tg-43

TG-43 U1 based dosimetric characterization of model 67-6520 Cs-137brachytherapy source

Ali S. Meigoonia!Department of Radiation Medicine, North Shore University Hospital, 300 Community Drive, Manhasset,New York 11030 and Department of Radiation Medicine, University of Kentucky Chandler MedicalCenter, Lexington, Kentucky 40536-0084

Clarissa Wright, Rafiq A. Koona, and Shahid B. AwanDepartment of Radiation Medicine, University of Kentucky Chandler Medical Center,Lexington, Kentucky 40536-0084

Domingo GraneroDepartment of Radiation Physics, ERESA, Hospital General Universitario, Avenida Tres Cruces,2, E-46014 Valencia, Spain

Jose Perez-CalatayudDepartment of Oncology, Physics Section, La Fe University Hospital, Avenida Campanar 21,E-46009 Valencia, Spain

Facundo BallesterDepartment of Atomic, Molecular and Nuclear Physics, University of Valencia, C/ Dr. Moliner 50,E-46100 Burjassot, Spain and Instituto de Fsica Corpuscular (IFIC), C/ Dr. Moliner 50,E-46100 Burjassot, Spain!Received 15 March 2009; revised 19 August 2009; accepted for publication 19 August 2009;published 16 September 2009"

Purpose: Brachytherapy treatment has been a cornerstone for management of various cancer sites,particularly for the treatment of gynecological malignancies. In low dose rate brachytherapy treat-ments, 137Cs sources have been used for several decades. A new 137Cs source design has beenintroduced !model 67-6520, source B3-561" by Isotope Products Laboratories !IPL" for clinicalapplication. The goal of the present work is to implement the TG-43 U1 protocol in the character-ization of the aforementioned 137Cs source.Methods: The dosimetric characteristics of the IPL 137Cs source are measured using LiF thermolu-minescent dosimeters in a Solid Water phantom material and calculated using Monte Carlosimulations with the GEANT4 code in Solid Water and liquid water. The dose rate constant, radialdose function, and two-dimensional anisotropy function of this source model were obtained fol-lowing the TG-43 U1 recommendations. In addition, the primary and scatter dose separation !PSS"formalism that could be used in convolution/superposition methods to calculate dose distributionsaround brachytherapy sources in heterogeneous media was studied.Results: The measured and calculated dose rate constants of the IPL 137Cs source in Solid Waterwere found to be 0.930!!7.3%" and 0.928!!2.6%" cGy h1 U1, respectively. The agreementbetween these two methods was within our experimental uncertainties. The Monte Carlo calculatedvalue in liquid water of the dose rate constant was "=0.948!!2.6%" cGy h1 U1. Similarly, theagreement between measured and calculated radial dose functions and the anisotropy functions wasfound to be within !5%. In addition, the tabulated data that are required to characterize the sourceusing the PSS formalism were derived.Conclusions: In this article the complete dosimetry of the newly designed 137Cs IPL source fol-lowing the AAPM TG-43 U1 dosimetric protocol and the PSS formalism is provided. 2009American Association of Physicists in Medicine. #DOI: 10.1118/1.3224462$

Key words: brachytherapy, Cs-137, PSS model, dosimetry, GEANT4, TG-43

I. INTRODUCTIONSoon after the discovery of 226Ra, the importance of theclinical application of brachytherapy sources wasdiscovered.1 However, problems of using 226Ra sources dueto the production of radon gas and the possibility of break-down of the source capsule and its releasing into the patientswere some of the strong motivators to find and use alterna-tive radioactive materials.2 137Cs sources were introduced as

a substitute for 226Ra in brachytherapy implants and havebeen used for several decades for low dose rate !LDR",3 mostcommonly in treatment of gynecological cancer patients,through intracavitary implants.4

A historical review of the existing publications will givean idea regarding the current status of dosimetric character-ization for 137Cs sources. In 1968, Meisberger et al.5 mea-sured the effective attenuation of different gamma-emitter

4711 4711Med. Phys. 36 10, October 2009 0094-2405/2009/3610/4711/9/$25.00 2009 Am. Assoc. Phys. Med.

radionuclides, including 137Cs, in water as compared to air.However, they did not determine the source self-absorptionand anisotropy of the radiation distribution around thesources. In 1972, Krishnaswamy6 showed the effects of ob-lique filtration through walls in 137Cs needles and tubes ascompared to equivalent radium sources. Based on this infor-mation, he calculated the dose distribution around 137Cssources, with different active lengths, using an along-and-away format !limited up to 5 cm distance from the sourcecenter". Despite the differences in the geometric structures ofvarious source models, the Krishnaswamy data are still beingused for a quick hand calculation or double check in brachy-therapy procedures with LDR 137Cs sources. In 1988,Williamson7 presented dose rate tables and treatment plan-ning data for two newly designed 137Cs sources, namely,gold matrix !series 67-800, intracavitary source, by Radia-tion Therapy Resources, Inc., Valencia, CA" and a discreteseed source !Oris Intracavitary tube source from CIS-US,Inc., Lake Success, NY". The Monte Carlo generated doserate tables are presented in an along-and-away format, whichare utilized in the Sievert integral based treatment planningsystems. In 1989, Waggener et al.8 wrote a computer pro-gram to generate the dose distribution around an asymmetric137Cs source !3M". They confirmed the accuracy of the cal-culated dose rate values by comparison with measured datausing the Fricke dosimeter. The angular distributions of thedose rates for different radial distances are presented in atabulated format. They have shown that the asymmetric dis-tribution of the 137Cs leads to a large difference !67%" be-tween the dose rates at the two ends of the source !radialdistance of 1 cm". These differences decrease to less than10% at points lying more than 3 cm from the source center.Using Monte Carlo simulation, in 1998, Williamson9 calcu-lated dose rate distributions around the Amersham CDCS.Jtype 137Cs source and also updated the dosimetric informa-tion of the 3M model 6500/6D6C 137Cs sources. The resultsand corresponding parameters were utilized in the Sievertintegral based treatment planning systems. The above notedinvestigations were published prior to the original10 orupdated11 Task Group No. 43 recommendations !TG-43 orTG-43 U1, respectively" by the American Association ofPhysicist in Medicine !AAPM".

