Verification of Ir-192 near source dosimetry using GAFCHROMIC® filmSou-Tung Chiu-Tsao, Tamara L. Duckworth, Neil S. Patel, Julianna Pisch, and Louis B. Harrison Citation: Medical Physics 31, 201 (2004); doi: 10.1118/1.1637733 View online: http://dx.doi.org/10.1118/1.1637733 View Table of Contents: http://scitation.aip.org/content/aapm/journal/medphys/31/2?ver=pdfcov Published by the American Association of Physicists in Medicine Articles you may be interested in Measurement of absorbed dose-to-water for an HDR 192Ir source with ionization chambers in a sandwich setup Med. Phys. 40, 092101 (2013); 10.1118/1.4816673 Influence of phantom material and dimensions on experimental I 192 r dosimetry Med. Phys. 36, 2228 (2009); 10.1118/1.3121508 Optimization of deterministic transport parameters for the calculation of the dose distribution around a high dose-rate I 192 r brachytherapy source Med. Phys. 35, 2279 (2008); 10.1118/1.2919074 Calculation of dose decrease in a finite phantom of a Ir 192 point source Med. Phys. 34, 3943 (2007); 10.1118/1.2775664 Approaches to calculating AAPM TG-43 brachytherapy dosimetry parameters for Cs 137 , I 125 , Ir 192 , Pd 103, and Yb 169 sources Med. Phys. 33, 1729 (2006); 10.1118/1.2199987

Verification of Ir-192 near source dosimetry using GAFCHROMIC ® filmSou-Tung Chiu-Tsao,a) Tamara L. Duckworth,b) Neil S. Patel,c) Julianna Pisch,d) andLouis B. Harrisone)

Department of Radiation Oncology, Beth Israel Medical Center and St. Luke’s-Roosevelt Hospital Center,New York, New York 10003

~Received 29 April 2003; revised 7 October 2003; accepted for publication 11 November 2003;published 9 January 2004!

Intravascular brachytherapy treatments of in-stent restenosis have been performed extensively usingIr-192 ribbon. Task Group 60 of the American Association of Physicists in Medicine~AAPM!recommends a dose reference point at 2 mm from the source center for these treatments. However,it is known that the source can be as close as 0.5 mm to the arterial wall if not centered in thelumen. Therefore, the source dosimetry needs to be characterized at these close distances to accu-rately determine the amount of dose delivered for noncentered cases. In this paper, we report theverification of the dose distributions around Ir-192 seed sources at radial distances from 0.5 mm to6 mm using GAFCHROMIC® film. We evaluated an Ir-192 single seed source and a train of 6 seedsspaced 1 mm apart enclosed in a nylon ribbon. Each source was placed in a homogeneous solidwater phantom directly below a stack of GAFCHROMIC® films ~MD-55-2!. The calibration curveof the lot of films used in the experiment was established for Ir-192 by exposing a set of calibrationfilms, one at a time, to an Ir-192 high dose rate~HDR! source. All films were scanned 5 or moredays after exposure with a Lumisys Model 150 microdensitometer. The data were acquired andevaluated using RIT113~Radiological Imaging Technology! software and analyzed using Excel andIDL ~Interactive Data Language! software. Isodose curve plots in the plane containing the source’slongitudinal axis and dose rate plots in the radial direction were obtained. For both configurations,the dose rates along the transverse axes agree to within the margin of error with previous MonteCarlo results. The isodose curve plots display hot spots near the seed ends, which is consistent withthe leakage of beta particles and electrons from the unsealed seed ends as predicted with MonteCarlo calculations. ©2004 American Association of Physicists in Medicine.@DOI: 10.1118/1.1637733#

Key words: Ir-192, GAFCHROMIC film, dosimetry, intravascular brachytherapy, beta leakage,photon

