Dose compensation of the total body irradiation therapy

Applied Radiation and Isotopes 55 (2001) 623–630

Dose compensation of the total body irradiation therapy

Jao-Perng Lina, Tieh-Chi Chua,*, Mu-Tai Liub

aDepartment of Nuclear Science, National Tsing Hua University, 101, Section 2 Kuang Fu road, Hsin Chu, Taiwan 300, ROCbDepartment of Radiation Oncology, Changhua Christian Hospital, Changhua, Taiwan 500, ROC

Received 13 February 2001; accepted 12 April 2001

Abstract

The aim of the study is to improve dose uniformity in the body by the compensator-rice and to decrease the dose tothe lung by the partial lung block. Rando phantom supine was set up to treat bilateral fields with a 15 MV linear

accelerator at 415 cm treatment distance. The experimental procedure included three parts. The first part was thebilateral irradiation without rice compensator, and the second part was with rice compensator. In the third part, ricecompensator and partial lung block were both used. The results of thermoluminescent dosimeters measurements

indicated that without rice compensator the dose was non-uniform. Contrarily, the average dose homogeneity with ricecompensator was measured within 75%, except for the thorax region. Partial lung block can reduce the dose which thelung received. This is a simple method to improve the dose homogeneity and to reduce the lung dose received.

The compensator-rice is cheap, and acrylic boxes are easy to obtain. Therefore, this technique is suitable for morestudies. r 2001 Elsevier Science Ltd. All rights reserved.

Keywords: Total body irradiation; Tissue compensator; Rando phantom; Thermoluminescent dosimeter

1. Introduction

Twelve years after Roentgen discovered X-rays, totalbody irradiation (TBI) was mentioned in the literature.

TBI is used for two purposes: one is for total malignantcell destruction and the other is for immunosuppression.TBI complements intensive chemotherapy, which is also

designed to achieve those objectives.There are many different ways of setting up a patient

for TBI treatment. For example, different energies, a

special treatment room (Lam et al., 1979; Galvin et al.,1980; Khan et al., 1980; Miralbell et al., 1994), two ormore simultaneous sources (Webster, 1960), a fixed

beam under which the patient moves (Quast, 1985) orkeeps a reclining position with the knees bent (Galvinet al., 1980; Khan et al., 1980), and AP/PA vs. bilateral

field (Lam et al., 1979; Shank, 1983; Breneman et al.,1990) may be used.

A major consideration of TBI treatments is the dosehomogeneity in the body. As for the homogeneity level,

the nominal 10% homogeneity was suggested as areasonable goal (Lam et al., 1979). Variations in thepatient’s body thickness and the presence of inhomoge-

neous tissues, particularly in the lungs, make it difficultto achieve a uniform dose over the whole body. In theareas of the patient’s irregular surface, the shape of the

isodose curves will be affected. Compensators areneeded to reduce the dose in the areas where thepatient’s body is thinner, for example, the head, neck,

lower legs, and feet (Galvin et al., 1980; Khan et al.,1980; Dyk, 1983; Shank, 1983).

The effectiveness of TBI is limited by the normaltissue tolerance of the lung, liver and kidney. The lung is

of particular concern. Since lungs have low tolerance toradiation and because pneumonitis (Dyk et al., 1981a, b)caused by radiation can result in the death of the patient,

it is imperative that the dose to the lung should beprecisely controlled. In the thorax the dose is often

*Corresponding author. Department of Nuclear Science,

National Tsing Hua University, 101, Section 2 Kuang Fu road,

Hsinchu 300, Taiwan, ROC. Tel.: +886-3-5710340; fax: +886-

3-572-7310.

E-mail address: [email protected] (T.-C. Chu).

0969-8043/01/ -see front matter r 2001 Elsevier Science Ltd. All rights reserved.

PII: S 0 9 6 9 - 8 0 4 3 ( 0 1 ) 0 0 1 2 9 - 4

excessive due to the low-density lung tissue where thebeam traverses. However, partial lung blocks for some

treatments can also reduce the dose to the lungs (Quast,1987; Miralbell et al., 1994; Ho et al., 1998).