Using the Monte Carlo simulation technique, Casal etal.12 obtained the absolute dose rates around the CDCS-M-type 137Cs source. These results were presented in bothTG-43 parameters and the along-and-away tabulated format.They compared their tabulated results with Krishnaswamysdata6 for 137Cs sources and demonstrated that for distancesgreater than 1 cm the differences were approximately 2%lower, and that these differences rise up to 4% at 0.5 cm.Other published data regarding the TG-43 or TG-43 U1based dosimetric parametrizations of different 137Cs sourcepertains to models CSM11,13 CSM3-a,14 CSM2, andCSM3,15 Walstam CDC.K type,16 CDC-type miniature,17 andthe Selectron pellet.18 Liu et al.19 determined the dosimetricparameters of the CSM-type 137Cs sources based on theoriginal TG-43 formalism. Prez-Calatayud et al.20 also ob-tained the two-dimensional !2D" dose rate distributionaround the CSM-type 137Cs sources and compared their re-sults with the calculated parameters by Liu et al.19

As the production of several of the above noted sourceshave been discontinued, a new 137Cs source design has beenintroduced !model 67-6520, source B3-561" by Isotope Prod-ucts Laboratories !IPL" to meet the demand of the currentclinical procedures. To date, there is no publication on thedosimetric parametrizations of this new source model.

The goal of this work is to determine the dosimetric char-acteristics of the new 137Cs source !model 67-6520, sourceB3-561" using the TG-43 U1 protocol.11 These determina-tions will be performed using experimental and Monte Carlosimulation techniques. In addition, the primary and scatterphoton dose contributions of the new source will be deter-mined for their application in convolution/superposition dosecalculation methods around brachytherapy implant in hetero-geneous media.

II. MATERIALS AND METHODSII.A. Source characteristics

Figure 1 shows the schematic diagram of the model 67-6520 137Cs source manufactured by IPL !24937 Avenue Tib-bitts, Valencia, CA" and distributed by Radiation ProductsDesign, Inc. !5218 Barthel Industrial Drive, Albertville,MN". In the following sections we refer to this source as the

FIG. 1. Schematic diagram of the model 67-6520 137Cs source !courtesy of the Isotope Product Laboratories". The source tip is at the side of the aluminumring.

4712 Meigooni et al.: Dose rate distribution of the IPL source 4712

Medical Physics, Vol. 36, No. 10, October 2009

IPL 137Cs source. The physical dimensions of the IPL 137Cssource are 3.05 mm in diameter and 20 mm in length. Thecapsule is composed of two types of 304 stainless steel lay-ers with a total wall thickness of 0.584 mm !outer layer of0.33 mm, inner layer of 0.254 mm". The active portion of thesource is 1.52 mm in diameter and 14.8 mm in length. Theradioisotope is uniformly distributed in the core of the ce-

sium oxide ceramic source, assuming approximately 5.29 mgCs2O in a 50 mCi source. The density of the active ceramicmaterial is 1.47 g /cm3 while the density of the ceramic itselfis 1.27 g /cm3. A color coded aluminum ring easily identifiesthe source activity and defines the source tip. The active coreis placed nonsymmetrically within the capsule with its tipand end located at 3 and 2 mm, respectively, from the cap-sule. The calibration of the air-kerma strength of this sourceis traceable to the primary standards by the National Instituteof Standard and Technology !NIST" through the AccreditedDosimetric Calibration Laboratory !ADCL" at the Universityof Wisconsin, Madison. The source calibration has been per-formed by the vendor !Eckert & Ziegler Isotope Products,Valencia, CA" prior to its shipment to our institution for do-simetric evaluation. A secondary calibration system has beendeveloped at our institution using a Capintec CRC-7R well-type chamber for daily verification process.

II.B. Experimental setup characteristicsfor thermoluminescent dosimeter measurementsin Solid Water

Dosimetric characteristics of the IPL 137Cs source weredetermined experimentally using 1.0#1.0#1.0 and 3.1#3.1#0.89 mm3 TLD-100 LiF thermoluminescent dosim-

FIG. 2. Comparison between the Monte Carlo simulated !solid line" andmeasured !symbol" radial dose functions of the IPL 137Cs source in SolidWater.

FIG. 3. Monte Carlo calculated radial dose functions for the IPL source. These data was fitted to a 5th order polynomial with coefficients a0=1.00765, a1=7.073#103 cm1, a2=5.680#104 cm2, a3=1.376#105 cm3, a4=1.117#106 cm4, and a5=1.608#108 cm6 between 0.25 and 20 cm.



eters !TLDs" !TLD-100, Harshaw/Bicron 6801 Cochran Rd.,Solon, OH 44139" in a 30#30#25 cm3 Solid Waterphantom. Our TLD procedures have been described in detailin our previous publications.21,22 As described in these pub-lications, slabs of the Solid Water phantom !RadiationMeasurements Inc., RMI, Middletown, WI" were custom de-signed and machined to accommodate the brachytherapysource and TLD chips for a particular dosimetric procedure.

One of these slabs of Solid Water was designed formeasurement of the radial dose function and dose rate con-stant. Four TLD chips, separated by 90 from each other,were placed at each radial distance, on the transverse bisectorplane of the source. In addition, the positions of the TLDswithin the phantom were selected in a specific pattern inorder to minimize interference of any one TLD to the ab-sorbed dose of any other TLD chip. A cylindrical plug wasdesigned to hold the source in the desired position and alsofacilitate a rapid placement and removal of the source withinthe phantom setup. The radial dose function was measured atradial distances ranging from 0.5 to 10 cm, with 0.5 cmincrements from 0.5 to 2 cm, and with 1 cm increments from2 to 10 cm. Dose rate constant was obtained from the TLDslocated at 1 cm radial distance from the source. The finaldose rate constant was extracted from the average of at least20 TLD chips.

