fetal dose conversion factor for fetal computed tomography examinations: A mathematical phantom study

Abstract This study aimed to examine the relationship between fetal dose and the dose–length product, and to evaluate the impact of the number of rotations on the fetal doses and maternal effective doses using a 320‐row multidetector computed tomography unit in a wide‐volume mode. The radiation doses for the pregnant woman and the fetus were estimated using ImPACT CT Patient Dosimetry Calculator software for scan lengths ranging from 176 to 352 mm, using a 320‐row unit in a wide‐volume mode and an 80‐row unit in a helical scanning mode. In the 320‐row unit, the fetal doses in all scan lengths ranged from 3.51 to 6.52 mGy; the maternal effective doses in all scan lengths ranged from 1.05 to 2.35 mSv. In the 80‐row unit, the fetal doses in all scan lengths ranged from 2.50 to 3.30 mGy; the maternal effective doses in all scan lengths ranged from 0.83 to 1.68 mSv. The estimated conversion factors from the dose–length product (mGy・cm) to fetal doses (mGy) for the 320‐row unit in wide‐volume mode and the 80‐row unit in helical scanning mode were 0.06 and 0.05 (cm−1) respectively. While using a 320‐row MDCT unit in a wide‐volume mode, operators must take into account the number of rotations, the beam width as automatically determined by the scanner, the placement of overlap between volumetric sections, and the ratio of overlapping volumetric sections.

Fetuses are more radiosensitive than adults and children. 8,9 Therefore, when pregnant women undergo CT examination, it is important to know the associated fetal radiation risk.
CT units generate patient-dose indexes of the volume CT dose index (CTDI vol ) and the dose-length product (DLP) that are measured in 16-and 32-cm diameter acrylic phantoms. The scientific literature contains only limited data regarding effective dose/DLP conversion factors, and there are, to our knowledge, no studies investigating the fetal dose/DLP conversion factors for fetal CT examination. [10][11][12] Elucidating the relationship between fetal doses and DLP could assist in estimating the fetal dose in clinical practice.
Recently, a 320-row multidetector CT (MDCT) unit has become available that allows axial volume-scanning with a 160-mm range.
For imaging neonates and small children, volume-scanning is of potential great advantage as the entire scan can be performed in one rotation. Because there is no over-ranging associated with axial volumetric scanning, this may reduce the patient's radiation dose. 13 In addition, a volume-scanning mode in a 320-row MDCT unit has been used in fetal CT. 14 However, Matsunaga et al. 14 reported that operators must take into account the number of rotations, beam width, and overlap between volumetric sections, because the volumetric scanning mode, using multiple contiguous sections in the 320-row MDCT unit, caused wide overlapping of the volumetric sections. Despite this concern, the relationship between the radiation dose and the number of rotations in a volume-scanning mode has not been described in previous reports.
This study examined the relationship between fetal dose and DLP, and evaluated the impact of the number of rotations on the fetal doses and maternal effective doses using a 320-row MDCT unit in a wide-volume mode.  15 This report provides normalized organ dose data for irradiation of a mathematical phantom. The radiation doses from a 320-row MDCT unit (Aquilion ONE, Toshiba Medical System Corporation, Otawara, Japan) and an 80-row MDCT unit (Aquilion PRIME, Toshiba Medical System Corporation, Otawara, Japan) were estimated. All scan parameters used in this study were based on those reported by Matsunaga et al. 14 (Table 1). The number of rotations and beam width cannot be adjusted manually in the 320-row MDCT unit in a wide-volume mode and were instead automatically determined by the scan length. The scan length is generally based on the fetus length; thus, scan lengths differ due to differences in fetal size. The radiation doses for the pregnant woman and the fetus were estimated by using the following scan lengths (beam width * number of rotations): 176 mm (120 mm * 2), 184 mm (128 mm * 2), 204 mm (140 mm * 2), 232 mm (160 mm * 2), 264 mm (120 mm * 3), 276 mm (128 mm * 3), 306 mm (140 mm * 3), 348 mm (160 mm * 3), and 352 mm (120 mm * 4). The scan length was entered separately from the number of rotations in the spreadsheet (Fig. 1), taking into account the overlap between volumetric sections. The sum of radiation doses from the number of rotations was defined as the estimated radiation dose. This method of estimating radiation doses was based on that used in the study by Matsunaga et al. 16 The 80-row MDCT unit employed a helical scanning mode. Since the 80-row MDCT unit uses a high-pitch factor, there is no overlap between the helical sections. 14 Therefore, the exposure parameters presented in Table 1 respectively. In the 320-row MDCT unit, as the scan length increased, the fetal doses increased. However, an increase in the number of rotations did not necessarily correspond to an increase in the fetal doses, e.g., the fetal dose did not increase between 348 mm (160 mm * 3) and 352 mm (120 mm * 4). The radiation dose to the organs in the pregnant woman included in the scan range (e.g., the colon, small intestine, and ovaries) also yielded similar results. The small intestine doses were the largest in each scan length, except at the uterus/fetus.

| ME TH ODS
In the 80-row MDCT unit, as the scan length increased, the fetal doses increased. A similar trend was observed in the radiation dose to the organs in the pregnant woman.
The ratios of the fetal doses (mGy) to DLP (mGy・cm) are shown in Table 2. The ranges of the ratios of the fetal doses (mGy) to DLP (mGy・cm) in the 320-and 80-row MDCT unit were 0.05-0.06 and 0.03-0.05 respectively.
In the 320-row MDCT unit, the maternal effective dose for each scan length (listed in Fig. 2  80-row MDCT unit in helical scanning mode were 0.06 and 0.05 (cm À1 ), respectively; to be conservative, it is recommended that the highest conversion factors be used. In the 320-row MDCT unit, the fetal doses estimated by multiplying the displayed DLP with the conversion factor calculated in this study (5.91 mGy) were almost the same as those derived from physical measurements using thermoluminescent dosimeters in a previous study (5.50 mGy). 14 Furthermore, the conversion factor calculated in this study yielded similar results for the 80-row MDCT unit. It may be important to understand that these factors will be influenced by the X-ray tube voltage, scan range, and patient size. These find- Scan lengths differ due to differences in fetal size. Therefore, it may not make sense to use the mathematical phantom in the ImPACT software because it does not scale for weight. Angel et al. 20 reported that the abdominal circumference of a pregnant woman, but not the gestational age of the fetus, is significantly correlated with fetal dose. Facilities in Japan perform fetal CT in order to screen for skeletal dysplasia after an average of 30 weeks of gestation. 21 In Japan, the mean abdominal circumference of a 28-to 32- week pregnant woman has been reported to be 89-92 cm. 22 16 reported that the fetal dose differences between the thermoluminescent dosimeters and the ImPACT software measurements were <1 mGy in pregnant patients with an abdominal circumference of 90 cm. Therefore, the fetal dose in a pregnant patient with this abdominal circumference can be estimated using the mathematical phantom in the ImPACT software.

| CONCLUSION S
The approximate fetal dose may be estimated by multiplying the dis-

CONFLI CTS OF INTEREST
The authors declare no conflict of interest.