FEASIBILITY STUDY OF USING PCXMC 2.0 TO ESTIMATE PATIENT DOSE ARISING FROM DEXA SCANS

FEASIBILITY STUDY OF USING PCXMC 2.0 TO ESTIMATE PATIENT DOSE ARISING FROM DEXA SCANS Abstract Patients undergoing dual energy X-ray absorption (DEXA) scans are exposed to small doses of ionizing radiation. Few papers have been published on the effective dose and organ dose for patients undergoing such scans on newer DEXA scanners. PCXMC 2.0 was used to calculate adult patient dose arising from DEXA scans. PCXMC 2.0 calculations were compared to published data and a discrepancy was noted for organ dose. Following this, effective and organ dose were measured on an anthropomorphic phantom using TLDs as a second comparison. The mean dose from 50 scans (minus background radiation) was measured. The dose calculated from PCXMC 2.0 compared to published data shows very good agreement for effective dose but a difference for organ dose. Our TLD data and PCXMC 2.0 data for organ dose have a closer agreement, within 20%. We are confident in using PCXMC 2.0 to calculate adult patient effective and organ dose arising from DEXA scans. INTRODUCTION Patients undergoing dual energy X-ray absorption (DEXA) scans are exposed to small doses of ionizing radiation. Early DEXA scanners focused mainly on lumbar spine and femur scans and the literature available reflects this. Recent developments in DEXA scanners mean that more extensive anatomic areas can be scanned. Few papers have been published on the effective dose (ED) and organ dose for adult patients undergoing such scans on newer DEXA scanners. Blake et al.(1) forms the basis of recent discussion regarding adult patient ED from DEXA scans. The study performed depth dose measurements using thermoluminescent dosemeters (TLDs) and a Rando phantom which were then mapped onto the Cristy mathematical phantoms for pediatrics and adults. ED was calculated using ICRP 60(2) tissue weighting factors. One of the highest organ doses recorded was to the stomach (2.6 μSv). Bandirali et al.(3) measured the ED on an anthropomorphic phantom undergoing lumbar and femoral DEXA scans, using three different scan modalities (fast, array and high-definition). A Rando phantom and TLD method was used. The most up-to-date ICRP 103(4) tissue weighting factors to calculate ED were used. The highest organ doses recorded was to the stomach (5.6 μSv). This is approximately double what Blake et al. recorded. Damilakis et al.(5) looked at the radiation exposure in X-ray based imaging used in osteoporosis; both CT and DEXA. The data used for the DEXA ED was taken from the study by Blake et al.(1). This study was done in 2005 and used a scanner which represented the latest capabilities at the time, but is now deemed outdated. This study also used older ICRP 60 weighting factors in the paper. There are two different methods to estimating organ dose; a direct measurement in an anthropomorphic phantom and using computational methods. An investigation using an anthropomorphic phantom and TLDs is time consuming as each scan uses a small amount of radiation, consequently multiple scans would be required to give a reasonable amount of dose to the TLDs. Additionally, the DEXA scanners are used daily and it would impact clinical service to take a scanner out of use for the purpose of a dose estimate. PCXMC 2.0(6) is a piece of software used to calculate dose in a variety of circumstances, e.g. skin dose from a fluoroscopic procedure. Anatomical data are based on the mathematical hermaphrodite phantom models of Cristy and Eckerman(7). These were developed based on the reference masses and organ definitions given in ICRP 89(8). Reasonable agreement of PCXMC 2.0 results have also been found in many comparisons with other dose calculations and phantom models(9). The main aim of this study was to compare the computed organ dose for adult patients undergoing DEXA scans with previously published directly measured organ dose. To achieve this aim, PCXMC 2.0 was used for the calculation. Adult ED and organ doses were calculated for both a Hologic Horizon A and Hologic Discovery A DEXA scanner using PCXMC 2.0 allowing comparison to Bandirali et al. The Hologic Horizon A and Discovery A scanners can perform a range of scans, including single and dual energy modes and be used in the anterior posterior (AP) and lateral position. The typical geometry parameters for an AP lumbar spine are shown in Table 1 and these were used in this investigation. Table 1. AP lumbar spine scan mode geometry parameters for Hologic Horizon A and Discovery A DEXA scanners and PCXMC 2.0 geometry. Mode  Scan length Horizon A (cm)  Scan length Discovery A (cm)  Scan width Horizon A and Discovery A (cm)  Collimator Horizon A and Discovery A (cm)  Number of PCXMC 2.0 slice  PCXMC 2.0 slice width (cm)  Express  20.6  20.6  11.4  4.73 × 0.16  70  0.03  Fast  20.3  20.6  11.4  4.73 × 0.08  137  0.15  Array  20.3  20.6  11.4  4.73 × 0.08  137  0.15  HD  20.3  20.6  11.4  4.73 × 0.04  273  0.075  Mode  Scan length Horizon A (cm)  Scan length Discovery A (cm)  Scan width Horizon A and Discovery A (cm)  Collimator Horizon A and Discovery A (cm)  Number of PCXMC 2.0 slice  PCXMC 2.0 slice width (cm)  Express  20.6  20.6  11.4  4.73 × 0.16  70  0.03  Fast  20.3  20.6  11.4  4.73 × 0.08  137  0.15  Array  20.3  20.6  11.4  4.73 × 0.08  137  0.15  HD  20.3  20.6  11.4  4.73 × 0.04  273  0.075  Table 1. AP lumbar spine scan mode geometry parameters for Hologic Horizon A and Discovery A DEXA scanners and PCXMC 2.0 geometry. Mode  Scan length Horizon A (cm)  Scan length Discovery A (cm)  Scan width Horizon A and Discovery A (cm)  Collimator Horizon A and Discovery A (cm)  Number of PCXMC 2.0 slice  PCXMC 2.0 slice width (cm)  Express  20.6  20.6  11.4  4.73 × 0.16  70  0.03  Fast  20.3  20.6  11.4  4.73 × 0.08  137  0.15  Array  20.3  20.6  11.4  4.73 × 0.08  137  0.15  HD  20.3  20.6  11.4  4.73 × 0.04  273  0.075  Mode  Scan length Horizon A (cm)  Scan length Discovery A (cm)  Scan width Horizon A and Discovery A (cm)  Collimator Horizon A and Discovery A (cm)  Number of PCXMC 2.0 slice  PCXMC 2.0 slice width (cm)  Express  20.6  20.6  11.4  4.73 × 0.16  70  0.03  Fast  20.3  20.6  11.4  4.73 × 0.08  137  0.15  Array  20.3  20.6  11.4  4.73 × 0.08  137  0.15  HD  20.3  20.6  11.4  4.73 × 0.04  273  0.075  The stomach region was chosen as this was identified as a high organ dose from Bandirali et al. If the investigation is successful we may extend using the PCXMC 2.0 method to other DEXA scanners and scan modes in the future. Partway through the investigation it became apparent that a comparison to dose measured using TLDs was necessary for a more comprehensive investigation. METHODS AND MATERIALS PCXMC 2.0 PCXMC 2.0 is a Monte Carlo simulation program for calculating patients’ organ doses and EDs in medical X-ray examinations. It can calculate the ED using both the present tissue weighting factors of ICRP 103 and the former tissue weighting factors of ICRP 60. There are several changes to the tissue weighting factors from ICRP 60 to ICRP 103; however, the area of interest in this investigation (stomach) has not changed. We used the software to