Another Solid Water phantom material slab was de-signed for measurement of 2D anisotropy function of the IPL137Cs model21,22 according to the TG-43 U1 protocol.11 Inthis slab, the TLDs were arranged in a circular fashion atradial distances of 2, 3, 5, and 7 cm from the source center.Measurements were performed in angular increments of 10from 0 to 350. With this design, either two sets of datapoints from 0 to 180 or four sets of data points in the rangeof 090 have been obtained.

The irradiated TLDs were read using a Harshaw model3500 TLD reader and was annealed using a standard anneal-ing technique.23 The response or readout !also known as TL"of the TLD chips was converted to dose by calibrating theresponses of several TLD chips from the same set using the6 MV photons beams from a Varian 2100CD linear accelera-tor. The TLD calibrations were performed using Solid Wa-ter phantom material !20#20#20 cm3". One of the slabsof the phantom materials was accurately machined to accom-modate the TLD chips in a plane perpendicular to the centralaxis of the beam. The TLD chips were placed at the depth ofthe maximum dose, around the central axis of the beam, withthe calibration beam geometry !i.e., 100 cm SSD and 10#10 cm2 field size". Five TLD chips were exposed to 25cGy and five were exposed to 100 cGy. The slope of thelinear curve obtained from the readout of these chips to theirradiated dose was found to be the calibration factor of thosechips. The output of this linear accelerator had recently beencalibrated !monthly QA procedures" with an ion chamber ina water phantom, following the AAPM TG-51 recommenda-tions. The measurements of the dose distributions around the137Cs source were carried out by predetermination of theexposure time for a dose range of 10100 cGy, where the

TLD response is linear.23 If larger doses are used during themeasurements, the calibration dose range increases to coverthe maximum dose and also provide the nonlinearity of theTLD response.

II.C. Monte Carlo calculations

The Monte Carlo transport Code GEANT4 version 8.024was used to calculate the air-kerma and the dose rate distri-butions around an IPL 137Cs source in liquid water and inSolid Water phantom materials. GEANT4 has been used indosimetric evaluation of different models of brachytherapysources, including 137Cs sources,20,18 and based on the rec-ommendations in the AAPM-ESTRO prerequisite report, ithas been proven to be a well established code in this field.25

In the present study, the $ spectrum of the 137Cs sourcewas not considered in the simulations, because its contribu-tion to the dose rate distribution was found to be negligible.26Photon interaction models for the Compton scattering, pho-toelectric effect, and Rayleigh scattering processes weretaken from the low energy package of GEANT4. In addition,this code uses the cross section data from the EPDL97library,27 as it is recommended in the TG-43 U1 report.11 Inthese simulations, the cutoff photon energy was set to be 10keV.

The track length estimator28 was used to estimate colli-sion kerma. In this model, it has been assumed that the sec-ondary electrons are under electronic equilibrium conditionand the collision kerma is equal to the absorbed dose. Con-sequently, no electron transport was considered in thesesimulations. For typical 137Cs sources, electronic equilibriumis reached at a radial distance of 0.35 cm from the sourcecenter.26 At 0.25 cm kerma is approximately 3% lower thandose.

In order to validate the accuracy of the source design,dimensions, and composition materials used in the MonteCarlo studies, the Monte Carlo simulations were performedin a 30#30#25 cm3 Solid Water phantom, which isidentical to that used in the experimental setup for TLD mea-surements. The chemical composition and density of SolidWater phantom were described in previous publications.29The kerma !to liquid water in Solid Water" was scored incylindrical shell tally cells !height and width of 0.05 cm"with their longitudinal axes coincident with the longitudinalaxis of the source.

After validation of the Monte Carlo algorithm and accu-racy of the source characteristics used in the simulation, doserate distribution in liquid water was calculated using thesame source input data. In these MC simulations, the IPL137Cs source was placed at the center of a 40 cm radiusspherical liquid water phantom, which was recommended asan unbounded phantom for dose calculation at distances ofup to 20 cm from the source center.30,31 To obtain the doserate distribution in an along-and-away format, D !y ,z", thekerma rates were scored in 400#800 cylindrical shell tallycells !height and width of 0.05 cm" with their longitudinalaxes coincident to the longitudinal axis of the source. Thesecond grid system was used to obtain the dose rate distri-



butions in polar coordinates, D !r ,%", to derive the TG-43 U1formalism parameters.11 This grid was composed of 400#180 cells bounded by two concentric spheres, 0.05 cm ra-dial differences, and two concentric cones with their apex atthe source center, 1 angular width. The center of this tallycell is located at !r ,%".

The air-kerma strength of the IPL 137Cs source has beendetermined in the void space according to the TG-43 U111recommendations. The air kerma SK was calculated at a dis-tance of 100 cm along the transverse bisector of the source.In these simulations, the tally cell was chosen to be a squaredannulus !1 cm thick and 1 cm high" with longitudinal axiscoincident with the longitudinal axis of the source.

