Feasibility study of using MOSFET detectors for in vivo dosimetry during permanent low-dose-rate prostate implants

Radiotherapy and Oncology 80 (2006) 296–301 www.thegreenjournal.com

MOSFET detectors

Feasibility study of using MOSFET detectors for in vivo dosimetry during permanent low-dose-rate prostate implants Joanna E. Cygler*, Abdelhamid Saoudi, Gad Perry, Christopher Morash, Choan E The Ottawa Hospital Regional Cancer Centre, Ottawa, Canada

Abstract Purpose: To investigate the feasibility of using new micro-MOSFET detectors for QA and in vivo dosimetry of the urethra during transperineal interstitial permanent prostate implants (TIPPB). Methods and materials: This study involves measurements for several patients who have undergone the implant procedure with iodine-125 seeds. A new micro-MOSFET detector is used as a tool for in vivo measurement of the initial dose rate within the urethra. MOSFETs are calibrated using a single special order calibration seed. The angular response is investigated in a 100 kVp X-ray beam. Results: micro-MOSFETs are found to have a calibration factor of 0.03 cGy/mV for low energy X-rays and a high isotropic response (within 2.5%). Prostate volume and shape changes during TIPPB due to edema caused by the trauma of needle insertion, making it difficult to achieve the planned implant geometry and hence the desired dose distribution. MOSFET measurements help us to evaluate the overall quality of the implant, by analyzing the maximum dose received by urethra, the prostate base coverage, the length of the prostatic urethra that is irradiated, and the apex coverage. Conclusions: We demonstrate that ease of use, quick calibration and the instantaneous reading of accumulated dose make micro-MOSFETs feasible for in vivo dosimetry during TIPPB. c 2006 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 80 (2006) 296–301.



Keywords: MOSFET detectors; In vivo dosimetry; Prostate permanent implants

Transperineal interstitial permanent prostate brachytherapy (TIPPB) using 125I seeds has become a widely practiced treatment for early stage prostate cancer [1–3]. Tumour control, free of normal tissue complications, depends on the dose received by the tumour and the organs at risk, such as urethra, bladder and rectum. This in turn depends on how accurately the actual implant geometry follows the preplanned seed distribution. In our institution the preplan is based on the trans-rectal ultrasound images acquired during the volume study. The implant is performed under trans-rectal ultrasound guidance with the assumption that the geometry at the time of implant is the same as that of the volume study. Although care is taken to reproduce for the implant the same patient and ultrasound probe positions as for the volume study, some deviations in patient positioning are common, since the implant usually happens a week or more after the volume study. During the implant procedure, the physician may encounter difficulties in placing the sources in the exact preplanned positions, due to increasing prostate volume caused by edema, and variable prostate texture which can cause prostate motion in response to needle insertion and seed migration after insertion.



For these reasons, the resultant seed (and therefore dose) distribution usually differs from the preplanned distribution. The seed distribution is indicated by fluoroscopy performed during the implant, however these images are projections in one plane, which cannot show the distribution in three dimensions. In addition, the prostate itself is not directly visible on fluoroscopy. One can use, to some degree, the ultrasound images to evaluate the seed distribution during the implant procedure, however the ultrasound system used in our department does not reconstruct 3D images and the seeds do not show very well. The best way to make sure that the preplanned dose distribution is closely reproduced during the implant procedure is to use intraoperative treatment planning [4]. If obvious areas of inadequate coverage are present, some extra seeds can be added, but it is not possible to remove seeds that have already been implanted. If intraoperative treatment planning is not available, as is the case in our institution, the final dose distribution is calculated and analyzed one month later, based on CT images taken at that time. Prostate edema is believed to be gone by that time. Because the prostate volume changes between the beginning of the implant at the time the CT scan is acquired, the actual dose

0167-8140/$ - see front matter c 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2006.07.008

J.E. Cygler et al. / Radiotherapy and Oncology 80 (2006) 296–301

distribution also changes in a continuous way in that period of time. Because of deviations in seed positioning between the preplan and the implant, organs at risk such as the urethra may receive a dose higher than the planned one. When intraoperative treatment planning is not available, then real-time in vivo dosimetry can provide useful information about the quality of the implant while the patient is still in the operating room. Commercially available MOSFET detectors have been used as in vivo dosimeters for a number of years. MOSFETs are solid-state detectors whose properties have been extensively characterized [5,6]. They are more convenient than thermoluminescent dosimeters (TLDs) because they are waterproof, give an instant reading of accumulated dose, and they do not require extra handling such as annealing. MOSFET detectors have been successfully used recently in real-time in vivo dosimetery during intra-operative electron beam radiotherapy [7]. We have used specially designed MOSFET detectors to measure the dose rate inside the bladder and along the urethra immediately post-implant. Although these measurements add extra time to the implantation procedure, they are useful as an indicator of the quality of the implant, and also as a verification of the entire dosimetry procedure, including seed calibration, treatment planning, and implant geometry. All the patients involved in this study gave an informed consent to the work.