simulate a DEXA scan by breaking it down into lots of small fields (slices) that correspond to each scan line (Table 1). This provides better matching of beam divergence and scatter than one big field. PCXMC 2.0 requires information on irradiation geometry, energy and quantity in order to calculate organ and EDs. The scan type AP lumbar spine and its four available scan modes; express, fast, array and HD were modeled on PCXMC 2.0. Standard patient size for PCXMC 2.0 uses a phantom height of 178.6 cm and weight of 73.2 kg. The Z reference (starting point) for PCXMC 2.0 for lower lumbar spine was estimated on each simulation. A Z reference point of 19 was used, with an end Z point of 39. Each slice was added to this starting reference point until the end of each lumbar scan was reached. PCXMC 2.0 calculates the total effective and organ dose received from each slice of the scan. This is combined at the end to give a total effective and organ dose for the whole scan. DEXA scanners The DEXA scanners used for this investigation were a Hologic Horizon A and a Hologic Discovery A. With the information available from the manufacturer, a target angle of 10° and 10% kV ripple was estimated. The Hologic user manual(10) was used to obtain the data for field size, focus to skin distance (FSD), scan length and X-ray beam geometry. The FSD was taken from the manual as the focal spot position is not marked on the tube housing for either scanner. The scan type AP lumbar spine has four different scan mode settings; fast, array, express and HD; only array mode was used during the investigation to allow comparison in a timely manner. An adult; height of 170 cm and weight 80 kg, was used for our test patient. This is the standard adult used for our quality assurance (QA) tests that are performed on DEXA scanners and was chosen for its simplicity. The AP lumbar spine scan is a dual energy scan, using 100 and 140 kVp switching. The Horizon A uses a nominal 0.2 mm Al filter for 100 kV and 0.2 mm Al plus 1.6 mm Brass filter for 140 kV. We were unable to find out the exact components of the 1.6 mm brass filter. Data provided in the manual stated that the 1.6 mm Brass filter was equivalent to 53 mm Al. IPEM 78(11) spectra has been used to estimate the likely error if we assumed the filter was all copper or all zinc or all aluminum. The Discovery A lumbar spine AP scan is also a dual energy scan; using 100 and 140 kVp switching, with a nominal 4.2 mm Al plus 0.08 mm Cu at 100 kV and 6.9 mm Al plus 0.08 mm Cu at 140 kV. The HVL measured for a single energy scan was comparable to nominal values for both scanner models. PCXMC 2.0 uses incident air kerma (IAK); this was measured free in air using a Radcal 9010 with 90 × 5–60 chamber and corrected using the inverse square law (ISL). No back or forward scatter correction was added to the measurements. Our QA measured IAK values were found to be higher than manufacturers stated values for all scan modes. It is unclear if the manufacturer’s stated values include scatter in their measurement. We investigated the scatter at the commissioning for the Horizon A scanner and found that our measured IAK could be ~11% higher than predicted IAK. Owing to this uncertainty we are using dose area product (DAP) values displayed for the scan mode used (Table 2). These are calculated and not directly measured by the scanner. DAP was measured using the IAK and the field size imaged using an Agfa CR cassette. The displayed DAP values were compared to measured DAP values and were found to be different for both types of scanner. Again, we believe this to incorporate scatter and as such are satisfied with using the displayed values. The DAP has been split equally between each slice, although giving a small error due to the higher exposure on first slice as seen on the CR image. We were unable to measure the dose at just this portion of the scan. Table 2. PCXMC 2.0 DAP calculations for AP lumbar spine scan. Mode  Total DAP displayed Horizon A (mGy cm2)  Total DAP displayed Discovery A (mGy cm2)  DAP per slice Horizon A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Horizon A (mGy cm2)a  DAP per slice Discovery A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Discovery A (mGy cm2)a  Express  15  16  0.1243  124.3  0.2286  0228.6  Fast  20  23  0.146  146.0  0.1679  167.9  Array  40  46  0.292  292.0  0.3358  335.8  HD  29  46  0.1062  106.2  0.1685  168.5  Mode  Total DAP displayed Horizon A (mGy cm2)  Total DAP displayed Discovery A (mGy cm2)  DAP per slice Horizon A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Horizon A (mGy cm2)a  DAP per slice Discovery A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Discovery A (mGy cm2)a  Express  15  16  0.1243  124.3  0.2286  0228.6  Fast  20  23  0.146  146.0  0.1679  167.9  Array  40  46  0.292  292.0  0.3358  335.8  HD  29  46  0.1062  106.2  0.1685  168.5  aThe DAP per slice has been multiplied by 1000 to give a usable number in PCXMC 2.0. The final data was divided by 1000 once calculated. Table 2. PCXMC 2.0 DAP calculations for AP lumbar spine scan. Mode  Total DAP displayed Horizon A (mGy cm2)  Total DAP displayed Discovery A (mGy cm2)  DAP per slice Horizon A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Horizon A (mGy cm2)a  DAP per slice Discovery A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Discovery A (mGy cm2)a  Express  15  16  0.1243  124.3  0.2286  0228.6  Fast  20  23  0.146  146.0  0.1679  167.9  Array  40  46  0.292  292.0  0.3358  335.8  HD  29  46  0.1062  106.2  0.1685  168.5  Mode  Total DAP displayed Horizon A (mGy cm2)  Total DAP displayed Discovery A (mGy cm2)  DAP per slice Horizon A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Horizon A (mGy cm2)a  DAP per slice Discovery A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Discovery A (mGy cm2)a  Express  15  16  0.1243  124.3  0.2286  0228.6  Fast  20  23  0.146  146.0  0.1679  167.9  Array  40  46  0.292  292.0  0.3358  335.8  HD  29  46  0.1062  106.2  0.1685  168.5  aThe DAP per slice has been multiplied by 1000 to give a usable number in PCXMC 2.0. The final data was divided by 1000 once calculated. At this point in our investigation it became clear that there was a difference between our PCXMC 2.0 data and published values. We decided to investigate further using an anthropomorphic phantom and TLDs. Rando anthropomorphic phantom The Rando(12) anthropomorphic phantom is constructed with a natural human skeleton that is cast inside a material that is radiologically equivalent to soft tissue. Both adult male and female phantoms are available. This investigation used the adult male phantom which is sliced into 2.5 cm sections totaling 35 slices. Grids of holes can be optionally drilled through the phantom’s soft tissue materials (not bone). Dose measurements can be obtained using individual dosemeters. A variety of commercially available dosemeters can be used in these grids. For this investigation TLDs supplied from Thermo Scientific were used. TLDs The TLD chips used are composed of lithium fluoride doped with magnesium, copper and phosphate at 10–100 ppm with a size of 3.2 mm square and 0.5 mm thick. These are