I. INTRODUCTION

Ir-192 seed ribbons have been used extensively in conven-tional brachytherapy of tumors and intravascular brachy-therapy~IVBT ! of in-stent restenosis.1–3 The American As-sociation of Physicists in Medicine~AAPM! Task Group 60~TG-60! recommends a reference dose point at 2 mm fromthe source center for IVBT treatments. However, since Ir-192ribbon is not actively centered in the arterial lumen duringthe treatment, the closest distance between the lumen walland source center can be 0.5 mm. It is, therefore, crucial tostudy the dosimetry of the source in the submillimeter rangebeginning at the surface of the ribbon containing the Ir-192seeds. Due to the steep dose gradient in this close distancerange, it has been very challenging to perform accurate dosemeasurement in high spatial resolution. There have been re-ports of Monte Carlo calculations and GAFCHROMIC filmdosimetry.4–8 So far, there is no published report on verifi-cation measurements of doses for Ir-192 ribbons, particularlyat very close distances down to 0.5 mm. The purpose of thiswork is to characterize the dosimetry of the Best IndustriesIr-192 single seed and multiple seed train at very close dis-tances to the source using GAFCHROMIC film. This data

will be useful in determining the amount of dose actuallygiven to arterial walls during intravascular brachytherapyprocedures performed using these sources.

In addition to photon contributions, even higher doses dueto leakage of beta particles and conversion electrons near theunencapsulated ends of each Ir-192 seed have been observed.At a radial distance of 0.5 mm, the doses from beta particlesand electron near the gap between neighboring seeds werefound to be nearly twice as large as the photon doses near thecenters of the seeds6 using Monte Carlo calculations. In thiswork, we also verify the hot spots near the gaps betweenseeds using GAFCHROMIC film dosimetry.

There are several dosimeters used for various purposes inmedical physics: ionization chambers, semiconductors, ther-moluminescent detectors~TLDs!, and films. GAFCHROMICfilm9 is an appropriate dosimeter for intravascular brachy-therapy because it provides superior spatial resolution. Thisis an important consideration when characterizing sourcesused in intravascular brachtherapy due to the steep dose gra-dients near the source. Also, GAFCHROMIC films are tissueequivalent with an effectiveZ of about 6.5. They are insen-sitive to room light so they can be easily handled. They areinitially transparent and darken in color upon exposure to

201 201Med. Phys. 31 „2…, February 2004 0094-2405 Õ2004Õ31„2…Õ201Õ7Õ$22.00 © 2004 Am. Assoc. Phys. Med.

ionizing radiation; therefore, chemical processing of thesefilms is not required.

There are many different models of GAFCHROMIC®

film made commercially available to date: HD810, MD55-1,MD55-2, HS and XR-T/R from International Specialty Prod-ucts ~ISP, Wayne, New Jersey!. Model HD810 is approxi-mately 0.1 mm thick, but its useful dose range is 50–2500Gy. Although the recommended dose range for HD810 filmis above the clinically relevant dose range, it has been usedto verify the dose distributions of intravascular brachy-therapy sources.10 MD55-1 has a more clinically relevantdose range of 3–100 Gy, but this film model was discontin-ued in September 1994. The most commonly usedGAFCHROMIC® film model is MD55-2, which has a rec-ommended dose range of 3–100 Gy and was used in all theexperiments described in this paper. There has been exten-sive research conducted determining the precision ofGAFCHROMIC® film model MD55-2 for dosimetric mea-surements for sources of different geometries andenergies.7,8,11–13After we completed the measurement usingMD55-2 films in this study, new models HS and XR becameavailable and they have higher sensitivity than the HD810and MD55-2 models.

GAFCHROMIC® films can be read out with densitom-eters. There are spot densitometers in which either the lightsource is stationary and the film is translated or in which thefilm is stationary and the light source moves. There are alsotwo-dimensional imaging systems in which a uniform lightsource, such as a light box, illuminates the film and an im-aging system measures the transmission. The densitometerused in this work is a scanning laser densitometer, Lumisysmodel 150, based on a HeNe laser.

II. MATERIALS AND METHODS

A. Ir-192 sources and GAFCHROMIC films

The Ir-192 sources used in the experiments are of model81-01, lot #9909I, manufactured by Best Industries~Spring-field, VA!. The single seed source consists of an Ir-192 seed,3 mm length and 0.5 mm diameter, enclosed in a nylon rib-bon @Fig. 1~a!#. The multiple seed source used consists of 6

Ir-192 seeds, spaced 1 mm apart, also enclosed in a nylonribbon @Fig. 1~b!#. The total train length was 23 mm. Theribbon outer diameter was 0.8 mm. The activity per seed wasabout 481 MBq~52 U!, based on the vendor-specified activ-ity that was NIST traceable. We verified the single seed ac-tivity and the total activity of the 6-seed train with a Capintecwell chamber, model CRC15R~Ramsey, New Jersey!. Theactivities of the sources agreed with the vendor-specified ac-tivities within the uncertainty of the calibration factor. Wecould not measure the activity of each individual seed in the6-seed ribbon to determine the actual activity distributionalong the source train. The vendor specified uniformitywithin 5% for the activities of the 6 seeds.