II.D. TG-43 U1 dosimetry protocol

As described in this protocol, the general 2D dose rateequation is defined as

D !r,%" = SK"GX!r,%"

GX!1 cm,&/2"gX!r"F!r,%" , !1"

where D !r ,%" is the dose rate at the point of interest at somedistance r from the center of the source and polar angle %relative to the longitudinal axis of the source, SK is the air-kerma strength, " is the dose rate constant in water, GX!r ,%"is the geometry function at the point of interest,GX!1 cm,& /2" is the geometry function at the referencepoint 1 cm from the source and at 90 from the longitudinalaxis of the source, gX!r" is the radial dose function, and

TABLE I. Propagation of errors estimated for the experimental and Monte Carlo procedures used in the presentinvestigations. The coverage index of the uncertainties is k=1. The components identified with a * were notutilized for derivation of the total uncertainty in the experimental radial dose function and 2D anisotropyfunction data, as the data are formed from ratios of measured doses, and are therefore unaffected by theidentified components. The underlined italic-bold entries are those that have been used for error bars of theradial dose and anisotropy functions.

Component

TLD measurementuncertainties

!%"

Monte Carlouncertainties

!%"

Type A Type B Type A Type B

Repetitive measurements 4TLD dose calibration !including LINAC calibration" 2.0LiF energy correction 5.0*Source and TLD position 2.0Air-kerma strength 3.0* 0.5Statistical uncertainties 0.5Cross sections 2.5Quadrature sum 4 6.5/2.8 0.7 2.5Total uncertainty 7.3/4.9 2.6

r (cm)0 2 4 6 8 10 12 14 16 18 20

)1 g2

/R(cm

2 rpi4)0=9

0

D(r,

0

0.005

0.01

0.015

0.02

0.025

0.03 total

total scatter

primary

single scatter

multiple scatter

FIG. 4. Separation of the different contributions to dose along the transverseaxis of the IPL source. Dose has been normalized to the total emitted radiantenergy of photons per unit solid angle, R /4&, and multiplied by the distanceto the source squared.

E (MeV)0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

)1

(dR/dE)/R(M

eV

610

510

410

310

210

110

1

10

210

310

FIG. 5. Energy-weighted photon spectrum per unit emitted radiant energy ofphotons exiting the IPL source.



TABL

EII

.D

ose

rate

peru

nit

air-k

erm

ast

reng

th!c

Gy

h1

U

1 "ar

ou

nd

the

IPL

sou

rce.

The

zax

isis

defin

edal

ong

the

sou

rce

posit

ive

tow

ard

the

sou

rce

tip.T

heo

rigin

isat

the

geom

etric

cen

ter

oft

heso

urc

e!n

ot

atth

eac

tive

cen

ter".

Dist

ance

alon

gz

!cm

"

Dist

ance

away

y!c

m"

00.

250.

50.

751

1.5

22.

53

45

68

1014

'14

0.00

426

0.00

425

0.00

423

0.00

419

0.00

416

0.00

413

0.00

408

0.00

407

0.00

401

0.00

390

0.00

374

0.00

355

0.00

311

0.00

265

0.00

184

'10

0.00

916

0.00

918

0.00

907

0.00

896

0.00

887

0.00

878

0.00

873

0.00

859

0.00

842

0.00

795

0.00

735

0.00

672

0.00

545

0.00

433

0.00

268

'8

0.01

463

0.01

488

0.01

465

0.01

444

0.01

429

0.01

419

0.01

398

0.01

365

0.01

320

0.01

209

0.01

082

0.00

955

0.00

727

0.00

547

0.00

315

'6

0.02

68

0.02

72

0.02

67

0.02

63

0.02

62

0.02

58

0.02

50

0.02

38

0.02

24

0.01

937

0.01

642

0.01

375

0.00

962

0.00

681

0.00

363

'5

0.03

94

0.04

00

0.03

89

0.03

84

0.03

83

0.03

72

0.03

54

0.03

30

0.03

03

0.02

50

0.02

03

0.01

646

0.01

094

0.00

749

0.00

386

'4

0.06

40

0.06

40

0.06

17

0.06

12

0.06

05

0.05

74

0.05

27

0.04

74

0.04

21

0.03

25

0.02

51

0.01

954

0.01

230

0.00

816

0.00

406

'3

0.11

67

0.11

65

0.11

28

0.11

14

0.10

79

0.09

65

0.08

32

0.07

05

0.05

93

0.04

21

0.03

05

0.02

28

0.01

358

0.00

875

0.00

422

'2.

50.

177

50.

172

10.

167

10.

162

20.

153

40.

130

10.

106

60.

086

50.

070

20.

047

40.

033

30.

024

30.

014

150.

009

000.

004

29'

20.

295

0.28

20.

272

0.25

50.

230

0.17

94

0.13

73

0.10

57

0.08

25

0.05

29

0.03

60

0.02

57

0.01

466

0.00

922

0.00

435

'1.

50.

606

0.56

10.

514

0.44

00.

366

0.25

00.

175

0.12

70

0.09

52

0.05

80

0.03

84

0.02

70

0.01

508

0.00

939

0.00

439

'1

1.

777

1.19

00.

816

0.58

70.

338

0.21

60.

147

70.

106

60.

062

10.

040

20.

027

90.

015

380.

009

520.

004

43'

0.75

4.

071.

821

1.07

90.

718

0.38

20.

234

0.15

64

0.11

12

0.06

38

0.04

09

0.02

83

0.01

551

0.00

958

0.00

444

'0.

5

6.45

2.44

1.32

60.

838

0.41

90.

249

0.16

31

0.11

47

0.06

50

0.04

15

0.02

85

0.01

558

0.00

961

0.00

445

'0.

25

7.30

2.81

1.49

20.

919

0.44

40.

258

0.16

75

0.11

69

0.06

58

0.04

18

0.02

87

0.01

564

0.00

962

0.00

446

0

7.50

2.92

1.54

90.

948

0.45

30.

261

0.16

90

0.11

77

0.06

59

0.04

18

0.02

87

0.01

564

0.00

963

0.00

445

0.25

7.

302.

811.

492

0.91

90.

444

0.25

80.

167

40.

116

90.

065

70.

041

70.

028

60.

015

620.

009

620.

004

450.

5

6.45

2.44

1.32

60.

838

0.41

90.