297

Xray Tube

3 cm

Fig. 1. Experimental setup for measurements of angular dependence of MOSFET response.

6.0 cm

MOSFET

1.0cm

Materials and methods The detectors used for the in vivo measurements are dual metal oxide semiconductor field effect transistors [5] (MOSFETs), model TN1002RDM (micro-MOSFETS) from Thomson & Nielsen Electronics Ltd., Ottawa, Canada. They are smaller (external dimensions 1 mm diameter and 0.5 mm thickness) and have a more isotropic response than the standard MOSFET detectors which were previously commercially available. Their metal leads have special 1 cm graduated marks, to allow for easy measurement of the detector position during the in vivo dosimetry procedure. In order to use MOSFET detectors for in vivo measurements, one needs to know their energy and angular responses, i.e. responses when irradiated at various angles. It has been shown that MOSFET sensitivity depends on radiation energy [8]. It is therefore necessary to calibrate each individual MOSFET detector in terms of the dose per unit detector reading using radiation of the same energy spectrum as encountered in the in vivo measurements. In our institution, type 6711 125I seeds (Nycomed-Amersham, Illinois) are used for prostate implants. At the time of this study the implants were performed using loose seeds and the needle placement was guided by ultrasound and a special template. Solid-state detectors are known to overrespond when exposed to low-energy photon radiation, due to the strong atomic number dependence of photoelectric effect. They are also known to under-respond to a certain extent to lower energies because of absorption by casings. Therefore, MOSFET detectors must be carefully calibrated when used for low photon energies such as from 125I (30 keV).

125I (6702) Source

6.0 cm

Solid Water Fig. 2. Experimental setup for MOSFET calibration with source.

125

I (6702)

The 100 kVp beam (mean beam energy approximately 30 keV) from an orthovoltage machine was used to evaluate the angular dependence of the response of the MOSFET detectors. Final calibration in terms of the dose per unit reading of each individual MOSFET detector was performed with a high activity 125I source (Nycomed-Amersham seed model 6702). The corresponding experimental setups are shown in Figs. 1 and 2.

MOSFET calibration procedures (a) Angular response Extensive characterization of the angular variation of the micro-MOSFET response for axial and normal to axial rotations was published by Roshau and Hintenlang [9] for 70 kVp X-rays. This measurement was done with the 100 kVp orthovoltage beam using a closed end square field cone of 10 · 10 cm2 at 50 cm FSD (focus-surface-distance). The angular response of the MOSFET detector was evaluated

298

MOSFET dosimetry in prostate implants

only for the rotation around the axis perpendicular to the beam axis. The MOSFET was embedded in a polystyrene phantom (25 cm · 25 cm · 3 cm) at a depth of 1.5 cm inside a special rotating insert. The insert could be rotated in steps of 30, see Fig. 1. (b) Calibration using an

125

I source

The response of solid-state detectors is a strong function of the photon beam energy at low photon energies [8], and scatter is a large component of the total particle fluence. Calibration of the MOSFETs in terms of the dose per unit detector reading was carried out in the solid water phantom using as the radiation source a high-activity 125I seed. This assured that the detectors were exposed to the same radiation energy spectrum during the calibration procedure as during the in vivo dosimetry. The type 6702 high-activity 125 I seed was used for MOSFETs calibrations, since it was available of higher activity than the type 6711, used for the prostate implants. Higher activity seed allowed shorter calibration times. The strength of a special high activity 125I (type 6702) source was measured using a Capintec CRC-12 well chamber (SN 12856), which had a calibration factor traceable to an ADCL. The calibration source and MOSFET detectors were positioned in a solid water phantom of dimensions, 30 cm · 30 cm · 13 cm. Three MOSFET detectors were used in the setup as shown in Fig. 2. The 125I source and each detector were embedded in a solid water phantom in such a way that each MOSFET was positioned at the TG-43 [10,11] reference point, i.e. on the transverse axis of the source, at the distance of 1 cm. There was a 6 cm thick solid water slab under the source as well as above the MOSFET to provide full scatter conditions. The experimental setup for MOSFET calibration is shown in Fig. 2.