All films used in this study are GAFCHROMIC® films,model MD55-2, lot #940818. Due to the small regions ofinterest, films were cut into appropriate sizes for each set ofexposures. For the exposures to the single seed, 6-seed trainand calibration high dose rate~HDR! source, the sizes of thecut films were 2 cm32 cm, 3 cm33.5 cm, and 3 cm34.5cm, respectively. The film thickness was specified as~240615! mm by the manufacturer~International Specialty Prod-ucts, Wayne, New Jersey! and verified with Vernier calipermeasurement. The two active emulsion layers~16 mm each!were sandwiched by three clear polyester bases with twodouble-sided adhesive tapes. For the purpose of this set ofmeasurements, we have approximated the detector at thecenter of the film thickness.

B. Calibration film irradiation

For irradiating the calibration films, an Ir-192 HDRsource~model 105.002, microSelectron-HDR V2, NucletronCorporation, Columbia, Maryland! was used. The Ir-192core has an active length of 3.6 mm and a diameter of 0.65mm by design.16 It is surrounded by a stainless steel encap-sulation that has an outer diameter of 0.9 mm. The source isattached to a delivery wire and is driven remotely into acatheter with an outer diameter of 2 mm.

The RMI solid water phantom~model 457!, manufacturedby Gammex~Middelton, WI!, used in the calibration expo-sures was composed of solid water slabs making the totaldimensions 30 cm330 cm320 cm. The center slab had achannel milled out in the center in which a catheter fitstightly, allowing the HDR source to be driven remotely toexpose a calibration film. The films were placed, one piece ata time, directly above the catheter but separated from it by a10 mm thick slab of solid water@Fig. 2~a!#. The films wereoriented such that the source and catheter run in the down-web direction~along the direction of emulsion coating! onthe film. This method of film placement is consistentthroughout all experiments and minimizes systematic errordue to variations in signal in the cross-web direction~perpen-dicular to the direction of emulsion coating!.

The dose rate delivered to each film center was calculatedusing the TG-43 formalism,2

D5Sk•L•@G~r ,u!/G~r 0 ,u0!#•g~r !•F~r ,u!.

FIG. 1. Geometrical configurations of the Ir-192 sources used in the~a!single seed and~b! 6-seed train experiments. Thick arrows designate thepositions where dose rates were analyzed over 0.5 mm~5 adjacent pixels inthe direction parallel to the source axis!. Black, dark gray, and light grayrectangles show the Ir-192 cores, steel capsules, and nylon ribbons, respec-tively. Air channel and air gaps form the pathways of beta leakage from theseed ends.

202 Chiu-Tsao et al. : Verification of Ir-192 near source dosimetry using GAFCHROMIC 202

Medical Physics, Vol. 31, No. 2, February 2004

In this equation,Sk is defined as the air kerma strength of thesource in unit ofU and is given by 0.109A, whereA denotesthe activity in MBq. At the time the calibration films wereexposed, the HDR source activity was 309 GBq~uncertainty2%!, based on measurement using a well chamber, StandardImaging model HDR1000, calibrated by an ADCL. The doserate constant for this source model is 1.108 c Gy U21 h21

~uncertainty 0.13%! based on Monte Carlo calculation byDaskalovet al.16 The geometry factor,G(r ,u), was calcu-lated using the line source approximation,

G~r ,u!5b/@L•r •sinu#,

whereb is the angle subtended by the active length of thesource,L50.36 cm, andr is the distance from the center ofthe source to the reference point~the center of the film emul-sion!, ~1.11260.046! cm. Finally, g(r ) and F(r ,u), the ra-dial dose function and anisotropy function, respectively, areequal to 1.0 at the reference angle chosen,u590°. The doserate at the center of the calibration film was about 5Gy min21 ~with 6% uncertainty!. The exposure times weused, ranging from 1 to 40 min, resulted in doses up to 200Gy at the film center.

