ORIGINAL ARTICLE Year : 2011  Volume : 36  Issue : 4  Page : 213219 The effects of cutouts on output, mean energy and percentage depth dose of 12 and 14 MeV electrons Navid Khaledy^{1}, Azim Arbabi^{2}, Dariush Sardari^{3}, ^{1} Young Researchers Club, Science and Research Branch, Islamic Azad University, Tehran; Department of Medical Physics, Imam Hosein Hospital, Shahid Beheshti Medical University, Tehran, Iran ^{2} Department of Medical Physics, Imam Hosein Hospital, Shahid Beheshti Medical University, Tehran, Iran ^{3} Department of Medical Radiation, Islamic Azad University, Science and Research Branch, Tehran, Iran Correspondence Address: Electron fieldshaping cerrobend cutouts on the linear accelerator applicator have some effects on the output and percentage depth dose. These effects which arise from the lateral scatter nonequilibrium are particularly evident in higher energies and in cutouts with smaller radius. Dose measurements for circular, square, and triangular cutouts as well as open field was performed in a 10 × 10 cm applicator, using plane parallel type ion chamber with a 100 cm source surface distance. The Percentage Depth Doses curves were drawn and the outputs were measured for each of these cutouts. The output factors, normalized to open 10 × 10 cm field, varied between 0.891 and 0.996 depending on the energy, cutout shape, and cavity area. With the use of cutouts, R _{100} shifted toward the surface. The shifts ranged from 9 to 0 mm and from 13 to 0 mm for 12 and 14 MeV, respectively, depending on the shape and cavity area. For R _{90} , R _{80} , and R _{50} the ranges for observed shifts narrowed down and practically no shifts were observed for R _{20} . We present these changes in the form of predictive formulas, which would be useful in clinical applications.
Introduction In treatment of cancers using electron beams, depending on the type and the shape of tumor, cast cerrobend alloy cutouts are used. Cerrobend is a fusible alloy usually containing of Tin, Lead, Bismuth, and Cadmium. We used the cadmiumfree type. The cutouts change the percentage depth dose (PDD) and the output, [1],[2],[3] in a way that the slope of descending part of PDD curve increases. Therefore, the values for R 100 , R 90 , R 80 , and R 50 in order show the highest decrease. Also E 0(mean) and OPF decrease while the cutout cavity area decreases. [4],[5] These changes in higher energy and smaller diameter cutouts are more profound, which is arising from the lack of lateral scatter equilibrium. R eq (minimum radius for the establishment of lateral scatter equilibrium) is calculated using the following equations: [6] [INLINE:1] In this formula, E p0 is the most probable energy at the phantom surface for the reference field size: [7],[8] [INLINE:2] In radius smaller than R eq values of R 100 , R 90 , R 80 , R 50 , E 0(mean) , and OPF are expected to change. [2] The main aim of this study was to fit the values of R 100 , R 90 , R 80 , R 50 , E 0(mean) , and OPF changes on a curve and formulizing these changes as a function of cutouts cavity area. There in this way there will be no need to carry on dosimetry for each patient individually. [2],[9] Materials and Methods Dosimetry system We used PTW (PTW Freiburg, Germany) hardware and software. The water phantom used had a 50 × 50 × 50 cm size and the model was MP3. A plane paralleltype chamber was used for absolute and relative dosimetry, and a thimbletype chamber was used as a reference chamber and was placed outside of the phantom near the radiation field corner. The plane parallel chamber model was PTW 34045 and its sensitive volume was 0.02 cm 3 . The cylindrical chamber model was PTW 31010. Controller software and collected data compiler was the MEPHYSTO® mc2. The IAEA TRS398 dosimetry protocol was used in this study. [10] Dosimetry Dosimetry was performed in 100 cm source surface distance (SSD) and using a 10 × 10 cm applicator. Dosimetry setup used for our work is shown in [Figure 1]. The PDD curve and output for the square, circle, and triangle cutouts and the 10 × 10 cm reference field was measured. The opening of Xray jaws was 19 × 19 cm. Linear accelerator used was a Siemens Primus model and the results were for the 12 and 14 MeV energies. The circle cutout diameters were 3, 4, 5, 6, and 8 cm. Squares had side lengths of 2.66, 3.54, 4.43, 5.32, and 7.1 cm. Also the equilateral triangles had 4.05, 5.4, 6.75, 8.1, and 10.58 cm side lengths. The cavity areas from the smallest cutout to the largest one for each shape were 7.06, 12.56, 19.62, 28.26, and 50.24 cm 2 , respectively.