Set- screw used as reference for distance measurement Balloon

In vivo measurements All MOSFET detectors were sterilized in CidexROPA solution (Johnson & Johnson Medical Products) for at least 20 min prior to use. Post-sterilization the detector was inserted into one branch of a Bardex three-way sterile urinary catheter (C. R. Bard Inc.), and then the catheter was inserted into the urethra. A small stainless-steel set screw was fixed to the tip of the catheter to act as a radio-opaque marker, to help visualize under fluoroscopy the initial MOSFET position in the bladder. MOSFET was pulled out along the urethra by 1 cm steps as indicated by the marks on the detector lead and checked using the ruler. For each measurement point, the distance between the MOSFET and the lower edge of the Foley balloon was recorded to provide the reference distance between the detector and the internal bladder wall. Fig. 3 shows a fluoroscopy image of the prostate implant with the modified catheter in place.

Results and discussion MOSFET calibration procedures (a) Angular response The angular isotropy of the detector response is crucial to assure no dosimetry artifacts. Fig. 4 shows the angular response of MOSFET detectors measured in a 100 kVp beam. This beam energy was chosen, because it has an effective energy of about 30 KeV, similar to that of an 125I source. The data on the graph are normalized to the detector response at 0 in the transversal plane. It can be seen that these detectors have a practically isotropic response. The largest deviation from isotropy is about 2.5%. This makes them ideal for in vivo dosimetry in a low-energy radiation field such as is present inside the 125I implanted prostate gland. Our results are in general agreement with those of Roshau and Hintenlang [9]. (b) Calibration using an

125

I source

We used TG-43 formalism [10,11] to calculate the dose rate at the MOSFET calibration point. According to TG-43

Seed

US Probe

Normalized response

120 110 100 90 80 70 60 50 0

30

60

90

120 150 180 210 240 270 300 330 360

Angle in degree

Fig. 3. Fluoroscopy image of the prostate implant with the modified Foley catheter in place.

Fig. 4. Angular dependence of MOSFET response in a 100 kVp photon beam. MOSFET readings are normalized to the reading at 0 transversal plane.

J.E. Cygler et al. / Radiotherapy and Oncology 80 (2006) 296–301

Detector

Average reading (mV/h)

Calibration factor (cGy/mV)

STD %

MOSFET#1 MOSFET#2 MOSFET#3

1354 1410 1490

33.48 E-3 32.14 E-3 30.42 E-3

2.2 3.0 1.6

_ 0 ; h0 Þ, for any cylindrical symmetrical sources, the dose, Dðr at the reference point (r0, h0) can be written as _ 0 ; h0 Þ ¼ SK K: Dðr

ð1Þ

The reference point, (r0, h0), is defined at r0 = 1 cm and h0 = 90. The air-kerma strength of the calibration seed was verified to be 45.2 U. Using Eq. (1) and applying the dose rate constant of 1.04 cGy/hU for this type of seed [11,12], the dose rate at the reference point, where MOSFET was positioned for the calibration, was 47.0 cGy/h. It has been assumed that the radiation field in the solid water phantom was exactly the same as in the corresponding phantom made of the liquid water. According to Mainegra et al.[13] this assumption is valid within 2.6%. Each MOSFET detector was left in the calibration geometry for 10 min. Therefore the total dose accumulated by each detector was 7.83 cGy. This allowed us to derive the calibration factors for each MOSFET used in the in vivo experiments. The results of the calibration for the individual dosimeters are listed in Table 1 and show 10% variation for this batch of detectors. This level of variation in MOSFET responses warrants the need for individual calibration factor for each detector used for patient measurements.