commercially known as TLD100H. They have a linear range of 1 μGy–10 Gy. The TLD-100H material has a near tissue-equivalent property and a very flat energy response(13). One hundred and twenty TLD100H were exposed with 50 scans for the Hologic Horizon A AP lumbar spine scan array mode. They were placed within the Rando phantom in the stomach region spread throughout slices 20–24, with two TLDs per hole. The TLDs, once exposed, were read out using a Harshaw 5500 reader. The resultant data is given in nC, which is multiplied by a calibration factor to give the dose received to each TLD. This is derived by giving a known dose to the calibration TLDs and dividing this dose by the resultant mean data of the TLD. Seven TLDS were kept out of the DXA room to measure background radiation and for calibration. An average was taken of the two TLDs in each hole and then an average overall as the dose to the stomach region. Once annealed the same 120 TLDs were again placed within the Rando phantom in the same holes as before and exposed with 50 scans for the Hologic Discovery A AP lumbar spine scan array mode. Background TLDs were kept outside of the room and a further calibration was performed. RESULTS The ED calculated using PCXMC 2.0 for both the Horizon A and Discovery A DEXA scanners is shown in Table 3. The calculations have been performed using ICRP 103 tissue weighting factors. The ED for 100 and 140 kV has been averaged for each mode and this value has been used to compare against published data. The array mode on the Horizon A gave an average ED of 13.3 μSv and the Discovery A an average of 14.2 μSv. Table 3. Effective dose (ED) from PCXMC 2.0 at 100 and 140 kV. Mode  ED 100 kV Horizon A (μSv)  ED 140 kV Horizon A (μSv)  Averaged ED Horizon A (μSv)  ED 100 kV Discovery A (μSv)  ED 140 kV Discovery A (μSv)  Averaged ED Discovery A (μSv)  PCXMC 2.0 statistical error (%)  Express  3.4  6.7  5.1  4.3  5.6  4.9  0.2  Fast  4.4  8.9  6.7  6.2  8.1  7.1  0.2  Array  8.8  17.8  13.3  12.2  16.1  14.2  0.2  HD  6.4  12.9  9.7  12.3  16.1  14.2  0.2  Mode  ED 100 kV Horizon A (μSv)  ED 140 kV Horizon A (μSv)  Averaged ED Horizon A (μSv)  ED 100 kV Discovery A (μSv)  ED 140 kV Discovery A (μSv)  Averaged ED Discovery A (μSv)  PCXMC 2.0 statistical error (%)  Express  3.4  6.7  5.1  4.3  5.6  4.9  0.2  Fast  4.4  8.9  6.7  6.2  8.1  7.1  0.2  Array  8.8  17.8  13.3  12.2  16.1  14.2  0.2  HD  6.4  12.9  9.7  12.3  16.1  14.2  0.2  Table 3. Effective dose (ED) from PCXMC 2.0 at 100 and 140 kV. Mode  ED 100 kV Horizon A (μSv)  ED 140 kV Horizon A (μSv)  Averaged ED Horizon A (μSv)  ED 100 kV Discovery A (μSv)  ED 140 kV Discovery A (μSv)  Averaged ED Discovery A (μSv)  PCXMC 2.0 statistical error (%)  Express  3.4  6.7  5.1  4.3  5.6  4.9  0.2  Fast  4.4  8.9  6.7  6.2  8.1  7.1  0.2  Array  8.8  17.8  13.3  12.2  16.1  14.2  0.2  HD  6.4  12.9  9.7  12.3  16.1  14.2  0.2  Mode  ED 100 kV Horizon A (μSv)  ED 140 kV Horizon A (μSv)  Averaged ED Horizon A (μSv)  ED 100 kV Discovery A (μSv)  ED 140 kV Discovery A (μSv)  Averaged ED Discovery A (μSv)  PCXMC 2.0 statistical error (%)  Express  3.4  6.7  5.1  4.3  5.6  4.9  0.2  Fast  4.4  8.9  6.7  6.2  8.1  7.1  0.2  Array  8.8  17.8  13.3  12.2  16.1  14.2  0.2  HD  6.4  12.9  9.7  12.3  16.1  14.2  0.2  The stomach dose calculated using PCXMC 2.0 for both the Horizon A and Discovery A DEXA scanners is shown in Table 4. The organ dose for 100 and 140 kV have been averaged for each mode and this value has been used to compare against published data. The array mode on the Horizon A gave an average stomach dose of 14.2 μGy and the Discovery A an average of 20.1 μGy. Table 4. Stomach dose from PCXMC 2.0 at 100 and 140 kV. Mode  Stomach dose Horizon A 100 kV (μGy)  Stomach dose Horizon A 140 kV (μGy)  Averaged stomach Horizon A dose (μGy)  Stomach dose Discovery A 100 kV (μGy)  Stomach dose Discovery A 140 kV (μGy)  Averaged stomach Discovery A dose (μGy)  PCXMC 2.0 statistical error (%)  Express  4.3  5.6  4.9  5.9  8.0  6.9  0.9  Fast  6.2  8.1  7.1  8.4  11.7  10.1  0.9  Array  12.2  16.1  14.2  16.8  23.5  20.1  0.9  HD  12.3  16.1  14.2  16.8  23.5  20.2  0.9  Mode  Stomach dose Horizon A 100 kV (μGy)  Stomach dose Horizon A 140 kV (μGy)  Averaged stomach Horizon A dose (μGy)  Stomach dose Discovery A 100 kV (μGy)  Stomach dose Discovery A 140 kV (μGy)  Averaged stomach Discovery A dose (μGy)  PCXMC 2.0 statistical error (%)  Express  4.3  5.6  4.9  5.9  8.0  6.9  0.9  Fast  6.2  8.1  7.1  8.4  11.7  10.1  0.9  Array  12.2  16.1  14.2  16.8  23.5  20.1  0.9  HD  12.3  16.1  14.2  16.8  23.5  20.2  0.9  Table 4. Stomach dose from PCXMC 2.0 at 100 and 140 kV. Mode  Stomach dose Horizon A 100 kV (μGy)  Stomach dose Horizon A 140 kV (μGy)  Averaged stomach Horizon A dose (μGy)  Stomach dose Discovery A 100 kV (μGy)  Stomach dose Discovery A 140 kV (μGy)  Averaged stomach Discovery A dose (μGy)  PCXMC 2.0 statistical error (%)  Express  4.3  5.6  4.9  5.9  8.0  6.9  0.9  Fast  6.2  8.1  7.1  8.4  11.7  10.1  0.9  Array  12.2  16.1  14.2  16.8  23.5  20.1  0.9  HD  12.3  16.1  14.2  16.8  23.5  20.2  0.9  Mode  Stomach dose Horizon A 100 kV (μGy)  Stomach dose Horizon A 140 kV (μGy)  Averaged stomach Horizon A dose (μGy)  Stomach dose Discovery A 100 kV (μGy)  Stomach dose Discovery A 140 kV (μGy)  Averaged stomach Discovery A dose (μGy)  PCXMC 2.0 statistical error (%)  Express  4.3  5.6  4.9  5.9  8.0  6.9  0.9  Fast  6.2  8.1  7.1  8.4  11.7  10.1  0.9  Array  12.2  16.1  14.2  16.8  23.5  20.1  0.9  HD  12.3  16.1  14.2  16.8  23.5  20.2  0.9  The TLD data for the Hologic Horizon A scanner stomach dose for array mode was 19.5 μGy. For the Discovery A scanner a stomach dose of 20.1 μGy was recorded. A summary of published data is available in Table 5. A comparison of PCXMC 2.0, TLD data and published data can be found in Table 6. Table 5. Summary of published data. Author  Scanner  Scan type  Mode  Method  Equivalent dose to stomach (μGy)  Effective dose (μSv)  Bandirali et al.  Hologic Discovery A  AP lumbar spine  Array  Anthropomorphic phantom and TLDs  5.6  n/a  Blake et al.  Discovery/QDR4500  AP lumbar spine  Array  TLD depth dose, ICRP 60 weighting factors  2.6  13.3a  Damilakis et al.  Discovery/QDR4500  AP lumbar spine  Array  Blake et al. paper  2.6  13.3a  Author  Scanner  Scan type  Mode  Method  Equivalent dose to stomach (μGy)  Effective dose (μSv)  Bandirali et al.  Hologic Discovery A  AP lumbar spine  Array  Anthropomorphic phantom and TLDs  5.6  n/a  Blake et al.  Discovery/QDR4500  AP lumbar spine  Array  TLD depth dose, ICRP 60 weighting factors  2.6  13.3a  Damilakis et al.  Discovery/QDR4500  AP lumbar spine  Array  Blake et al. paper  2.6  13.3a  aGender average. Table 5. Summary of published data. Author  Scanner  Scan type  Mode  Method  Equivalent dose to stomach (μGy)  Effective dose (μSv)  Bandirali et al.  Hologic Discovery A  AP lumbar spine  Array  Anthropomorphic phantom and TLDs  5.6  n/a  Blake et al.  Discovery/QDR4500  AP lumbar spine  Array  TLD depth dose, ICRP 60 weighting factors  2.6  13.3a  Damilakis et al.  Discovery/QDR4500  AP lumbar spine  Array  Blake et al. paper  2.6  13.3a  Author  Scanner  Scan type  Mode  Method  Equivalent dose to stomach (μGy)  Effective dose (μSv)  Bandirali et al.  