C. Ir-192 seed experiments

1. Single seed

The ribbon containing the single seed was positioned in astraight channel with a diameter of 0.9 mm, milled out of aslab of solid water. A film stack was positioned above the

seed, all surrounded by a solid water phantom, 30 cm330cm320 cm @see Fig. 2~b!#. The experimental films werenumbered from 0 to 9, film 0 being the closest to the sourcesat a distance of 0.057 cm. The following films, 1–7, wereeach an additional 0.024 cm from the closest film. Threeseparate experiments were performed on three different days.The first experiment involved four films in a stack, labeled0a–3a, using an exposure time of 38 min~correspondingwith 32 U h!. There were eight films, labeled 0b–3b and4–7, in a stack in the second experiment, using an exposuretime of 2.1 h~98 U h!. Films 8 and 9@not shown in Fig. 2~b!#were exposed for 2 h and 15 min~109 U h! with separationfrom the source by a 2 mmthick slab of solid water.

2. Multiple seed

The setup for the multiple seed experiments was identicalto that for the single seed@Fig. 2~b!#. The films were exposedin two separate experiments on the same day. The films inthe first stack, named 0–7, were exposed for 34 min~172U h!. Films 8 and 9 were irradiated next using 383 U h.

D. Film scanning and data acquisition

All films, calibration and experimental, were scanned atleast 5 days post irradiation with a Lumisys model 150 den-sitometer. The Lumisys 150 used a 2 mWHeNe laser with awavelength of 633 nm and had a set of six lenses that allowa variable spot size ranging from 100mm to 420mm. Thescanner was run using a software package designed by Ra-diological Imaging Technology, RIT 113 vs 3.07. This soft-ware initiated film scanning, conducted data acquisition andhelped perform quality assurance measurements. The scan-ning resolution was set to 100mm. The signal values weresaved inASCII format and exported into Excel.

For the calibration curve, the minimum signal value oneach calibration film, corresponding to the maximum dose,was obtained from the area on the film centered on the HDRsource. This signal value was converted into optical density~OD!. The background OD, obtained from an unirradiatedfilm piece, was subtracted from the OD data to obtain the netoptical density~NOD!. NOD vs Dose~Gy! for each of the 9calibration films was plotted in Excel and fit with a thirdorder polynomial.

The signal values from the areas of interest on the filmsfrom the Ir-192 seed experiments, 3 cm33 cm for singleseed and 3 cm33.5 cm for the multiple seed train, wereconverted to doses using the established calibration curve.Dividing by the exposure time~h! and source air kermastrength~U!, we obtained dose rate values. The dose ratedata from the films, in theX–Z plane, were converted tothose in ar –Z plane using the equation,r25X21d2, whered corresponds to the vertical distance between the film emul-sion and the source center,X is the horizontal coordinateperpendicular to the seed axis, andr is the radial distance foreach data point@see Fig. 2~b!#. At each radial distancer, thedose rates were average over 5 adjacent pixels~correspond-ing with 0.5 mm! aligned parallel to the source axis andcentered at the perpendicular bisector of the single seed@Fig.1~a!#, and then averaged over the left and right side of each

FIG. 2. Schematic diagrams of the set-up~a! for the calibration film expo-sures by Ir-192 HDR source, and~b! for Ir-192 single seed or multiple seedtrain exposures. Source and film are surrounded by 30 cm330 cm320 cmof solid water. In~a!, each calibration film is positioned, one at a time, abovethe HDR source at a distance of 1.112 cm between source center and filmemulsion layer. In~b!, a stack of 8 films is shown placed above the source.

203 Chiu-Tsao et al. : Verification of Ir-192 near source dosimetry using GAFCHROMIC 203

Medical Physics, Vol. 31, No. 2, February 2004

film. Grand averages over dose rates from films in the twosets, a and b, were further obtained for films 0–3. The aver-age dose rates were plotted against the radial distance, forcomparison with those generated from Monte Carlo calcula-tions.

For the 6-seed train, the conversion from theX–Z planeto ther –Z plane was also performed to yieldD (r,Z) val-ues. The average over 5 pixel lines at each of the perpendicu-lar bisectors of the two central seeds@lines A and B in Fig.1~b!# and averaged over the left and right sides were obtainedfor each film. These averaged dose rates were plotted vsradial distances from the seed train center. Isodose curveswere plotted using IDL software, v. 5.4.