In this work, we used bilateral fields with 15MV

energy and placed tissue compensators-rice all over thebody in order to improve dose uniformity. Partial lungblocks were used to reduce the lung dose received. Thebilateral field technique permits us to treat at distances

(415 cm) which are long enough to obtain sufficientportal sizes to encompass the total body. Throughoutthe procedure Rando phantom could lie in a comfor-

table and supine position.

2. Methods and materials

2.1. Patient positioning and field arrangement

Irradiation is bilateral with 15 MV at a source-to-surface distance (SSD) of 415 cm. The treatment field of

40� 40 cm2 at 1m would project the diagonal diameterof 234 cm at the treatment distance, and the collimator isrotated 451. The field size is large enough to suit mostadult patients.

The patient is supine and the photon beam is directedhorizontally across the treatment room (Fig. 1). Beforetreating each patient, a number of measurements were

made with a Rando phantom (Alderson ResearchLaboratories, Stamford, CT). This phantom simulatesa male body and consists of three sections: head and

neck, thorax and pelvis with upper thighs, but thephantom lacks two arms. The lung material in thethorax was made of epoxy, which is a resin-based lung

substitute material. In this work, emphasis is given onthe dose distribution of Rando phantom.

2.2. Studies design

An acrylic box was designed to place the phantom andthe compensator-rice in it. Fig. 1 depicts the geometry of

the acrylic box and the phantom. Its dimensions are200 cm long, 60 cm wide, and 60 cm high. The thickness

of the acrylic box is 1 cm. The phantom was coveredslightly by the rice in order to simulate the real situation:a patient should breathe freely in the acrylic box with

rice. The phantom was compensated to a uniform depthby rice. Treatment practice is to apply bolus suitablysuch as to obtain a constant value for the dosecalculated at midplane. The phantom was given 2 Gy

to the phantom midline at the umbilicus of the pelvis,which is the prescribed point.

The experimental procedure included three parts. The

first part is the bilateral irradiation without ricecompensator, and the second part is with rice compen-sator. In the third part, both rice compensator and

partial lung block are used. Materials of lung blockconsist of bismuth, lead and tin. The attenuationcoefficient of that block is 0.4085 cm�1. The lung blocks

are fixed by a plastic tray, which is then positioned onthe gantry head. The size and place of lung block wasobtained by Pinnacle3 treatment planning system.Computed tomography scans (1 cm slice spacing) were

used and the data were transferred into planning system.Lithium fluoride (LiF) thermoluminescent dosimeters(TLDs, TLD-100) (3.1� 3.1� 0.89mm3) were loaded

into the phantom at each body section and used todirectly evaluate the dose distribution to midline in thephantom.

2.3. Dosimetry

The extreme conditions under which TBI is deliveredcreate a unique set of problems in terms of determiningthe dose. Standard dosimetric measurements (theequivalent thickness of acrylic relative to rice, central

axis depth dose, beam profile, and beam output factor)were used to calculate the monitor unit (MU) at theprescribed point.

First, because the whole phantom was fully compen-sated with rice and beam would cross the acrylic boxfirstly (Fig. 1), the relation of rice and water (soft tissue)

has to be obtained in order to calculate the dose. Thevalues are measured by an ion chamber at the thickness5, 10, 15, 20 cm rice and acrylic separately. The ratio

from their reading values was taken to obtain theequivalent thickness of acrylic relative to rice. Second,because the acrylic box was set at a constant distance(SSD 415 cm), the absorbed doses in the acrylic box

which varied with depth would be measured. Depth dosealong the central beam axis was measured in a30� 30 cm2 acrylic phantom at 415 cm SSD, and the

depth was from 0.7 to 50 cm in the acrylic box. Third,adequate coverage on the patient with a reasonably flatbeam is one of the commonest considerations. The beam

profile was scanned horizontally at the points in themidplane in the acrylic at 415 cm SSD. Finally, the beam

Fig. 1. Opposed lateral field and the phantom with rice in the

acrylic box.

J.-P. Lin et al. / Applied Radiation and Isotopes 55 (2001) 623–630624

output factor was obtained in an acrylic phantom at itsdepth of 5 cm at SSD 415 and 100 cm, separately, to

compare the difference between the two. All measure-ments employed a Farmer ion chamber (type 2581) ofvolume 0.6 cm3.