{Figure 1} To obtain PDD both chambers were on. The obtained curves were the percentage depth ionization (PDI) and then converted to the PDD by MEPHYSTO® mc2 software. After the percentage depth dose curves were drawn, the information were extracted as R 100 , R 90, R 80 , R 50 , R 20 , R p , and E 0(mean) . The mean energy of the electron field in the phantom surface was calculated by Eq. (3). [7],[8] [INLINE:3] For measuring the output, the plane parallel chamber was placed at the depth of maximum dose. The maximum depth dose was calculated for each cutout using the PDD curve, because the cutouts' d max are different to a 10 × 10 cm field. We measured the output after turning off the thimble chamber. The measurement was done four times for each field and the average was compared with the reference field (10 × 10 cm) output. Results Percentage depth dose curves Percentage depth dose curve on the central axis for 12 MeV energy We calculated R p using PDD curve, which was equal to 5.7 cm and using the Eq. (2) the E p0 was 11.53 MeV. Also the minimum radius required for lateral scatter equilibrium in 12 MeV electron energy by using Eq. (1) was about 3 cm. Meaning the minimum circular field diameter necessary for 12 MeV electron lateral scatter equilibrium establishment is about 6 cm, which means that changes in PDD for circular field with diameter of less than 6 cm are expected. [2] As is clear from [Table 1], shields cause more change in higher depth doses, especially R 100 , R 90 , and R 80 , and it has a little effect (up to 1 mm) on the R 20 , which is negligible. Triangular shields changed R 100 , R 90 , R 80 , and R 50 more than the circular and square shields. The reduction in R 100 in the triangle shield was from 27 to 18 mm, but for the circular and square shield it was from 27 to 19 mm. For circular and square fields larger than 28.26 cm 2 , due to the establishment of electron lateral scattering equilibrium effect there is no difference between the values of R 100 , R 90 , R 80 , and R 50 compared to reference field. Also, the mean energy of the electron field in smaller fields is lower. By fitting these values on the curve, we devised Eq. (4) (8) formulas to calculate the parameter change based on cutouts area.{Table 1} Dynamic fit curve was found to be the best method for matching abovementioned values on a proper curve and attaining its computed formula. With formulization of these changes for R 100 , Eq. (4) was obtained: [INLINE:4] This formula is a Polynomial type, where S is the area of the shield cavity. Other methods, such as A/P [11] in which A is the area of the cavity and P is the periphery of the cutout hole, were not utilized due to the lack of favorable results in formulization. For R 90 , R 80 , and R 50 these equations were calculated: [INLINE:5] In the formula for E 0(mean), the fitted curve was a Sigmoidal (Gompertz) type. Percentage depth dose curve on the central axis for 14 MeV energy Here the R p was equal to 6.2 cm and the E p0 was 12.66 MeV. As a result R eq = 3.2 cm. PDD data for 14 MeV is demonstrated in [Table 2].{Table 2} In this energy the highest change was observed in R 100 . Here again the decrease on R 100 , R 90 , R 80 , and R 50 by the triangular shields is rather than circular and square shields. The R 100 for the circular shape decreased from 30 to 20 mm in comparison with the reference field. For the square cutout it changed from 30 to 19 mm and for triangle cutout it changed from 30 to 17 mm. The R 20 for the 7.06 cm 2 area for all fields reduced from 59 to 57 mm. In the largest area (50.24 cm 2 ) for 14 MeV in the circular and triangle fields the change of R 100 was 1 mm and for the square cutout it remained unchanged. For 14 MeV similar to 12 MeV the mean energy for smaller fields was lower than the larger fields. The maximum reduction in mean energy was for the triangle shape, decreasing from 12.19 to 10.71 MeV. [Figure 2],[Figure 3],[Figure 4] are obtained by fitting the Mean values of [Table 1] and [Table 2] on the curves. In these figures, the error bars represent plus and minus standard deviations for each point. Some points have same value in several measurements, for these points the standard deviations are zero and no error bar exists. [INLINE:6] [INLINE:7]{Figure 2}{Figure 3}{Figure 4} Output factor Output factor in 12 MeV energy Dose rates in Gy/min were measured at the depth of maximum for all fields and then calculated as a ratio of the reference dose rate value to obtain the OPF for each shield. The dose rates were measured 4 times for each field. Averaging the output values in [Table 3] can provide a more general formula for all forms: [INLINE:8]{Table 3} As the cutout area becomes smaller the dose rate and consequently the OPF for the circle, square, and triangle are reduced from 0.993 to 0.907, from 0.995 to 0.895, and from 0.992 to 0.891, respectively. The lowest output was observed for the triangle. Eq. (14) is a more general formula and can be used for all shapes. Output factor in 14 MeV energy Here again, the average of OPF values for [Table 4] can be used to make a more general formula for all forms in 14 MeV: [INLINE:9]{Table 4} Eqs (14) and (15) are the formulas for the [Figure 5] curves, which were obtained from [Table 3] and [Table 4] data. The curves in [Figure 5] are very close together; hence the error bars have had overlap, so the error bars are omitted. However, the standard deviations are available in [Table 3] and [Table 4].