Hologic Discovery A  AP lumbar spine  Array  Anthropomorphic phantom and TLDs  5.6  n/a  Blake et al.  Discovery/QDR4500  AP lumbar spine  Array  TLD depth dose, ICRP 60 weighting factors  2.6  13.3a  Damilakis et al.  Discovery/QDR4500  AP lumbar spine  Array  Blake et al. paper  2.6  13.3a  aGender average. Table 6. Comparison of all data. Array mode  Bandirali et al.  Blake et al.  Damilakis et al.  PCXMC 2.0 Horizon A  PCXMC 2.0 Discovery A  TLD Horizon A  TLD Discovery A  Effective dose (μSv)  n/a  13.3  13.3  13.3  14.2  n/a  n/a  Stomach dose (μGy)  5.6  2.6  2.6  14.2  20.1  19.5  20.1  Array mode  Bandirali et al.  Blake et al.  Damilakis et al.  PCXMC 2.0 Horizon A  PCXMC 2.0 Discovery A  TLD Horizon A  TLD Discovery A  Effective dose (μSv)  n/a  13.3  13.3  13.3  14.2  n/a  n/a  Stomach dose (μGy)  5.6  2.6  2.6  14.2  20.1  19.5  20.1  Table 6. Comparison of all data. Array mode  Bandirali et al.  Blake et al.  Damilakis et al.  PCXMC 2.0 Horizon A  PCXMC 2.0 Discovery A  TLD Horizon A  TLD Discovery A  Effective dose (μSv)  n/a  13.3  13.3  13.3  14.2  n/a  n/a  Stomach dose (μGy)  5.6  2.6  2.6  14.2  20.1  19.5  20.1  Array mode  Bandirali et al.  Blake et al.  Damilakis et al.  PCXMC 2.0 Horizon A  PCXMC 2.0 Discovery A  TLD Horizon A  TLD Discovery A  Effective dose (μSv)  n/a  13.3  13.3  13.3  14.2  n/a  n/a  Stomach dose (μGy)  5.6  2.6  2.6  14.2  20.1  19.5  20.1  DISCUSSION AND CONCLUSIONS The aim of our investigation was to see if using PCXMC 2.0 to estimate patient dose arising from DEXA scans is feasible. Following the investigation, we are satisfied in using PCXMC 2.0 to calculate adult patient effective and organ dose arising from DEXA scans. Our investigation has given us the confidence to use PCXMC 2.0 as a time saving exercise in producing patient dose estimates. There was good agreement between PCXMC 2.0 and directly measured organ doses. The PCXMC 2.0 stomach dose estimates are within 20% for the Hologic Horizon A scanner and 4% for the Hologic Discovery scanner for the array mode when comparing to TLD dose data. The TLD results show that PCXMC 2.0 underestimates the dose slightly. Our measured organ doses using TLDs and PCXMC 2.0 are closer together than either are to the published doses. A comparison of the PCXMC 2.0 and TLD data with published data shows a large discrepancy for the stomach dose but very good agreement with ED. Bandirali et al. stomach dose was ~60% lower than that calculated with PCXMC 2.0 although they were different scanners. Blake et al. ED was within 1% of that calculated with PCXMC 2.0. PCXMC 2.0 relies upon correct information regarding X-ray tube filters used and unfortunately we were unable to obtain this information. Our assumptions may possibly account for some of this % error. Also, Bandirali et al. used 90 TLDs over the whole body where as we used 120 TLDs in the stomach region. Our lowest TLD readings came from the outer edge of the phantom and the higher readings towards the mid line. Placing a TLD towards the outer edge of the phantom would give a bigger difference than one placed closer to the mid line. We used nominal DAP values from the DEXA scanner to allow comparison for other centers, however DAP calibration factors should be taken into account. Our QA measurements of IAK indicate a higher dose than expected, ~40%, for the Hologic Horizon A compared to the manufacturer specifications. For the Hologic Discovery A our QA measurements of tube output are similar to those specified by the manufacturer. In conclusion given the errors involved with the computation of patient ED and organ doses, PCXMC 2.0 is a reliable piece of software for calculating patient doses arising from DEXA scans. This will save time over scanning a phantom multiple times and reading TLDs. Using PCXMC 2.0 was much faster than using Rando and TLDS. It is time consuming to prepare the TLDs, load them into the correct slices, arrange time on the DEXA scanner and read out the TLDS. Also there is uncertainty with TLDs as demonstrated with the variation between our data and that published by Bandirali et al. Future work include using other DEXA scan modes used clinically for completeness, placing TLDS throughout the phantom to give a comparison of other organ doses arising from DEXA scans and comparing stomach dose data to ICRP 116. REFERENCES 1 Blake, G. M., Naeem, M. and Boutros, M. Comparison of effective dose to children and adults from dual X-ray absorptiometry examinations. Bone  38( 6), 935– 942 ( 2006). Google Scholar CrossRef Search ADS PubMed  2 International Commission on Radiological Protection. Recommendations of the International Commission on Radiological Protection. ICRP Publication 60 (Elsevier Science) ( 1991). 3 Bandirali, M. et al.  . Dose absorption in lumbar and femoral dual energy X-ray absorptiometry examinations using three different scan modalities: an anthropomorphic phantom study. J. Clin. Densitom.  16( 3), 279– 282 ( 2013). Google Scholar CrossRef Search ADS PubMed  4 International Commission on Radiological Protection. The 2007 recommendations of the International Commission on Radiological Protection. ICRP Publication 103 (Elsevier Science) ( 2007). 5 Damilakis, J., Adams, J. E., Guglielmi, G. and Link, T. M. Radiation exposure in X-ray-based imaging techniques used in osteoporosis. Eur. Radiol.  20, 2707– 2714 ( 2010). Google Scholar CrossRef Search ADS PubMed  6 Tapiovaara, M. and Siiskonen, T. PCXMC 2.0 A Monte Carlo Program for Calculating Patient Doses in Medical X-ray Examinations , second edn. Helsinki: Finnish Centre for Radiation and Nuclear Safety) ( 2008) STUK-A231. 7 Stabin, M. G., Emmons, M. A., Segars, W. P. and Fernald, M. J. Realistic reference adult and pediatric phantom series for internal and external dosimetry. Radiat. Prot. Dosim.  149( 1), 56– 59 ( 2012). Google Scholar CrossRef Search ADS   8 International Commission on Radiological Protection. Basic anatomical and physiological data for use in radiological protection: reference values. ICRP Publication 89 (Elsevier Health) ( 2003) 9 Golikov, V., Barkovsky, A., Wallstrōm, E. and Cederblad, Å. A comparative study of organ doses assessment for patients undergoing conventional X-ray examinations: phantom experiments vs. calculations. Radiat. Prot. Dosim.  178( 2), 223– 234 ( 2018). Google Scholar CrossRef Search ADS   10 DEXA manual horizon Hologic Technical Specifications Manual (MAN-03283 Revision 002). 11 Birch, R., Marshall, M. and Ardran, G. M. Catalogue of spectral data for diagnostic X-rays. Institute of physics and Engineering in Medicine (IPEM) ( 1979). 12 Rando Phantom Product Information. Available on https://www.imagingsol.com.au/product/1838/Rando-Phantom.html (accessed November 27, 2017). 13 Luo, L. Z. and Rotunda, J. E. Performance of Harshaw TLD-100H two-element dosemeter. Radiat. Prot. Dosim.  120( 1–4), 324– 330 ( 2006). Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Radiation Protection Dosimetry Oxford University Press

FEASIBILITY STUDY OF USING PCXMC 2.0 TO ESTIMATE PATIENT DOSE ARISING FROM DEXA SCANS

Loading next page...