III. RESULTS

A. Calibration curve

The Ir-192 calibration curve is displayed in Fig. 3. Thefilm response is nonlinear. The data were fit with a third-order polynomial. This equation was used to convert the dataon the experimental films from NOD to dose. The equation is

Dose ~Gy)59.2704* ~NOD!329.3574* ~NOD!2

169.319* ~NOD!; ~R250.9998!.

Using this lot of film and Ir-192, one needs to deliver ap-proximately 70 Gy to achieve a NOD equal to unity. Com-bining the uncertainties in the dose rate constant, sourcestrength, the distance from the source to film emulsion, ex-posure time, film uniformity and densitometer readings, theuncertainty in the conversion factor is about 9%~see TableI!.

B. Single seed

Along the perpendicular bisector of the seed, the doserates obtained from all films, 0–9, are plotted against theradial distances,r, in Fig. 4~a!. Previously presented MonteCarlo data6 out tor50.6 cm are also plotted for comparison.Agreement is found within the margin of error. To demon-strate the contributions from the beta leakage from the seedends, the longitudinal dose profiles from films 0~average of0a and 0b! and films 1~average of 1a and 1b! are displayedin Fig. 5. The top line, films 0 atd50.057 cm, shows a sharppeak atZ520.2 cm and a small bump atZ50.2 cm. Thesepeaks are asymmetric and lie just beyond the ends of theseed. The sharp peak has a dose rate of 190 c Gy U21 h21,quite close to the average dose rate along the central 0.5 mmof the source length at this close distance. The beta peaks inthe profile from films 1 atd50.081 cm are still visible, al-though not as prominent. The longitudinal profiles atd50.105 cm~average of films 2a and 2b! and 0.201 cm~fromfilm 6! are also included in the plot. Note the asymmetry ofthe film measurement data in theZ-direction along the seedlength and beyond the seed ends, indicating a variation ofactivity distribution along the seed length and also differentbeta leakage from two ends of a seed. The Monte Carlogenerated dose profiles at the corresponding distances areplotted for comparison. General agreement is noted, exceptfor the heights of the beta peaks.

C. Multiple seed train

Figure 4~b! displays the average dose rates along the per-pendicular bisectors of the two central seeds of the 6-seedtrain @positions A and B from Fig. 1~b!# from all nine films.

FIG. 3. Calibration curve for Ir-192 using MD55-2 GAFCHROMIC® film,lot number 940818, scanned with a Lumisys model 150 microdensitometer,wavelength5633 nm.

TABLE I. Uncertainty analysis for dose calibration.

Quantity Value Uncertainty Percent Combined

Distance between source center and bottom surface of solid water slab 0.1 cm 0.045 cmThickness of solid water slab 1 cm 0.01 cmHalf thickness of film 0.012 cm 0.00075 cmDistance between source center and film emulsion 1.112 cm 0.046 cm 4.15%Dose rate constant, Ref. 16 0.13%Source strength 309 GBq 2%Dose rate 5 Gy min21 6.2%Exposure time~minimum! 60 s 2 s 3.33%Exposure time~maximum! 2400 s 2 s 0.08%Dose~minimum! 5 Gy 7.0%Dose~maximum! 200 Gy 6.2%Film uniformity 5%Densitometer reading 2%Dose conversion factor~low dose point! 8.9%Dose conversion factor~high dose point! 8.2%

204 Chiu-Tsao et al. : Verification of Ir-192 near source dosimetry using GAFCHROMIC 204

Medical Physics, Vol. 31, No. 2, February 2004

These data are in agreement with the values obtained bysummation of doses over 6 seeds based on Monte Carlo data6

for a single seed with air channel in the ribbon at both ends.As shown in Fig. 6, the isodose curves for this 6-seed

train source obtained from the closest film, film 0, display

‘‘hot spots’’ near the gaps in between the seeds and at the endof the seed train. Beyondr50.1 cm, the longitudinal dosedistribution is fairly uniform. The isodose lines are plottedfor the values predicted by Monte Carlo: 35, 15, 9, and 6.5c Gy U21 h21, for r50.05, 0.1, 0.15, and 0.2 cm, respec-tively. The hot spots are due to the beta leakage from theseed ends. The maximum dose rate is as high as 95c Gy U21 h21 and the average dose rate for these five spikes

FIG. 4. Measured dose rates vs radial distances obtained withGAFCHROMIC® film measurements indicated by small diamonds,~a!along the transverse axis of single Ir-192 seed source, and~b! averaged overtwo perpendicular bisectors~arrows A and B in Fig. 1! of the two centralseeds in a 6-seed train. Monte Carlo results, shown by big open circles,including dose contributions from photon, beta, conversion electron, andAuger electron, are also plotted for comparison. Corresponding semilogplots for ~a! and ~b! are shown in~c! and ~d!.