{Figure 5} Discussion The chart of changes for R 100 , R 90 , R 80 , and R 50 shows that while increasing the shield cavity area, after a certain area the values of these parameters reach a saturation status. This means that there is a threshold for the field size influence on R 100 , R 90 , R 80 and R 50. This threshold is close to the R eq . [2],[6] Su et al. measured the cutout OPF for 6 MeV electron beam by radiographic film and 2D small volume ion chamber array. The aim of their study was not about the cutout effects on the output, but in their research this effect can also be observed. [1] Xu et al. have demonstrated that for 6 MeV beam and circular cutout, reducing the diameter reduces the amount of R 100 , R 50 , and OPF. [2] Besides, Sharma et al. observed similar changes in R 100 , R 90 , and R 50 for quadrangular cutouts and 6 MeV energy, due to the changes of R 100 , they propose using of bolus to decrease the skin dose arisen by the reduction of d max . [12] Zhang et al. found that the OPF in 6, 12, 20, 30, and 40 MeV for circular cutouts is reduced when the field radius is smaller than the minimum needed radius for the establishment of lateral scatter equilibrium. [13] Sharma et al. instead of R eq did their measurements for several cutouts with a radius less or more than R p . For 6 MeV electron beam the difference between the amount of R p and R eq is about 9 mm, however, in their study the exact threshold of change had not been investigated. [12] Other researchers got the same result for other energies in different methods. [3],[4],[9],[11] Our findings are in line with the mentioned studies except that these studies have been limited to a particular shape (circular or square) and mostly for one energy level, and none of them have formulized the changes caused by using different cutouts. In 12 MeV and 50.24 cm 2 area for circular, square, and triangle shapes the values of R 100 , R 90 , R 80 , R 50 , and R 20 are equal to the values of the 10 × 10 cm field. The most changes in PDD parameters was observed for the triangle shape, so that R 100 and R 90 for 7.06 cm 2 area were changed 1 mm more in comparison with circle and square shapes. In 14 MeV, the minimum radius for establishment of lateral scatter equilibrium was increased. So, in the 50.24 cm 2 area, which had the largest radius, the R 100 for both the circular and the triangle shapes was different to the 10 × 10 cm field. For the triangle the R 90 , R 80 , and R 50 values were different from the 10 × 10 cm field. Here again the most changes in the values of R 100 , R 90 , R 80 , R 50 , and R 20 were observed with the triangle. E 0(mean) will decrease, if the cutout size is reducing. Whereas E 0(mean) is related to the R 50 , the reduction of R 50 would cause reduction in E 0(mean) as well. The minimum value of E 0(mean) in 12 and 14 MeV was achieved by using 7.06 cm 2 square and 7.06 cm 2 triangle cutouts, respectively, which was equal to 10.05 MeV for 12 MeV and 10.71 MeV for 14 MeV. The reason why the output and percentage depth dose for triangle are different to circular and square shapes is arising from the nature of the triangle which has smaller width of field in the corners in comparison to square and circle fields, which prevents the establishment of lateral scatter equilibrium more than circular and square shapes. Conclusions It is clear that if we use smaller cutouts, in higher depth doses (R 100 , R 90 , R 80 , and R 50 ), more reduction of depth would be seen. [2],[12] However, this reduction is much more tangible in the depth of R 100 and R 90. Considering the d max changes due to the small cutouts, it is obvious that the chamber should be placed at the correct d max in output measurements. [12] The output of the machine which is one of the most important factors in treatment, decreases with reducing the diameter of cutout. [1],[2],[9],[12] This reduction is more pronounced for triangles rather than circle or square. Also, the changes in PDD and outputs in clinical cases must be considered. For example, to compensate the changes of d max , the bolus might be used [12] and output changes should be considered in the treatment calculation. The formulas presented in this study, are clinically useful to calculate the changes of PDD and OPF for each regular and semiregular geometry cutouts on the central axis. Knowledge of these variations might be effective in having an accurate dosimetry and treatment. Acknowledgment We thank Dr. Kourosh Sheibani the chief researcher at Clinical Research and Development Centre, Imam Hossein Medical Centre, for his valuable assistance in preparing and revising the manuscript. We would like to appreciate Mr. Mehdi Goodarzi for editing some parts of the manuscript. References