The MU at a prescribed point can be calculated usingthe equation:

MU ¼Dose

OP�DD�BF�LF; ð1Þ

where OP is the beam output factor at 415 cm, DD thedepth dose along the central beam axis, BF the backscatter factor, and LF the lateral scatter factor. Since theoutput factor is measured by acrylics phantom, it must

be divided by BF and LF to achieve the full scatter.The absorbed doses in the phantom are calculated as

follows:

Dc ¼ DpðPDDÞcðPDDÞp

fpfc

� �2

OAR; ð2Þ

where the Dc and Dp are the calculated point and theprescribed dose, respectively. fp and fc are the distancesof target to the prescribed point and to the calculatedpoint. OAR is the off-axis ratio of the prescribed point.

The absorbed dose in the patients who receive TBIwas simulated by the use of an anatomical phantom inconjunction with TLD for direct measurement. The

values measured by TLD were compared with thatcalculated at the point of measurement.

3. Results

3.1. Dosimetry

Such beam profiles provide the data for dosimetric

calculations and the adequate coverage of the treatmentfield. The relative distribution of the beam is shown inFig. 2. At 90 cm from the midpoint in the horizontal

direction, the beam intensity is reduced to 88% of themaximum value. In common treatment field size(180 cm) the off-axis ratio was approximately 0.9.

Since the beam passed through 1 cm thick acrylic, thecentral axis depth dose in acrylic was measured (Fig. 3curve B). The central axis depth dose of water phantomat SSD 415 cm could not be measured due to the

limitation of the machine used. Therefore, it should be

measured in the water phantom at SSD 100 cm (Fig. 3curve C). These measured results were compared withthe data transferred from 100 cm SSD to 415 cm SSD by

Mayneord factor (Fig. 3). The depth dose at a depth of30 cm measured in acrylic (curve B) was 6% lower thanin water phantom transferred by the Mayneord factor(curve A). The difference resulted from the mass energy

absorption coefficient of water and acrylic and from thescatter condition between acrylic and water phantom.

The results of thickness of rice are 5, 10, 15, 20 cm

relative to the corresponding equivalent thickness ofacrylic 4.95, 9.66, 14.01, 18.17 cm. The distance of theprescribed point measured to the surface of the acrylic

box is 31 cm. According to these data and Fig. 3, weknow that the equivalent depth of rice is 32.2 cm, this isclose to the measured value (31 cm). Rice is therefore a

suitable compensator of equivalent tissue.For 415 cm SSD, the measured output at 5 cm depth

in a phantom was 4% lower than the values computedby the inverse square law from standard 100 cm data.

For extended distances, the measured values were lowerthan the predicted from the inverse square law. Thereason may be air scatter. Consequently, the output

should be measured at the treatment distance.The phantom was given 2Gy at the prescribed point.

The predicted MU, taking the output factor, depth dose

and phantom thickness into account but assuming ahomogeneous medium, is tabulated in Table 1 for

Fig. 2. Horizontal beam profiles at midplane measured at

415 cm.

Table 1

Monitor units and treatment time to deliver 2Gy at treatment distance

Technique of irradiation Irradiation with

non-compensator

Irradiation with

compensator-rice

Irradiation with compensator-

rice and lung block

Monitor dose rate (MU/min) 500 500 500

Monitor unit (MU) 3994 6100 6192

Treatment time (min) 8 12.2 12.4

J.-P. Lin et al. / Applied Radiation and Isotopes 55 (2001) 623–630 625

different techniques. The monitor unit of irradiation

with compensation is larger than that without it, and thetreatment time is also longer. The MUs of irradiationwith compensation and lung block are slightly morethan that with non-lung block, which resulted from

plastic tray.

3.2. Dose distribution in phantom

The precision for our method using TLD was estimatedto be within 75%. The dose distribution for the bilateralfield technique was generated from the average TLD

measurements. The distribution is shown in Fig. 4.

Fig. 3. Comparison depth doses. Curve C is the DD measured at SSD 100 cm (water phantom). Curve A is the DD at SSD 415 cm

transferred from Curve C by the Mayneord factor. Curve B is the DD measured at 415 cm (acrylic).

Fig. 4. Dose distribution in a Rando phantom without compensator and with compensator. The data are normalized to 100% in the

prescribed point.