FIG. 5. Longitudinal dose rate profiles plotted along the length of the singleseed for films at 0.057 cm~top line, open diamonds!, 0.081 cm~middle line,open squares!, 0.105 cm~lower line, open triangles!, and 0.201 cm~lowestline, open circles! away from the center of the source. The Monte Carlogenerated values at the corresponding distances are shown by closed sym-bols connected with dotted lines~to guide the eye!.

FIG. 6. Isodose curve plot from film 0 in contact with the seed train withschematic drawing of the multiple seeds~rectangles! enclosed in the ribbon.Line values are~from outside toward the center! 6.5, 9, 15, and 35c Gy U21 hr21. These values are the Monte Carlo predicted values forr50.2, 0.15, 0.1, and 0.05 cm, respectively. Hot spots near the air gaps aredue to beta leakage from the unencapsulated ends of the seeds.

205 Chiu-Tsao et al. : Verification of Ir-192 near source dosimetry using GAFCHROMIC 205

Medical Physics, Vol. 31, No. 2, February 2004

is 67 c Gy U21 h21, which is double of the average dose rates29 c Gy U21 h21 at the central 0.5 mm segments of the indi-vidual seeds.15

IV. DISCUSSION

IVBT is a relatively new therapy modality in the treat-ment of in-stent restenosis. The Food and Drug administra-tion ~FDA! has approved a few delivery systems, which in-clude the Ir-192 seed ribbons studied in this work. While thecurrent dose prescription has been shown effective in treatingrestenosis, it may not be the optimal prescription for all pa-tients. In order to fine tune the dose prescription schemes,which would be based on correlation studies between dosedistributions and clinical outcomes, it is crucial to accuratelydetermine dose characteristics with high precision for allsources used or proposed for use in IVBT.

The near source measurements for both the single seedand the multiple seed sources of Ir-192 demonstrate the im-portance of verification of Monte Carlo calculations usingfilm measurement. At very close distances, applicable toIVBT, doses from the Ir-192 seed train vary greatly along thesource length. High doses at hot spots due to beta leakagefrom the seed ends may be twice as large as the doses nearthe perpendicular bisectors of individual seeds. These find-ings are different from some published Monte Carlodata,4,8,14,17 which did not reflect the increase in dose ratebetween the seeds caused by beta leakage from the unencap-sulated ends of the seeds. Although quantifying the exactratio of beta dose to gamma dose is difficult~due to seed toseed variations!, the drastic increase in doses near the seedends is undeniable and can be clinically significant. RecentMonte Carlo calculations by Patelet al.6,15 and Yanget al.5

demonstrated the high beta doses near the interseed gaps,which were verified in this GAFCHROMIC® film study. Thedifference between the film data and the Monte Carlo resultsmay be explained as follows. A uniform source activity wasassumed in the Monte Carlo calculations, while the actualseed activities can vary up to 5% based on the vendor’sspecifications. In addition, the observed asymmetry~equato-rial and longitudinal! of the doses may be explained by~1!the imperfections of the actual seeds in the end cutting and inthe activity distribution in the core, and~2! the fact that theseeds are not always colinear and seed spacing may not be

TABLE II. Uncertainty analysis for dose rate measurement.

Quantity Value Uncertainty Percent Combined

Film uniformity 5%Densitometer reading 2%Dose conversion factor 8.9%Source strength 5%Exposure time~minimum! 2280 s 2 s 0.09%Dose rate 11.5%

TABLE III. Uncertainty analysis of radial distance determination.