248

0.16

31

0.11

47

0.06

50

0.04

14

0.02

85

0.01

557

0.00

960

0.00

445

0.75

4.

071.

821

1.07

90.

718

0.38

20.

234

0.15

64

0.11

12

0.06

38

0.04

09

0.02

83

0.01

549

0.00

957

0.00

444

1

1.77

91.

190

0.81

60.

587

0.33

80.

216

0.14

77

0.10

66

0.06

21

0.04

02

0.02

79

0.01

539

0.00

953

0.00

443

1.5

0.58

00.

557

0.51

50.

440

0.36

50.

250

0.17

50

0.12

70

0.09

52

0.05

80

0.03

84

0.02

69

0.01

508

0.00

939

0.00

440

20.

283

0.27

80.

271

0.25

50.

230

0.17

95

0.13

73

0.10

56

0.08

25

0.05

28

0.03

60

0.02

57

0.01

466

0.00

922

0.00

435

2.5

0.16

90

0.16

92

0.16

61

0.16

20

0.15

34

0.13

01

0.10

66

0.08

648

0.07

03

0.04

74

0.03

33

0.02

43

0.01

417

0.00

900

0.00

429

30.

115

50.

114

40.

112

00.

111

00.

107

80.

096

50.

083

30.

070

470.

059

30.

042

10.

030

50.

022

70.

013

560.

008

750.

004

224

0.06

18

0.06

25

0.06

10

0.06

09

0.06

03

0.05

74

0.05

27

0.04

741

0.04

21

0.03

26

0.02

51

0.01

954

0.01

229

0.00

816

0.00

406

50.

038

20.

039

20.

038

40.

038

10.

038

10.

037

20.

035

30.

032

960.

030

30.

025

00.

020

30.

016

460.

010

940.

007

490.

003

866

0.02

68

0.02

69

0.02

63

0.02

61

0.02

60

0.02

57

0.02

50

0.02

382

0.02

24

0.01

938

0.01

641

0.01

375

0.00

962

0.00

681

0.00

363

80.

014

480.

014

620.

014

450.

014

270.

014

180.

014

110.

013

940.

013

630.

013

190.

012

070.

010

830.

009

550.

007

270.

005

480.

003

1510

0.00

879

0.00

899

0.00

896

0.00

885

0.00

879

0.00

872

0.00

866

0.00

856

0.00

840

0.00

793

0.00

736

0.00

672

0.00

545

0.00

433

0.00

267

140.

004

170.

004

180.

004

140.

004

140.

004

110.

004

080.

004

060.

004

040.

004

010.

003

900.

003

730.

003

540.

003

110.

002

650.

001

84



F!r ,%" is the 2D anisotropy function. The subscript X in thegeometry and radial dose functions indicates the point !re-place X by P" or line !replace X by L" source approximation.A detailed description of the above parameters can be foundin the updated AAPM TG-43 U1 protocol.11

II.E. Calculations of primary and scatterdose separation

The primary and scatter dose separation !PSS" formalismpublished by Russell et al.32 can be used in convolution/superposition methods33 to calculate dose distributionsaround brachytherapy sources in heterogeneous media. ThisPSS formalism has been recently used to characterize dosi-metric properties of HDR 192Ir and 169Yb sources.34 PSSdose formalism is based on the separation of energy deposi-tion into primary and scattered dose components in a givenphantom material. Therefore, this model is based on the as-sumption that the main mechanisms for energy loss from theradiation field of photons leaving the source capsule are welldescribed by the primary and scattered exponential portionsof the attenuation. Thus, the total dose is considered to be thesuperposition of the primary and scattered components suchthat Dtotal=Dprim+Dscat. Following this PSS formalism a pho-ton is considered as a primary photon when it leaves thesource capsule without taking account if it has previouslyinteracted inside the source, that is, every photon that es-capes from the capsule in the outward direction is a primaryphoton. When this photon makes its first interaction in thephantom material, it is considered as single scatter until itmakes another interaction when it will be considered as amultiple-scattered photon. In the present study, the primary,single-scatter and multiple-scatter photon contributions todose have been scored separately. The energy spectrum ofthe primary photons as defined above leaving the capsule hasbeen also binned as described by Taylor and Rogers.34

III. RESULTS AND DISCUSSIONIII.A. TLD measurements and Monte Carlocalculations of dose rate distributionsin Solid Water

Dosimetric characteristics of the IPL 137Cs source havebeen determined by TLD measurements and Monte Carlosimulation and the uncertainties of the final results with thetwo dosimetric techniques were obtained following theTG-43 U1 recommendation.11 Table I shows the error propa-gation for the TLD measurement and Monte Carlo data usedin these investigations. It should be noted that in this table,the values with * sign were not utilized for derivation ofthe total uncertainties in experimental radial dose functionand 2D anisotropy function. The reason for this exclusionwas the fact that these parameters are relative quantities andthey are independent of the uncertainties of source calibra-tion and energy correction of the TLD response. Therefore,the underlined italic-bold values are those that have beenused for error bars of the radial dose function and 2D aniso-tropy function.

The dose rate constant of the IPL 137Cs source TLD mea-sured and MC simulated in the experimental Solid Waterphantom were found to be 0.930!!7.3%" and0.928!!2.6%" cGy h1 U1, respectively. The small differ-ence between both values reflects the accuracy of the sourceand phantom geometries used in Monte Carlo simulationmethod. Therefore, this Monte Carlo simulation in water !de-scribed in Sec. III B" could be used for clinical application ofthis source.

The radial dose function of the IPL 137Cs source was mea-sured and calculated in Solid Water using the linear sourceapproximation, gL!r". Figure 2 compares the TLD measuredand MC calculated radial dose functions of the IPL sourcemodel.