In vivo measurements The in vivo measurements of the dose in the urethra were carried out by moving the MOSFET detector along the urethra in 1 cm steps. The estimated uncertainty of the MOSFET position was within ±1 mm. The detector was left at each position to accumulate dose for 10 min, long enough to provide a reasonable measurement signal. The signal measured during 10 min interval ranged from 15 to 330 mV, depending on the detector position. We have measured the initial dose rate along the urethras of several patients. Based on our calibration measurement reproducibility, we estimate the accuracy of the in vivo measurements to be ±3% at each spatial position. An example of measured and calculated initial dose rates for the urethra as a function of the distance from the bladder wall is shown in Fig. 5 for one patient. The pre-plan dose profile along the urethra was based on dosimetric calculations performed on ultrasound images of the prostate acquired in 5 mm steps during the volume study procedure. At that time the aerated gel was injected into the Foley catheter to help identify the urethra. The pre-plan dose distribution calculations were done using the image based Brachytherapy Module of Theraplan Plus version 3.8 software. It is clearly seen in Fig. 5 that the dose rate increases for the parts of the urethra surrounded by the prostate implant. For this

12 150% mPD

Initial Dose Rate (cGy/h)

Table 1 Calibration factors for MOSFET detectors measured with iodine125 seed

299

10 8

mPD

6

90% mPD

Prostate Apex

Prostate Base

4

Prostate length = 50 mm

2 0 0

10

20

30

40

50 60 70 Distance (mm)

80

90

100 110

Fig. 5. Initial pre-plan (d) and measured post-implant (m) dose rates inside the urethra.

patient, the maximum initial dose rate to the urethra was 12 cGy/h. Applying the following equation for the total integral dose, DOSE ¼ D_ 0  1:44  T 1=2 ;

ð2Þ

where D_ 0 is the initial dose rate, T1/2 = 59.4 days is the halflife of 125I, the maximum dose to the urethra is 246.3 Gy. This is 1.7 times larger than the minimum peripheral dose (mPD) of 145 Gy prescribed to the prostate gland. For this patient, the full width at half maximum of the dose-versus-distance curves for the urethra is 5.2 cm. This means that 5.2 cm of the urethra receives the dose larger than 123 Gy and less than 246.3 Gy. The value of the maximum urethral dose rates for different patients in this study ranged from 10 to 16 cGy/h, corresponding to a total absorbed dose of 205–328 Gy. The shape of the dose-rate vs. distance-curve can help to evaluate the overall implant quality. Patient data are summarized in Table 2. All patients in this study had a prostate volume about 50 cm3 or less and an average number of seeds per unit volume of 2.1 seeds/cm3. The patients who received a greater number of seeds per unit volume were those having a small prostate size (e.g. patient #3). For this patient, the smaller prostate volume resulted in a higher proportion of seeds close to the urethra, causing a higher urethral dose. The analysis of the measured data for all patients has been done in terms of the following parameters, shown in Table 2: UL – pre-implant prostatic urethra length (mm) based on ultrasound images acquired during the volume study N – number of implanted seeds IDRmax – maximum initial dose rate along the urethra (cGy/h) measured during post-implant in vivo dosimetry procedure PUL(mPD) – immediate post-implant prostatic urethra length (mm) covered by dose = mPD PUL(150%mPD) – immediate post-implant prostatic urethra length (mm) covered by dose = 150%mPD L (PB  PmPD) – immediate post-implant distance along the prostatic urethra from the bladder/base (Base) interface, PB, to the point PmPD, at which the urethral dose reaches the mPD value.

300

MOSFET dosimetry in prostate implants

Table 2 Compiled data from measurements Pat#

UL (mm)

N

IDRmax cGy/h

PUL(mPD) (mm)

PUL(150%mPD) (mm)

L(PBPmPD) (mm)

PUL UL

#1 #2 #3 #4 #5

35 50 25 40 35

93 108 51 89 93

12.2 11.1 15.6 10.8 10.4

47 47 23 26 41

27 15 13 0 0

0 5.4 8.6 8.2 2.2

133% 105% 125% 86% 123%

 100%

UL – pre-implant prostatic urethra length (mm) based on ultrasound images acquired during the volume study; N – number of implanted seeds; IDR – maximum initial dose rate along the urethra (cGy/h), measured during post-implant in vivo dosimeter procedure; PUL(mPD) – immediate post-implant prostatic urethra length (mm) covered by dose = mPD; PUL(150%mPD) – immediate post-implant prostatic urethra length (mm) covered by dose = 150%mPD; L (PB  PmPD) – immediate post-implant distance along the prostatic urethra from the Base (bladder/base interface) to the point at which the dose reaches mPD value. It corresponds to the length of the prostate not covered by mPD.