Quantity Value Uncertainty Percent

Distance between source center and bottom 0.045 cm 0.005 cmSurface of film stack or solid water slabThickness of solid water slab 0.2 cm 0.01 cmHalf-thickness of film 0.012 cm 0.00075 cmThickness of film 0.024 cm 0.0015 cmx coordinate perpendicular to the source axis varying 0.01 cmDistance from source center to emulsion of film 0 0.057 cm 0.00506 cm 8.87%Distance from source center to emulsion of film 1 0.081 cm 0.00527 cm 6.51%Distance from source center to emulsion of film 2 0.105 cm 0.00548 cm 5.22%Distance from source center to emulsion of film 3 0.129 cm 0.00568 cm 4.41%Distance from source center to emulsion of film 4 0.153 cm 0.00588 cm 3.84%Distance from source center to emulsion of film 5 0.177 cm 0.00607 cm 3.43%Distance from source center to emulsion of film 6 0.201 cm 0.00625 cm 3.11%Distance from source center to emulsion of film 7 0.225 cm 0.00643 cm 2.86%Distance from source center to emulsion of film 8 0.257 cm 0.01121 cm 4.36%Distance from source center to emulsion of film 9 0.281 cm 0.01131 cm 4.02%

Quantity

Percent uncertainty

Maximum Mimimum

Radial distance,r, from film 0 15% 2%Radial distance,r, from film 1 10% 2%Radial distance,r, from film 2 8% 2%Radial distance,r, from film 3 7% 2%Radial distance,r, from film 4 6% 2%Radial distance,r, from film 5 5% 2%Radial distance,r, from film 6 4.5% 2%Radial distance,r, from film 7 4% 2%Radial distance,r, from film 8 6% 2%Radial distance,r, from film 9 6% 2%

206 Chiu-Tsao et al. : Verification of Ir-192 near source dosimetry using GAFCHROMIC 206

Medical Physics, Vol. 31, No. 2, February 2004

exactly 1 mm within the nylon ribbon~Figs. 5 and 6!.An analysis of the major components of uncertainty that

contribute to the uncertainties in the measured dose rates andradial distances has been carried out. These are listed inTables II and III. For radial distances,r, from each film, thepercent uncertainties vary with the horizontal coordinates,X.The maximum and minimum percent uncertainties are tabu-lated.

Furthermore, a stack of multiple films embedded in asolid water phantom was exposed to Ir-192 seed or seed trainin a dose measurement experiment. BecauseGAFCHROMIC® film emulsion and base materials arenearly tissue equivalent9 with effective atomic number of6.5, the experimental configuration is nearly water-equivalent for the photon contributions, although not identi-cal to water medium used in the Monte Carlo calculations.As for the dose contributions from beta particles and elec-trons, the uncertainty may be higher than estimated.

V. CONCLUSION

The dose distributions for the Ir-192 single seed and mul-tiple seed train used in IVBT have been verified usingGAFCHROMIC® film dosimetry between 0.057 cm and 0.6cm. Agreement within the margin of error between film dataand Monte Carlo predictions has been found. The hot spotsnear the interseed gaps must be taken into consideration inevaluating correlation of doses with clinical outcomes.

ACKNOWLEDGMENT

This work was supported in part by NIH Grant No.RO1-HL57923.

a!Author to whom correspondence should be addressed. Phone:~212!844-8096; FAX: ~212!844-8037; electronic mail: schiutsa@bethisraelny.org

b!Current address: Radiation Oncology Department, New York UniversityMedical Center, New York, New York; electronic mail:td255@yahoo.com

c!Electronic mail: bigpoppaneil@yahoo.comd!Electronic mail: jpisch@bethisraelny.orge!Electronic mail: lharriso@bethisraelny.org1R. Nath, H. Amols, C. Coffey, D. Duggan, S. Jani, Z. Li, M. Schell, C. G.Soares, J. Whiting, P. E. Cole, I. Crocker, and R. Schwartz, ‘‘Intravascu-lar brachytherapy physics: Report of the AAPM Radiation Therapy Com-mittee Task Group No. 60,’’ Med. Phys.26, 119–152~1999!.

2R. Nath, L. L. Anderson, G. Luxton, K. A. Weaver, J. F. Williamson, andA. S. Meigooni, ‘‘Dosimetry of interstitial brachytherapy sources: Rec-

ommendations of the AAPM Radiation Therapy Committee Task GroupNo. 43,’’ Med. Phys.22, 209–234~1995!.