 
/lp/ou_press/feasibility-study-of-using-pcxmc-2-0-to-estimate-patient-dose-arising-SNVM0o5vCM
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com
ISSN
0144-8420
eISSN
1742-3406
D.O.I.
10.1093/rpd/ncy044
Publisher site
See Article on Publisher Site

Abstract

Abstract Patients undergoing dual energy X-ray absorption (DEXA) scans are exposed to small doses of ionizing radiation. Few papers have been published on the effective dose and organ dose for patients undergoing such scans on newer DEXA scanners. PCXMC 2.0 was used to calculate adult patient dose arising from DEXA scans. PCXMC 2.0 calculations were compared to published data and a discrepancy was noted for organ dose. Following this, effective and organ dose were measured on an anthropomorphic phantom using TLDs as a second comparison. The mean dose from 50 scans (minus background radiation) was measured. The dose calculated from PCXMC 2.0 compared to published data shows very good agreement for effective dose but a difference for organ dose. Our TLD data and PCXMC 2.0 data for organ dose have a closer agreement, within 20%. We are confident in using PCXMC 2.0 to calculate adult patient effective and organ dose arising from DEXA scans. INTRODUCTION Patients undergoing dual energy X-ray absorption (DEXA) scans are exposed to small doses of ionizing radiation. Early DEXA scanners focused mainly on lumbar spine and femur scans and the literature available reflects this. Recent developments in DEXA scanners mean that more extensive anatomic areas can be scanned. Few papers have been published on the effective dose (ED) and organ dose for adult patients undergoing such scans on newer DEXA scanners. Blake et al.(1) forms the basis of recent discussion regarding adult patient ED from DEXA scans. The study performed depth dose measurements using thermoluminescent dosemeters (TLDs) and a Rando phantom which were then mapped onto the Cristy mathematical phantoms for pediatrics and adults. ED was calculated using ICRP 60(2) tissue weighting factors. One of the highest organ doses recorded was to the stomach (2.6 μSv). Bandirali et al.(3) measured the ED on an anthropomorphic phantom undergoing lumbar and femoral DEXA scans, using three different scan modalities (fast, array and high-definition). A Rando phantom and TLD method was used. The most up-to-date ICRP 103(4) tissue weighting factors to calculate ED were used. The highest organ doses recorded was to the stomach (5.6 μSv). This is approximately double what Blake et al. recorded. Damilakis et al.(5) looked at the radiation exposure in X-ray based imaging used in osteoporosis; both CT and DEXA. The data used for the DEXA ED was taken from the study by Blake et al.(1). This study was done in 2005 and used a scanner which represented the latest capabilities at the time, but is now deemed outdated. This study also used older ICRP 60 weighting factors in the paper. There are two different methods to estimating organ dose; a direct measurement in an anthropomorphic phantom and using computational methods. An investigation using an anthropomorphic phantom and TLDs is time consuming as each scan uses a small amount of radiation, consequently multiple scans would be required to give a reasonable amount of dose to the TLDs. Additionally, the DEXA scanners are used daily and it would impact clinical service to take a scanner out of use for the purpose of a dose estimate. PCXMC 2.0(6) is a piece of software used to calculate dose in a variety of circumstances, e.g. skin dose from a fluoroscopic procedure. Anatomical data are based on the mathematical hermaphrodite phantom models of Cristy and Eckerman(7). These were developed based on the reference masses and organ definitions given in ICRP 89(8). Reasonable agreement of PCXMC 2.0 results have also been found in many comparisons with other dose calculations and phantom models(9). The main aim of this study was to compare the computed organ dose for adult patients undergoing DEXA scans with previously published directly measured organ dose. To achieve this aim, PCXMC 2.0 was used for the calculation. Adult ED and organ doses were calculated for both a Hologic Horizon A and Hologic Discovery A DEXA scanner using PCXMC 2.0 allowing comparison to Bandirali et al. The Hologic Horizon A and Discovery A scanners can perform a range of scans, including single and dual energy modes and be used in the anterior posterior (AP) and lateral position. The typical geometry parameters for an AP lumbar spine are shown in Table 1 and these were used in this investigation. Table 1. AP lumbar spine scan mode geometry parameters for Hologic Horizon A and Discovery A DEXA scanners and PCXMC 2.0 geometry. Mode  Scan length Horizon A (cm)  Scan length Discovery A (cm)  Scan width Horizon A and Discovery A (cm)  Collimator Horizon A and Discovery A (cm)  Number of PCXMC 2.0 slice  PCXMC 2.0 slice width (cm)  Express  20.6  20.6  11.4  4.73 × 0.16  70  0.03  Fast  20.3  20.6  11.4  4.73 × 0.08  137  0.15  Array  20.3  20.6  11.4  4.73 × 0.08  137  0.15  HD  20.3  20.6  11.4  4.73 × 0.04  273  0.075  Mode  Scan length Horizon A (cm)  Scan length Discovery A (cm)  Scan width Horizon A and Discovery A (cm)  Collimator Horizon A and Discovery A (cm)  Number of PCXMC 2.0 slice  PCXMC 2.0 slice width (cm)  Express  20.6  20.6  11.4  4.73 × 0.16  70  0.03  Fast  20.3  20.6  11.4  4.73 × 0.08  137  0.15  Array  20.3  20.6  11.4  4.73 × 0.08  137  0.15  HD  20.3  20.6  11.4  4.73 × 0.04  273  0.075  Table 1. AP lumbar spine scan mode geometry parameters for Hologic Horizon A and Discovery A DEXA scanners and PCXMC 2.0 geometry. Mode  Scan length Horizon A (cm)  Scan length Discovery A (cm)  Scan width Horizon A and Discovery A (cm)  Collimator Horizon A and Discovery A (cm)  Number of PCXMC 2.0 slice  PCXMC 2.0 slice width (cm)  Express  20.6  20.6  11.4  4.73 × 0.16  70  0.03  Fast  20.3  20.6  11.4  4.73 × 0.08  137  0.15  Array  20.3  20.6  11.4  4.73 × 0.08  137  0.15  HD  20.3  20.6  11.4  4.73 × 0.04  273  0.075  Mode  Scan length Horizon A (cm)  Scan length Discovery A (cm)  Scan width Horizon A and Discovery A (cm)  Collimator Horizon A and Discovery A (cm)  Number of PCXMC 2.0 slice  PCXMC 2.0 slice width (cm)  Express  20.6  20.6  11.4  4.73 × 0.16  70  0.03  Fast  20.3  20.6  11.4  4.73 × 0.08  137  0.15  Array  20.3  20.6  11.4  4.73 × 0.08  137  0.15  HD  20.3  20.6  11.4  4.73 × 0.04  273  0.075  The stomach region was chosen as this was identified as a high organ dose from Bandirali et al. If the investigation is successful we may extend using the PCXMC 2.0 method to other DEXA scanners and scan modes in the future. Partway through the investigation it became apparent that a comparison to dose measured using TLDs was necessary for a more comprehensive investigation. METHODS AND MATERIALS PCXMC 2.0 PCXMC 2.0 is a Monte Carlo simulation program for calculating patients’ organ doses and EDs in medical X-ray examinations. It can calculate the ED using both the present tissue weighting factors of ICRP 103 and the former tissue weighting factors of ICRP 60. There are several changes to the tissue weighting factors from ICRP 60 to ICRP 103; however, the area of interest in this