It corresponds to the length of the prostate not covered by mPD. Based on these data the following clinical information can be extracted: 1. Quality of the prostate implant in terms of the target coverage. Minimum peripheral dose, mPD, is the dose that encompasses the target volume (prostate with appropriate margin) [12]. The goal of the permanent prostate implant with 125I seeds is to deliver mPD equal to 145 Gy. However mPD describes the dose to the prostate surface only. Actual doses to the central portions of the gland and tumour may be considerably higher. Since urethra is positioned centrally inside the implanted volume, it usually receives doses equal to or higher than mPD. The dose to urethra can vary significantly, depending on the strength of the sources and the implant technique. During generation of the pre-implant treatment plans the utmost care is taken to minimize the dose to urethra without compromising the tumour coverage. The measured dose versus distance-curve along the prostatic urethra contains also information about two distinct parameters related to the quality of the implant: the dose rate at the base of the prostate, and the length of the urethra, that receives the minimum peripheral dose, PUL(mPD), compared to the pre-planned length of the urethra to receive the mPD, UL. Properly implanted base of the gland will result in a steep dose gradient at the superior end of the prostatic urethra. The value of mPD will be reached very rapidly, within millimeters from the prostate wall, as it does for patients #1 and 5, where this distance is 0– 2 mm. This rapid increase is expected, especially when the base is preplanned to receive mPD at the external wall of the prostate (the bladder-prostate interface), see Fig. 5. In the case of patients #3 and 4, the distance over which the prostatic urethra reaches mPD is greater than 8 mm. This clearly demonstrates a deficiency of the dose delivered to the base. It also underscores the potential value of realtime dosimetry using MOSFETs, since a measurement made immediately post-implant could allow correction of the under-dose of the prostate base. For patient #2 it is difficult to conclude that the base is under-dosed especially with a 10 mm measurement step.

In this case the dose sampling should have been done with a maximum 5 mm step in the base region. The length of the prostatic urethra before implant, UL, is the same or smaller than the length of the prostate covered by the mPD, PUL(mPD), in the superior–inferior direction immediately post-implant. This is expected due to the prostate swelling during the implant procedure. In Table 2 the last column indicates that for patients #1, 2, 3 and 5 the apex of the prostate is well covered by the mPD. However for patient #4, about 14% of the initial length of the prostate appears to be under-dosed. The reason why the base and/or the apex are often not well covered is probably related to prostate movement with needle insertion and retraction, and the frequent change of the relative base position during the implant procedure due to edema caused by the trauma of needle insertion [14,15]. In order to minimize the potential of under-dosing base and/or apex of the gland one has to ensure that the length of the prostate covered by mPD in the superior–inferior direction is greater than or equal to the length of the prostatic urethra before implant, UL. This can be done during pre-planning by adding an appropriate superior–inferior margin. 2. Percentage of the urethra receiving dose higher than 150% of mPD. The urethra is the main organ at risk in TIPPB. Urethral irritation is the main cause of symptoms following prostate brachytherapy. It has been recommended that urethra should receive dose not higher than 200% of mPD1. Table 2 shows that for patients #1–3 the length of the prostatic urethra getting more than 150% of mPD varies between 13 and 27 mm. For patient #3 the initial dose rate in the central part of the urethra was 15.62 cGy/h, which corresponds to a total dose of 320.7 Gy. This is greater than the recommended dose limit, which is 1.5 to 2 times mPD. For this patient, the smaller prostate volume resulted in a higher proportion of seeds close to the urethra, causing a higher urethral dose. After the implant, there is no practical way to reduce this urethral dose, but in vivo measurements might suggest that such patient be followed closely and considered to be at increased risk for urethral complications. It also underscores the potential value of real-time dosimetry using MOSFETs. The drawback of using one MOSFET detector

J.E. Cygler et al. / Radiotherapy and Oncology 80 (2006) 296–301

to measure the dose along the entire length of the prostatic urethra is that it takes about one hour (approximately 10 min per position). This is excessively long if the measured data are to be used as the basis for adjusting the implant. However, the data are still useful for evaluation of the overall quality of the implant. In order to decrease the measurement time to a practical limit, we are currently working on designing a special linear array of MOSFET detectors, to enable the simultaneous reading of the initial dose rate at several positions inside the urethra.

Conclusions Specially designed MOSFET detectors are very useful for in vivo dosimetry of permanent prostate implants. When inserted into the urethra, they can measure in real-time the initial dose rate received by this organ. This can serve not only as an indicator of possible treatment complications due to excessive dose to the urethra, but also as a measure of the overall quality of the implant. In addition, if the measurements indicate under-dose of base and/or apex of the prostate, additional seeds can be inserted to correct this problem.