3A. E. Ajani, H. S. Kim, and R. Waksman, ‘‘Clinical trials of vascularbrachytherapy for in-stent restenosis: Update,’’ Cardiovasc. Radiat. Med.2, 107–113~2001!.

4D. Baltas, P. Karaiskos, P. Papgiannis, L. Sakelliou, E. Loeffler, and N.Zamboglou, ‘‘Beta versus gamma dosimetry close to Ir-192 brachy-therapy sources,’’ Med. Phys.28, 1875–1882~2001!.

5N. C. Yang and R. Chan, ‘‘Beta particle and conversion electron effectson Ir-192 dosimetry in intravascular brachytherapy,’’ Med. Phys.28, 1182~2001!.

6N. S. Patel, S. T. Chiu-Tsao, J. A. Shih, H. S. Tsao, and L. B. Harrison,‘‘Effect of conversion, Auger, and primary beta rays on the dose distri-bution of an Ir-192 seed/train used in intravascular brachytherapy,’’ Med.Phys.28, 1182–1183~2001!.

7R. Chan, C. G. Soares, and S. T. Chiu-Tsao, ‘‘GAFCHROMIC film do-simetry for Ir-192 seeds at distances close to the source,’’ Med. Phys.24,1019 ~1997!.

8N. Reynaert, M. V. Eijkeren, Y. Taeymans, and H. Thierens, ‘‘Dosimetryof Ir-192 sources used for endovascular brachytherapy,’’ Phys. Med. Biol.46, 499–516~2001!.

9A. Niroomand-Rad, C. R. Blackwell, B. M. Coursey, K. P. Gall, J. M.Galvin, W. L. McLaughlin, A. S. Meigooni, R. Nath, J. E. Rodgers, andC. G. Soares, ‘‘Radiochromic film dosimetry: Recommendations ofAAPM Radiation Therapy Committee Task Group 55,’’ Med. Phys.25,2093–2115~1999!.

10D. M. Duggan, J. L. Lobdell, and M. C. Schell, ‘‘Radiochromic filmdosimetry of a high dose rate beta source for intravascular brachy-therapy,’’ Med. Phys.26, 2461–2464~1999!.

11N. V. Klassen, L. van der Zwan, and J. Cygler, ‘‘GAFCHROMIC MD-55:Investigated as a precision dosimeter,’’ Med. Phys.24, 1924–1934~1997!.

12J. F. Dempsey, D. A. Low, S. Mutic, J. Markman, A. S. Kirov, G. H.Nussbaum, and J. F. Williamson, ‘‘Validation of a precision radiochromicfilm dosimetry system for quantitative two-dimensional imaging of acuteexposure dose distributions,’’ Med. Phys.27, 2462–2475~1996!.

13T. D. Bohm, D. W. Pearson, and R. K. Das, ‘‘Measurements and MonteCarlo calculations to determine the absolute detector response of radio-chromic film for brachytherapy dosimetry,’’ Med. Phys.28, 142–146~2001!.

14R. Wang and X. A. Li, ‘‘A Monte Carlo calculation of dosimetric param-eters of90Sr/90Y and 192Ir SS sources for intravascular brachytherapy,’’Med. Phys.27, 2528–2535~2000!.

15N. S. Patel, S. T. Chiu-Tsao, Y. Ho, T. Duckworth, J. A. Shih, H. S. Tsao,H. Quon, and L. B. Harrison, ‘‘High beta and electron dose from192Ir:implications for ‘gamma’ intravascular brachytherapy,’’ Int. J. Radiat.Oncol., Biol., Phys.54, 972–980~2002!.

16G. M. Daskalov, E. Loffler, and J. F. Williamson, ‘‘Monte Carlo-aideddosimetry of a new high dose-rate brachytherapy source,’’ Med. Phys.25,2200–2208~1998!.

17S. J. Ye, E. Ishmael, and J. J. Feldmeier, ‘‘Dosimetric characteristics of alinear array ofg or b-emitting seeds in intravascular irradiation: MonteCarlo studies for the AAPM TG-43/60 formalism,’’ Med. Phys.30, 403–414 ~2003!.

207 Chiu-Tsao et al. : Verification of Ir-192 near source dosimetry using GAFCHROMIC 207

Medical Physics, Vol. 31, No. 2, February 2004