investigation (stomach) has not changed. We used the software to simulate a DEXA scan by breaking it down into lots of small fields (slices) that correspond to each scan line (Table 1). This provides better matching of beam divergence and scatter than one big field. PCXMC 2.0 requires information on irradiation geometry, energy and quantity in order to calculate organ and EDs. The scan type AP lumbar spine and its four available scan modes; express, fast, array and HD were modeled on PCXMC 2.0. Standard patient size for PCXMC 2.0 uses a phantom height of 178.6 cm and weight of 73.2 kg. The Z reference (starting point) for PCXMC 2.0 for lower lumbar spine was estimated on each simulation. A Z reference point of 19 was used, with an end Z point of 39. Each slice was added to this starting reference point until the end of each lumbar scan was reached. PCXMC 2.0 calculates the total effective and organ dose received from each slice of the scan. This is combined at the end to give a total effective and organ dose for the whole scan. DEXA scanners The DEXA scanners used for this investigation were a Hologic Horizon A and a Hologic Discovery A. With the information available from the manufacturer, a target angle of 10° and 10% kV ripple was estimated. The Hologic user manual(10) was used to obtain the data for field size, focus to skin distance (FSD), scan length and X-ray beam geometry. The FSD was taken from the manual as the focal spot position is not marked on the tube housing for either scanner. The scan type AP lumbar spine has four different scan mode settings; fast, array, express and HD; only array mode was used during the investigation to allow comparison in a timely manner. An adult; height of 170 cm and weight 80 kg, was used for our test patient. This is the standard adult used for our quality assurance (QA) tests that are performed on DEXA scanners and was chosen for its simplicity. The AP lumbar spine scan is a dual energy scan, using 100 and 140 kVp switching. The Horizon A uses a nominal 0.2 mm Al filter for 100 kV and 0.2 mm Al plus 1.6 mm Brass filter for 140 kV. We were unable to find out the exact components of the 1.6 mm brass filter. Data provided in the manual stated that the 1.6 mm Brass filter was equivalent to 53 mm Al. IPEM 78(11) spectra has been used to estimate the likely error if we assumed the filter was all copper or all zinc or all aluminum. The Discovery A lumbar spine AP scan is also a dual energy scan; using 100 and 140 kVp switching, with a nominal 4.2 mm Al plus 0.08 mm Cu at 100 kV and 6.9 mm Al plus 0.08 mm Cu at 140 kV. The HVL measured for a single energy scan was comparable to nominal values for both scanner models. PCXMC 2.0 uses incident air kerma (IAK); this was measured free in air using a Radcal 9010 with 90 × 5–60 chamber and corrected using the inverse square law (ISL). No back or forward scatter correction was added to the measurements. Our QA measured IAK values were found to be higher than manufacturers stated values for all scan modes. It is unclear if the manufacturer’s stated values include scatter in their measurement. We investigated the scatter at the commissioning for the Horizon A scanner and found that our measured IAK could be ~11% higher than predicted IAK. Owing to this uncertainty we are using dose area product (DAP) values displayed for the scan mode used (Table 2). These are calculated and not directly measured by the scanner. DAP was measured using the IAK and the field size imaged using an Agfa CR cassette. The displayed DAP values were compared to measured DAP values and were found to be different for both types of scanner. Again, we believe this to incorporate scatter and as such are satisfied with using the displayed values. The DAP has been split equally between each slice, although giving a small error due to the higher exposure on first slice as seen on the CR image. We were unable to measure the dose at just this portion of the scan. Table 2. PCXMC 2.0 DAP calculations for AP lumbar spine scan. Mode  Total DAP displayed Horizon A (mGy cm2)  Total DAP displayed Discovery A (mGy cm2)  DAP per slice Horizon A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Horizon A (mGy cm2)a  DAP per slice Discovery A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Discovery A (mGy cm2)a  Express  15  16  0.1243  124.3  0.2286  0228.6  Fast  20  23  0.146  146.0  0.1679  167.9  Array  40  46  0.292  292.0  0.3358  335.8  HD  29  46  0.1062  106.2  0.1685  168.5  Mode  Total DAP displayed Horizon A (mGy cm2)  Total DAP displayed Discovery A (mGy cm2)  DAP per slice Horizon A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Horizon A (mGy cm2)a  DAP per slice Discovery A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Discovery A (mGy cm2)a  Express  15  16  0.1243  124.3  0.2286  0228.6  Fast  20  23  0.146  146.0  0.1679  167.9  Array  40  46  0.292  292.0  0.3358  335.8  HD  29  46  0.1062  106.2  0.1685  168.5  aThe DAP per slice has been multiplied by 1000 to give a usable number in PCXMC 2.0. The final data was divided by 1000 once calculated. Table 2. PCXMC 2.0 DAP calculations for AP lumbar spine scan. Mode  Total DAP displayed Horizon A (mGy cm2)  Total DAP displayed Discovery A (mGy cm2)  DAP per slice Horizon A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Horizon A (mGy cm2)a  DAP per slice Discovery A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Discovery A (mGy cm2)a  Express  15  16  0.1243  124.3  0.2286  0228.6  Fast  20  23  0.146  146.0  0.1679  167.9  Array  40  46  0.292  292.0  0.3358  335.8  HD  29  46  0.1062  106.2  0.1685  168.5  Mode  Total DAP displayed Horizon A (mGy cm2)  Total DAP displayed Discovery A (mGy cm2)  DAP per slice Horizon A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Horizon A (mGy cm2)a  DAP per slice Discovery A (mGy cm2)  DAP per slice for entry to PCXMC 2.0 Discovery A (mGy cm2)a  Express  15  16  0.1243  124.3  0.2286  0228.6  Fast  20  23  0.146  146.0  0.1679  167.9  Array  40  46  0.292  292.0  0.3358  335.8  HD  29  46  0.1062  106.2  0.1685  168.5  aThe DAP per slice has been multiplied by 1000 to give a usable number in PCXMC 2.0. The final data was divided by 1000 once calculated. At this point in our investigation it became clear that there was a difference between our PCXMC 2.0 data and published values. We decided to investigate further using an anthropomorphic phantom and TLDs. Rando anthropomorphic phantom The Rando(12) anthropomorphic phantom is constructed with a natural human skeleton that is cast inside a material that is radiologically equivalent to soft tissue. Both adult male and female phantoms are available. This investigation used the adult male phantom which is sliced into 2.5 cm sections totaling 35 slices. Grids of holes can be optionally drilled through the phantom’s soft tissue materials (not bone). Dose measurements can be obtained using individual dosemeters. A variety of commercially available dosemeters can be used in these grids. For this investigation TLDs supplied from Thermo Scientific were used. TLDs The TLD chips used are composed of lithium fluoride doped with magnesium, copper and phosphate at 10–100 ppm with a size of 3.2 