Acknowledgements We gratefully acknowledge Thomson–Nielsen for providing us with special MOSFET detectors and Nycomed Amersham for donating a high activity I-125 seed. * Corresponding author. Joanna E. Cygler, Department of Medical Physics, Ottawa Hospital Regional Cancer Centre, 501 Smyth Road, Box 927, Ottawa, Ont., Canada K1H 8L6. E-mail address: [email protected] Received 1 February 2006; received in revised form 31 May 2006; accepted 7 July 2006; Available online 14 August 2006

References [1] Nag S, Bice W, DeWyngaert K, Prestidge B, Stock R, Yu Y. The American Brachytherapy Society recommendations for permanent prostate brachytherapy postimplant dosimetric analysis. Int J Radiat Oncol Biol Phys 2000;46:221–30. [2] Nag S, Beyer D, Friedland J, Grimm P, Nath R. American Brachytherapy Society (ABS) recommendations for transperineal permanent brachytherapy of prostate cancer. Int J Radiat Oncol Biol Phys 1999;44:789–99.

301

[3] Merrick G, Butler W. Modified uniform seed loading for prostate brachytherapy: rationale, design and evaluation. Tech Urol 2000;6:78–84. [4] Zelefsky MJ, Yamada Y, Cohen G, Venkatraman ES, Fung AYC, Furhang E, Silvern D, Zaider M. Postimplantation analysis of permanent transperineal prostate implantation: improved dose distributions with an intraoperative computer-optimized conformal planning technique. Int J Radiat Oncol Biol Phys 2000;48:601–8. [5] Soubra M, Cygler J, Mackay G. Evaluation of dual bias dual metal oxide-silicon semiconductor field effect transistor detector as radiation dosimeter. Med Phys 1994;21:567–72. [6] Ramani R, Russell S, O’Brien P. Clinical dosimetry using MOSFETs. Int J Radiat Oncol Biol Phys 1997;37:959–64. [7] Ciocca M, Piazzi V, Lazzari R, Vavassori A, Luini A, Veronesi P, Galimberti V, Intra M, Guido A, Tosi G, Veronesi U, Orecchia R. Real-time in vivo dosimeter using micro-MOSFET detectors during intraoperative electron beam radiation therapy in early-stage breast cancer. Radiother Oncol 2006;78:213–6. [8] Kron T, Duggan L, Smith T, Rosenfeld A, Butson M, Kaplan G, Howlett S, Hyodo K. Dose response of various radiation detectors to synchrotron radiation. Phys Med Biol 1998;43:3235–59. [9] Roshau JN, Hintenlang DE. Characterization of the angular response of an ‘‘isotropic’’ MOSFET dosimeter. Health Phys 2003;84:376–9. [10] Nath R, Anderson LL, Luxton G, Weaver KA, Williamson JF, Meigooni AS. Dosimetry of interstitial brachytherapy sources: Recommendations of the AAPM Radiation Therapy Committee Task Group No. 43. Med Phys 1995;22:209–34. [11] Williamson JF, Coursey BM, DeWerd LA, Hanson WF, Nath R, Ibbot G. Guidance to Users of Nycomed Amersham and North American Scientific, Inc. I-125 Interstitial Sources: Dosimetry and Calibration Changes: Recommendation of the American Association of Physicists in Medicine Radiation Therapy Committee Ad Hoc Subcommittee on Low-Energy Seed Dosimetry. Medical Physics 1999;26:570–573. [12] Yu Y, Anderson LL, Li Z, et al. Permanent prostate seed implant brachytherapy: Report of the American Association of Physicists in Medicine Task Group No. 64. Med Phys 1999;26:2054–76. [13] Mainegra E, Capote R, Lopez E. Dose rate constants for 125I, 103 Pd, 192Ir, 169Yb brachytherapy sources: an EGS4 Monte Carlo Study. Phys Med Biol 1998;43:1557–66. [14] Waterman FM, Yu N, Corn BW, Dicker AP. Edema associated with I-125 or Pd-103 prostate brachytherapy and its impact on post-implant dosimetry: an analysis based on serial CT acquisition. Int J Rad Oncol Biol Phys 1998;41:1069–77. [15] Yu N, Dicker AP, Nath R, Waterman FM. The impact of edema on planning 125I and 103Pd prostate implants. Med Phys 1999;26:763–7.