The Monte Carlo derived 2D anisotropy function of theIPL source in Solid Water for angles ranging from 0 to90 were obtained by averaging the values from the fourquadrants around the source and are shown in Fig. 3. Takinginto account the TLD uncertainties, both TLD measurementsand Monte Carlo calculation are compatible. The agreementbetween these two methods validates the proper selection ofthe source design, dimensions, and composition materials,particularly the thickness of the end caps, in the simulationtechnique.

III.B. Monte Carlo calculations of dose ratedistributions in liquid water

Dose rate distribution in liquid water of the IPL sourcewas obtained using Monte Carlo techniques as described inSec. II C. Table II shows the dose rate distribution for thenew source model in water in an along-and-away look-uptable.

The Monte Carlo simulated dose rate constant ofthe IPL source in water was found to be "

TABLE III. Monte Carlo calculated radial dose functions for the IPL source.These data were fitted to a fifth order polynomial with coefficients a0=1.007 65, a1=7.073#103, a2=5.680#104, a3=1.376#105, a4=1.117#106, and a5=1.608#108 between 0.25 and 20 cm.

Distance r!cm" gL!r" !L=1.48 cm"

0.25 1.0070.5 1.003

0.75 1.0021.0 1.0001.5 0.9962 0.9913 0.9814 0.9705 0.9576 0.9437 0.9288 0.912

10 0.87612 0.83615 0.77220 0.657



=0.948!!2.6%" cGy h1 U1. In addition, the simulated ra-dial dose function !calculated for L=1.48 cm" of this sourcein water is presented in Table III. Moreover, a fifth orderpolynomial fit to these data has been introduced for theirclinical applications in the treatment planning systems. Thecoefficients of this polynomial fit are presented in Table III.The Monte Carlo simulated 2D anisotropy function of thesource model in liquid water is shown in Table IV at radialdistances of distances ranging from 0.25 to 20 cm.

III.C. Primary and scatter dose separation data

Primary and scatter dose rate data are tabulated at 12 ra-dial distances ranging from 0.25 to 20 cm and from 0 to180 polar angles with minimum resolutions of 1, 2, 5,and 10 depending on the angle interval.35 Figure 4 indicates

the separation of the primary, single, and multiple scatteredand the total dose portion of the IPL 137Cs source in liquidwater. In addition, the energy-weighted photon spectrum forthis source model is shown in Figure 5.35

IV. CONCLUSIONSDosimetric characteristics of the IPL 137Cs source !model

67-6520" have been obtained following the AAPM TG-43U1 formalism using TLD and Monte Carlo methods. Theagreement between the measured and Monte Carlo calcu-lated values indicated the proper selection of design, dimen-sions, and composition material of both the source and thephantom in the Monte Carlo simulations. Using the validatedMonte Carlo simulation technique, the clinically recom-mended TG-43 U1 parameters for this source model have

TABLE IV. The anisotropy function F!r ,%" of the IPL source. The origin is at the geometric center of the source !not at the active center" and the origin of thepolar angle is at the tip side of the source !a color coded aluminum ring defines the source tip".

% !deg"

r

!cm"

0.25 0.5 0.75 1 1.5 2 3 4 5 6 7 8 10 12 15 20

0 0.912 0.903 0.899 0.899 0.898 0.905 0.910 0.904 0.910 0.923 0.920 0.9241 0.916 0.907 0.906 0.906 0.908 0.911 0.915 0.915 0.919 0.927 0.927 0.9362 0.917 0.912 0.910 0.911 0.913 0.914 0.917 0.921 0.922 0.927 0.932 0.9423 0.917 0.914 0.910 0.909 0.911 0.913 0.915 0.918 0.921 0.925 0.932 0.9394 0.918 0.912 0.905 0.904 0.906 0.909 0.911 0.913 0.919 0.925 0.929 0.9365 0.917 0.909 0.901 0.900 0.902 0.904 0.907 0.910 0.916 0.923 0.928 0.9366 0.913 0.903 0.897 0.898 0.900 0.903 0.907 0.910 0.915 0.922 0.929 0.9378 0.904 0.898 0.898 0.900 0.905 0.908 0.910 0.914 0.920 0.925 0.932 0.94010 0.911 0.907 0.907 0.910 0.914 0.918 0.920 0.923 0.928 0.933 0.940 0.94515 0.968 0.944 0.940 0.937 0.937 0.940 0.941 0.943 0.944 0.948 0.950 0.954 0.95720 1.004 0.981 0.966 0.960 0.957 0.958 0.959 0.960 0.961 0.961 0.964 0.965 0.967 0.96825 1.014 1.002 0.988 0.978 0.974 0.971 0.971 0.971 0.971 0.971 0.971 0.972 0.974 0.975 0.97630 1.009 1.000 0.993 0.985 0.983 0.980 0.978 0.979 0.979 0.979 0.979 0.980 0.981 0.981 0.98240 1.003 0.999 0.997 0.993 0.992 0.990 0.990 0.990 0.989 0.989 0.989 0.989 0.990 0.989 0.99050 1.002 0.999 0.998 0.996 0.996 0.995 0.994 0.995 0.994 0.994 0.994 0.994 0.994 0.995 0.99460 1.005 1.001 0.999 0.999 0.999 0.999 0.998 0.998 0.998 0.998 0.998 0.997 0.997 0.997 0.998 0.99770 1.002 1.001 1.000 1.000 0.999 1.000 1.000 0.999 0.999 1.000 0.999 0.999 0.999 0.999 0.999 0.99980 1.000 1.000 0.999 1.000 1.000 1.000 1.000 1.000 1.001 1.000 1.000 1.000 0.999 1.000 1.000 0.99990 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1100 1.001 1.000 1.000 1.000 0.999 1.000 1.000 1.000 1.000 1.000 1.000 0.999 0.999 1.000 1.000 1.000110 1.002 1.001 1.000 1.000 0.999 1.000 0.999 0.999 1.000 0.999 0.999 0.999 0.999 0.999 0.999 0.999120 1.004 1.001 1.000 0.999 0.998 0.999 0.997 0.997 0.998 0.997 0.997 0.997 0.997 0.998 0.998 0.997130 1.001 0.999 0.998 0.997 0.996 0.996 0.994 0.995 0.994 0.994 0.994 0.994 0.995 0.994 0.994140 1.004 0.999 0.997 0.992 0.992 0.990 0.989 0.990 0.989 0.990 0.989 0.989 0.990 0.990 0.990150 1.009 1.001 0.993 0.985 0.983 0.980 0.979 0.980 0.980 0.980 0.980 0.980 0.981 0.982 0.983155 1.014 1.001 0.988 0.977 0.974 0.971 0.971 0.971 0.972 0.972 0.972 0.973 0.975 0.976 0.977160 1.004 0.981 0.964 0.960 0.958 0.958 0.960 0.961 0.961 0.962 0.964 0.965 0.967 0.970165 0.967 0.945 0.941 0.939 0.940 0.942 0.944 0.946 0.947 0.950 0.953 0.956 0.960170 0.917 0.915 0.915 0.916 0.920 0.922 0.926 0.927 0.932 0.937 0.943 0.947172 0.914 0.909 0.908 0.910 0.913 0.916 0.919 0.922 0.927 0.931 0.939 0.945174 0.928 0.918 0.911 0.910 0.912 0.915 0.918 0.920 0.925 0.930 0.936 0.941175 0.936 0.927 0.918 0.917 0.917 0.918 0.920 0.922 0.927 0.931 0.936 0.942176 0.942 0.934 0.925 0.923 0.924 0.925 0.926 0.927 0.933 0.935 0.939 0.947177 0.943 0.939 0.934 0.931 0.931 0.930 0.931 0.933 0.936 0.938 0.942 0.949178 0.943 0.939 0.937 0.935 0.935 0.935 0.935 0.938 0.939 0.941 0.944 0.948179 0.940 0.935 0.931 0.932 0.934 0.934 0.933 0.936 0.939 0.939 0.943 0.951180 0.932 0.929 0.922 0.925 0.929 0.927 0.928 0.930 0.937 0.930 0.939 0.959