mm square and 0.5 mm thick. These are commercially known as TLD100H. They have a linear range of 1 μGy–10 Gy. The TLD-100H material has a near tissue-equivalent property and a very flat energy response(13). One hundred and twenty TLD100H were exposed with 50 scans for the Hologic Horizon A AP lumbar spine scan array mode. They were placed within the Rando phantom in the stomach region spread throughout slices 20–24, with two TLDs per hole. The TLDs, once exposed, were read out using a Harshaw 5500 reader. The resultant data is given in nC, which is multiplied by a calibration factor to give the dose received to each TLD. This is derived by giving a known dose to the calibration TLDs and dividing this dose by the resultant mean data of the TLD. Seven TLDS were kept out of the DXA room to measure background radiation and for calibration. An average was taken of the two TLDs in each hole and then an average overall as the dose to the stomach region. Once annealed the same 120 TLDs were again placed within the Rando phantom in the same holes as before and exposed with 50 scans for the Hologic Discovery A AP lumbar spine scan array mode. Background TLDs were kept outside of the room and a further calibration was performed. RESULTS The ED calculated using PCXMC 2.0 for both the Horizon A and Discovery A DEXA scanners is shown in Table 3. The calculations have been performed using ICRP 103 tissue weighting factors. The ED for 100 and 140 kV has been averaged for each mode and this value has been used to compare against published data. The array mode on the Horizon A gave an average ED of 13.3 μSv and the Discovery A an average of 14.2 μSv. Table 3. Effective dose (ED) from PCXMC 2.0 at 100 and 140 kV. Mode  ED 100 kV Horizon A (μSv)  ED 140 kV Horizon A (μSv)  Averaged ED Horizon A (μSv)  ED 100 kV Discovery A (μSv)  ED 140 kV Discovery A (μSv)  Averaged ED Discovery A (μSv)  PCXMC 2.0 statistical error (%)  Express  3.4  6.7  5.1  4.3  5.6  4.9  0.2  Fast  4.4  8.9  6.7  6.2  8.1  7.1  0.2  Array  8.8  17.8  13.3  12.2  16.1  14.2  0.2  HD  6.4  12.9  9.7  12.3  16.1  14.2  0.2  Mode  ED 100 kV Horizon A (μSv)  ED 140 kV Horizon A (μSv)  Averaged ED Horizon A (μSv)  ED 100 kV Discovery A (μSv)  ED 140 kV Discovery A (μSv)  Averaged ED Discovery A (μSv)  PCXMC 2.0 statistical error (%)  Express  3.4  6.7  5.1  4.3  5.6  4.9  0.2  Fast  4.4  8.9  6.7  6.2  8.1  7.1  0.2  Array  8.8  17.8  13.3  12.2  16.1  14.2  0.2  HD  6.4  12.9  9.7  12.3  16.1  14.2  0.2  Table 3. Effective dose (ED) from PCXMC 2.0 at 100 and 140 kV. Mode  ED 100 kV Horizon A (μSv)  ED 140 kV Horizon A (μSv)  Averaged ED Horizon A (μSv)  ED 100 kV Discovery A (μSv)  ED 140 kV Discovery A (μSv)  Averaged ED Discovery A (μSv)  PCXMC 2.0 statistical error (%)  Express  3.4  6.7  5.1  4.3  5.6  4.9  0.2  Fast  4.4  8.9  6.7  6.2  8.1  7.1  0.2  Array  8.8  17.8  13.3  12.2  16.1  14.2  0.2  HD  6.4  12.9  9.7  12.3  16.1  14.2  0.2  Mode  ED 100 kV Horizon A (μSv)  ED 140 kV Horizon A (μSv)  Averaged ED Horizon A (μSv)  ED 100 kV Discovery A (μSv)  ED 140 kV Discovery A (μSv)  Averaged ED Discovery A (μSv)  PCXMC 2.0 statistical error (%)  Express  3.4  6.7  5.1  4.3  5.6  4.9  0.2  Fast  4.4  8.9  6.7  6.2  8.1  7.1  0.2  Array  8.8  17.8  13.3  12.2  16.1  14.2  0.2  HD  6.4  12.9  9.7  12.3  16.1  14.2  0.2  The stomach dose calculated using PCXMC 2.0 for both the Horizon A and Discovery A DEXA scanners is shown in Table 4. The organ dose for 100 and 140 kV have been averaged for each mode and this value has been used to compare against published data. The array mode on the Horizon A gave an average stomach dose of 14.2 μGy and the Discovery A an average of 20.1 μGy. Table 4. Stomach dose from PCXMC 2.0 at 100 and 140 kV. Mode  Stomach dose Horizon A 100 kV (μGy)  Stomach dose Horizon A 140 kV (μGy)  Averaged stomach Horizon A dose (μGy)  Stomach dose Discovery A 100 kV (μGy)  Stomach dose Discovery A 140 kV (μGy)  Averaged stomach Discovery A dose (μGy)  PCXMC 2.0 statistical error (%)  Express  4.3  5.6  4.9  5.9  8.0  6.9  0.9  Fast  6.2  8.1  7.1  8.4  11.7  10.1  0.9  Array  12.2  16.1  14.2  16.8  23.5  20.1  0.9  HD  12.3  16.1  14.2  16.8  23.5  20.2  0.9  Mode  Stomach dose Horizon A 100 kV (μGy)  Stomach dose Horizon A 140 kV (μGy)  Averaged stomach Horizon A dose (μGy)  Stomach dose Discovery A 100 kV (μGy)  Stomach dose Discovery A 140 kV (μGy)  Averaged stomach Discovery A dose (μGy)  PCXMC 2.0 statistical error (%)  Express  4.3  5.6  4.9  5.9  8.0  6.9  0.9  Fast  6.2  8.1  7.1  8.4  11.7  10.1  0.9  Array  12.2  16.1  14.2  16.8  23.5  20.1  0.9  HD  12.3  16.1  14.2  16.8  23.5  20.2  0.9  Table 4. Stomach dose from PCXMC 2.0 at 100 and 140 kV. Mode  Stomach dose Horizon A 100 kV (μGy)  Stomach dose Horizon A 140 kV (μGy)  Averaged stomach Horizon A dose (μGy)  Stomach dose Discovery A 100 kV (μGy)  Stomach dose Discovery A 140 kV (μGy)  Averaged stomach Discovery A dose (μGy)  PCXMC 2.0 statistical error (%)  Express  4.3  5.6  4.9  5.9  8.0  6.9  0.9  Fast  6.2  8.1  7.1  8.4  11.7  10.1  0.9  Array  12.2  16.1  14.2  16.8  23.5  20.1  0.9  HD  12.3  16.1  14.2  16.8  23.5  20.2  0.9  Mode  Stomach dose Horizon A 100 kV (μGy)  Stomach dose Horizon A 140 kV (μGy)  Averaged stomach Horizon A dose (μGy)  Stomach dose Discovery A 100 kV (μGy)  Stomach dose Discovery A 140 kV (μGy)  Averaged stomach Discovery A dose (μGy)  PCXMC 2.0 statistical error (%)  Express  4.3  5.6  4.9  5.9  8.0  6.9  0.9  Fast  6.2  8.1  7.1  8.4  11.7  10.1  0.9  Array  12.2  16.1  14.2  16.8  23.5  20.1  0.9  HD  12.3  16.1  14.2  16.8  23.5  20.2  0.9  The TLD data for the Hologic Horizon A scanner stomach dose for array mode was 19.5 μGy. For the Discovery A scanner a stomach dose of 20.1 μGy was recorded. A summary of published data is available in Table 5. A comparison of PCXMC 2.0, TLD data and published data can be found in Table 6. Table 5. Summary of published data. Author  Scanner  Scan type  Mode  Method  Equivalent dose to stomach (μGy)  Effective dose (μSv)  Bandirali et al.  Hologic Discovery A  AP lumbar spine  Array  Anthropomorphic phantom and TLDs  5.6  n/a  Blake et al.  Discovery/QDR4500  AP lumbar spine  Array  TLD depth dose, ICRP 60 weighting factors  2.6  13.3a  Damilakis et al.  Discovery/QDR4500  AP lumbar spine  Array  Blake et al. paper  2.6  13.3a  Author  Scanner  Scan type  Mode  Method  Equivalent dose to stomach (μGy)  Effective dose (μSv)  Bandirali et al.  Hologic Discovery A  AP lumbar spine  Array  Anthropomorphic phantom and TLDs  5.6  n/a  Blake et al.  Discovery/QDR4500  AP lumbar spine  Array  TLD depth dose, ICRP 60 weighting factors  2.6  13.3a  Damilakis et al.  Discovery/QDR4500  AP lumbar spine  Array  Blake et al. paper  2.6  13.3a  aGender average. Table 5. Summary of published data. Author  Scanner  Scan type  Mode  Method  Equivalent dose to stomach (μGy)  Effective dose (μSv)  Bandirali et al.  Hologic Discovery A  AP lumbar spine  Array  Anthropomorphic phantom and TLDs  5.6  n/a  Blake et al.  Discovery/QDR4500  AP lumbar spine  Array  TLD depth dose, ICRP 60 weighting factors  2.6  13.3a  Damilakis et al.  Discovery/QDR4500  AP lumbar spine  Array  Blake et al. paper  2.6  13.3a  Author  Scanner  Scan type  Mode  Method  Equivalent dose to stomach (μGy)  Effective dose (μSv)  Bandirali et al.  Hologic Discovery A  AP lumbar spine  Array  Anthropomorphic phantom and TLDs  5.6  n/a  Blake et al.  Discovery/QDR4500  AP lumbar spine  Array  TLD depth dose, ICRP 60 weighting factors  2.6  13.3a  Damilakis et al.  Discovery/QDR4500  AP lumbar spine  Array  Blake et al. paper  2.6  13.3a  aGender average. Table 6. Comparison of all data. Array mode  Bandirali et al.  Blake et al.  Damilakis et al.  PCXMC 2.0 Horizon A  PCXMC 2.0 Discovery A  TLD Horizon A  TLD Discovery A  Effective dose (μSv)  n/a  13.3  13.3  13.3  14.2  n/a  n/a  Stomach dose (μGy)  5.6  2.6  2.6  14.2  20.1  19.5  20.1  Array mode  Bandirali et al.  Blake et al.  Damilakis et al.  PCXMC 2.0 Horizon A  PCXMC 2.0 Discovery A  TLD Horizon A  TLD Discovery A  Effective dose (μSv)  n/a  13.3  13.3  13.3  14.2  n/a  n/a  Stomach dose (μGy)  5.6  2.6  2.6  14.2  20.1  19.5  20.1  Table 6. Comparison of all data. Array mode  Bandirali et al.  Blake et al.  Damilakis et al.  PCXMC 2.0 Horizon A  PCXMC 2.0 Discovery A  TLD Horizon A  TLD Discovery A  Effective dose (μSv)  n/a  13.3  13.3  13.3  14.2  n/a  n/a  Stomach dose (μGy)  5.6  2.6  2.6  14.2  20.1  19.5  20.1  Array mode  Bandirali et al.  Blake et al.  Damilakis et al.  PCXMC 2.0 Horizon A  PCXMC 2.0 Discovery A  TLD Horizon A  TLD Discovery A  Effective dose (μSv)  n/a  13.3  13.3  13.3  14.2  n/a  n/a  Stomach dose (μGy)  5.6  2.6  2.6  14.2  20.1  19.5  20.1  DISCUSSION AND CONCLUSIONS The aim of our investigation was to see if using PCXMC 2.0 to estimate patient dose arising from DEXA scans is feasible. Following the investigation, we are satisfied in using PCXMC 2.0 to calculate adult patient effective and organ dose arising from DEXA scans. Our investigation has given us the confidence to use PCXMC 2.0 as a time saving exercise in producing patient dose estimates. There was good agreement between PCXMC 2.0 and directly measured organ doses. The PCXMC 2.0 stomach dose estimates are within 20% for the Hologic Horizon A scanner and 4% for the Hologic Discovery scanner for the array mode when comparing to TLD dose data. The TLD results show that PCXMC 2.0 underestimates the dose slightly. Our measured organ doses using TLDs and PCXMC 2.0 are closer together than either are to the published doses. A comparison of the PCXMC 2.0 and TLD data with published data shows a large discrepancy for the stomach dose but very good agreement with ED. Bandirali et al. stomach dose was ~60% lower than that calculated with PCXMC 2.0 although they were different scanners. Blake et al. ED was within 1% of that calculated with PCXMC 2.0. PCXMC 2.0 relies upon correct information regarding X-ray tube filters used and unfortunately we were unable to obtain this information. Our assumptions may possibly account for some of this % error. Also, Bandirali et al. used 90 TLDs over the whole body where as we used 120 TLDs in the stomach region. Our lowest TLD readings came from the outer edge of the phantom and the higher readings towards the mid line. Placing a TLD towards the outer edge of the phantom would give a bigger difference than one placed closer to the mid line. We used nominal DAP values from the DEXA scanner to allow comparison for other centers, however DAP calibration factors should be taken into account. Our QA measurements of IAK indicate a higher dose than expected, ~40%, for the Hologic Horizon A compared to the manufacturer specifications. For the Hologic Discovery A our QA measurements of tube output are similar to those specified by the manufacturer. In conclusion given the errors involved with the computation of patient ED and organ doses, PCXMC 2.0 is a reliable piece of software for calculating patient doses arising from DEXA scans. This will save time over scanning a phantom multiple times and reading TLDs. Using PCXMC 2.0 was much faster than using Rando and TLDS. It is time consuming to prepare the TLDs, load them into the correct slices, arrange time on the DEXA scanner and read out the TLDS. Also there is uncertainty with TLDs as demonstrated with the variation between our data and that published by Bandirali et al. Future work include using other DEXA scan modes used clinically for completeness, placing TLDS throughout the phantom to give a comparison of other organ doses arising from DEXA scans and comparing stomach dose data to ICRP 116. REFERENCES 1 Blake, G. M., Naeem, M. and Boutros, M. Comparison of effective dose to children and adults from dual X-ray absorptiometry examinations. Bone  38( 6), 935– 942 ( 2006). Google Scholar CrossRef Search ADS PubMed  2 International Commission on Radiological Protection. Recommendations of the International Commission on Radiological Protection. ICRP Publication 60 (Elsevier Science) ( 1991). 3 Bandirali, M. et al.  . Dose absorption in lumbar and femoral dual energy X-ray absorptiometry examinations using three different scan modalities: an anthropomorphic phantom study. J. Clin. Densitom.  16( 3), 279– 282 ( 2013). Google Scholar CrossRef Search ADS PubMed  4 International Commission on Radiological Protection. The 2007 recommendations of the International Commission on Radiological Protection. ICRP Publication 103 (Elsevier Science) ( 2007). 5 Damilakis, J., Adams, J. E., Guglielmi, G. and Link, T. M. Radiation exposure in X-ray-based imaging techniques used in osteoporosis. Eur. Radiol.  20, 2707– 2714 ( 2010). Google Scholar CrossRef Search ADS PubMed  6 Tapiovaara, M. and Siiskonen, T. PCXMC 2.0 A Monte Carlo Program for Calculating Patient Doses in Medical X-ray Examinations , second edn. Helsinki: Finnish Centre for Radiation and Nuclear Safety) ( 2008) STUK-A231. 7 Stabin, M. G., Emmons, M. A., Segars, W. P. and Fernald, M. J. Realistic reference adult and pediatric phantom series for internal and external dosimetry. Radiat. Prot. Dosim.  149( 1), 56– 59 ( 2012). Google Scholar CrossRef Search ADS   8 International Commission on Radiological Protection. Basic anatomical and physiological data for use in radiological protection: reference values. ICRP Publication 89 (Elsevier Health) ( 2003) 9 Golikov, V., Barkovsky, A., Wallstrōm, E. and Cederblad, Å. A comparative study of organ doses assessment for patients undergoing conventional X-ray examinations: phantom experiments vs. calculations. Radiat. Prot. Dosim.  178( 2), 223– 234 ( 2018). Google Scholar CrossRef Search ADS   10 DEXA manual horizon Hologic Technical Specifications Manual (MAN-03283 Revision 002). 11 Birch, R., Marshall, M. and Ardran, G. M. Catalogue of spectral data for diagnostic X-rays. Institute of physics and Engineering in Medicine (IPEM) ( 1979). 12 Rando Phantom Product Information. Available on https://www.imagingsol.com.au/product/1838/Rando-Phantom.html (accessed November 27, 2017). 13 Luo, L. Z. and Rotunda, J. E. Performance of Harshaw TLD-100H two-element dosemeter. Radiat. Prot. Dosim.  120( 1–4), 324– 330 ( 2006). Google Scholar CrossRef Search ADS   © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Radiation Protection DosimetryOxford University Press

Published: Mar 16, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off