been determined in liquid water. In addition, a tabulated doserate distribution in an along-and-away format is provided forbenchmark of a treatment planning system or utilization in aquick hand calculation purpose. Moreover, the dose distribu-tions around this source have been parametrized with thePSS formalism. The primary and scattered dose rate tables aswell as the energy-weighted photon spectrum are providedfor treatment planning systems based on convolution/superposition methods.

ACKNOWLEDGMENTSOne of the authors !A.S.M." would like to thank Rebecca

F. Duncan, Venkata Rachabatthula, and Curtis Baker for theirvaluable contributions during the TLD dosimetry procedures.This study was supported in part by Radiation Products andDesign, Inc. and by the Generalitat Valencia !Project No.Prometeo/2008/114" and the Spanish Ministerio de Ciencia yTecnologa !Project Nos. FPA200765013-C02-01 andFPA2006-12120-C03-02".

a"Author to whom correspondence should be addressed. Electronic mail:[email protected]; Telephone: !516"562-2491; Fax: !516"562-1592.

1C. A. Prez and L. W. Brady, Principles and Practice of Radiation On-cology, 3rd ed. !Lippincott Williams and Wilkins, Philadelphia, 1998".

2D. R. Shearer, Recent Advances in Brachytherapy Physics !AmericanInstitute of Physics, New York, 1981".

3P. W. Grigsby, J. F. Williamson, and C. A. Prez, Source configurationand dose rates for the Selectron afterloading equipment for ginecologicapplicators, Int. J. Radiat. Oncol., Biol., Phys. 24, 321327 !1992".

4The GEC ESTRO Handbook of Brachytherapy, edited by A. Gerboulet, R.Poetter, J.-J. Mazeron, H. Meertens, and E. Van Limbergen !ESTRO,Leuven, 2002".

5L. L. Meisberger, R. J. Keller, and R. J. Shalek, The effective attenuationin water of the gamma rays of gold-198, Ir-192, Cs-137, Ra-226, andCo-60, Radiology 90, 953957 !1968".

6V. Krishnaswamy, Dose distributions about Cs-137 sources in tissue,Radiology 105, 181184 !1972".

7J. F. Williamson, Monte Carlo and analytic calculation of absorbed dosenear Cs-137 intracavitary sources, Int. J. Radiat. Oncol., Biol., Phys. 15,227237 !1988".

8R. Waggener, J. Lange, J. Feldmeier, P. Eagan, and S. Martin, Cs-137dosimetry table for asymmetric source, Med. Phys. 16, 305308 !1989".

9J. F. Williamson, Monte Carlo based dose rate tables for the AmershamCDCS.J and 3M Model 6500 Cs-137 tubes, Int. J. Radiat. Oncol., Biol.,Phys. 41, 959970 !1998".

10R. Nath, L. L. Anderson, G. Luxton, K. A. Weaver, J. F. Williamson, andA. S. Meigooni, Dosimetry of interstitial brachytherapy sources: Recom-mendations of the AAPM Radiation Therapy Committee Task GroupNo.43, Med. Phys. 22, 209234 !1995".

11M. J. Rivard, B. M. Coursey, L. A. DeWerd, W. F. Hanson, M. SaifulHuq, G. S. Ibbott, M. G. Mitch, R. Nath, and J. F. Williamson, Update ofAAPM Task Group No. 43 Report: A revised AAPM protocol for brachy-therapy dose calculations, Med. Phys. 31, 663674 !2004".

12E. Casal, F. Ballester, J. L. Lluch, J. Perez-Calatayud, and F. Lliso,Monte Carlo calculations of dose rate distribution around the AmershamCDCS-M-type Cs-137 source, Med. Phys. 27, 132140 !2000".

13F. Ballester, J. L. Lluch, Y. Limami, M. A. Serrano, E. Casal, J. Prez-Calatayud, and F. Lliso, A Monte-Carlo investigation of the dosimetriccharacteristics of the CSM11 137Cs source from CIS, Med. Phys. 27,21822189 !2000".

14J. Prez-Calatayud, F. Lliso, F. Ballester, M. A. Serrano, J. L. Lluch, Y.Limami, V. Puchades, and E. Casal, A Monte-Carlo study of dose ratedistributions around the specially asymmetric CSM3-a Cs-137 source,Phys. Med. Biol. 46, N169N173 !2001".

15J. Prez-Calatayud, D. Granero, E. Casal, F. Ballester, and V. Puchades,

Monte Carlo and experimental derivation of TG-43 dosimetric param-eters for CSM-type Cs-137 sources, Med. Phys. 32, 2836 !2005".

16J. Prez-Calatayud, F. Ballester, J. L. Lluch, M. A. Serrano-Andrs, E.Casal, V. Puchades, and Y. Limami, Monte Carlo calculation of dose ratedistributions around the Walstam CDC.K-type 137Cs sources, Phys. Med.Biol. 46, 20292040 !2001".

17J. Prez-Calatayud, F. Ballester, M. A. Serrano-Andrs, J. L. Lluch, V.Puchades, Y. Limami, and E. Casal, Dosimetric characteristics of theCDC-type miniature cylindrical 137Cs brachytherapy sources, Med.Phys. 29, 538543 !2002".

18J. Prez-Calatayud, D. Granero, F. Ballester, V. Puchades, and E. Casal,Monte Carlo dosimetric characterization of the Cs-137 selectron/LDRsource: Evaluation of applicator attenuation and superposition approxima-tion effects, Med. Phys. 31, 493499 !2004".

19L. Liu, S. C. Prasad, and D. A. Bassano, Determination of Cs-137 do-simetry parameters according to the AAPM TG-43 formalism, Med.Phys. 31, 477483 !2004".

20J. Prez-Calatayud, D. Granero, F. Ballester, E. Casal, R. Cases, and S.Agramunt, Technical note: Monte Carlo derivation of TG-43 dosimetricparameters for radiation therapy resources and 3M Cs-137 sources, Med.Phys. 32, 24642470 !2005".

21S. Meigooni, D. M. Gearheart, and K. Sowards, Experimental determi-nation of dosimetric characteristics of Best 125I brachytherapy source,Med. Phys. 27, 21682173 !2000".

22A. S. Meigooni, Z. Bharucha, M. Yoe-Sein, and K. Sowards, Dosimetriccharacteristics of the Best double-wall 103Pd brachytherapy source,Med. Phys. 28, 25682575 !2001".

23A. S. Meigooni, H. Panth, V. Mishra, and J. F. Williamson, Instrumen-tation and dosimeter-size artifacts in quantitative thermoluminescence do-simetry of low-dose fields, Med. Phys. 22, 555561 !1995".

24S. Agostinelli et al., Geant4a simulation toolkit, Nucl. Instrum. Meth-ods Phys. Res. A 506, 250303 !2003"; see also http://geant4.web.cern.ch/geant4/.

25Z. Li, R. K. Das, L. A. DeWerd, G. S. Ibbott, A. S. Meigooni, J. Perez-Calatayud, M. J. Rivard, R. S. Sloboda, and J. F. Williamson, Dosimetricprerequisites for routine clinical use of photon emitting brachytherapysources with average energy higher than 50 keV, Med. Phys. 34, 3740!2007".

26J. Perez-Calatayud, D. Granero, M. Rivard, C. Melhus, M. Pujades, and F.Ballester, Evaluation of electronic equilibrium conditions near brachy-therapy sources, Med. Phys. 35, 2971 !2008".

27D. Cullen, J. H. Hubbell and L. Kissel, EPDL97: The evaluated photondata library, 97 version, Lawrence Livermore National Laboratory Re-port No. UCRL-LR-50400, 6, Rev. 5 !1997".

28J. F. Williamson, Monte Carlo evaluation of kerma at a point for photontransport problems, Med. Phys. 14, 567576 !1987".

29A. S. Meigooni, S. B. Awan, N. S. Thompson, and S. A. Dini, UpdatedSolid Water to water conversion factors for 125I and 103Pd brachy-therapy sources, Med. Phys. 33, 39883992 !2006".

30C. S. Melhus and M. J. Rivard, Approaches to calculating AAPM TG-43brachytherapy dosimetry parameters for 137Cs, 125I, 192Ir, 103Pd, and169Yb sources, Med. Phys. 33, 17291737 !2006".

31J. Prez-Calatayud, D. Granero, and F. Ballester, Phantom size inbrachytherapy source dosimetric studies, Med. Phys. 31, 20752081!2004".

32K. R. Russell, A. K. Carlsson-Tedgren, and A. Ahnesj, Brachytherapysource characterization for improved dose calculations using primary andscatter dose separation, Med. Phys. 32, 27392752 !2005".

33J. F. Williamson, R. S. Baker, and Z. Li, A convolution algorithm forbrachytherapy dose computation in heterogeneous geometries, Med.Phys. 18, 12561265 !1991".

34R. E. P. Taylor and D. W. O. Rogers, EGSnrc Monte Carlo calculateddosimetry parameters for 192Ir and 169Yb brachytherapy sources, Med.Phys. 35, 49334944 !2008".

35See EPAPS Document No. E-MPHYA6-36-054910 for the numerical val-ues of the parameters and functions of TG-43 U1 formalism for the IPL inEXCEL spreadsheet format. The primary and scatter photon dose contri-butions as well as the energy-weighted photon spectrum are also pro-vided. For more information on EPAPS, see http://www.aip.org/pubservs/epaps.html.



tg-43

Documents