The measured dose uniformity in the non-compen-sated phantom at midplane is correspondent with the

phantom thickness data. The dose near the headand neck is higher than that at the prescribed pointbecause of the small thickness of the phantom. The high

dose in the lung is due to its lower tissue density.As expected the dose uniformity in compensatedphantom without lung tissue was quite good, i.e., inthe order of 75% of prescribed mid-line dose.

Particularly, the dose of head and neck with compen-sator reduced to 5% less than that without compensator.Moreover, the lung dose received with compensator is

less than that without compensator. However, the lungregion shows an increase of about 12% relative to theprescribed mid-line dose. It is because the reduced

equivalent thickness would result in a higher dose inthe lung.

A comparison between the difference of the measured

data and the calculated data is shown in Fig. 5. The neckdoses are higher than what we calculated and they are

higher probably due to the presence of an air cavity. Inthe thorax region, we have to determine the lung density

for the Rando phantom. The steps are as follows: first,outline the lung on the CT images. Second, use the‘‘track cursor’’ in the treatment planning system.

Through CT images, we have an average lung densityof 0.35 g/cm3 for Rando phantom. Then we adjusted thethickness of lung to the equivalent thickness of water.The difference between the measured data and the

calculated ones is within 710% (most data are within5%), and the calculated data are lower than themeasured ones. There is qualitative agreement between

the prescribed dose and the measured dose.The results from treatment planning system agreed

well with the measured dose by TLD. Since our system

was used for the calculation, the input data for thecomputer program had been measured at 100 cm SSD.The tissue maximum ratio and off axis ratio should be

verified to produce more accurate calculation at theextended treatment distance used in TBI.

Fig. 5. Measured and calculated doses from the midplane axis in a phantom. The midplane pelvis dose is given 200 cGy: (a) without

compensator, (b) with compensator.


3.3. Lung shielding block

Since the electron density of each lung slice isdifferent, we measured every point of each lung slice to

evaluate the lung dose received. The average dose tothe lung with compensator measured was 2.39Gy. Themaximum (slice 18) and minimum (slice 10) doses to the

lung should be considered. The design of lungblock’s thickness is referred to the average lung dose(2.39 Gy). If we should limit the lung dose approxi-mately to 1.95 or 1.30 Gy, the thickness of lung block

should be 0.5 or 1.5 cm. Lung blocks for bilateralposition have to be verified by the port films (Fig. 6).Fig. 7 shows the dose distributions of maximum

lung dose slice, and Fig. 8 shows the relative dose with0.5 and 1.5 cm block shielding. Measurements showthat a thickness of 0.5 cm block provides approximately

75% transmission; however, this was predicted as 82%broad beam transmission. Meanwhile, a block ofthickness 1.5 cm provides 61% transmission, which

was expected to be 54%. In fact, the measureddose should be larger than the calculated dose becauseof the scatter effect. But the thinner block (0.5 cm)cut by hand may result in an error for the lower

transmission.

4. Discussion

Most magna-field radiotherapy procedures areperformed with AP/PA fields, although some institu-

tions use a lateral field for TBI. Since the thicknessof the body in the AP configuration is more uniformthan in the lateral position, a more homogeneous

dose distribution can be expected (Dyk, 1983). ForAP/PA treatment, some authors (Glasgow et al.,1988; Breneman et al., 1990; Miralbell et al., 1994) havedesigned and used special TBI patient supports.

They are not easily amenable to most radiation therapycenters, however. Our technique took rice as tissueequivalent compensation with bilateral irradiation,

and we designed an acrylic box to contain patientand rice only. It is easier than to design a specialmachine.

In our study, we measured the dose distribution inphantom only and used that data as coming from realpatients to evaluate the dose distributions. However,

real patients are more complex than the phantom. So theabsorbed dose is verified at some institutions (Ribaset al., 1998) by measurement made for patients withdiodes or TLD placed at the entrance and exit side of

body. In addition, real patients would breathe and

Fig. 6. A verification film of the phantom and lung shielding block. (The solid line is the contour of lung, and the dotted line is region

of block.).


move opposite to the Rando phantom’s immovability.Therefore, a real patient should be immobilizated and

then be set up into the acrylic box.

When a prescribed tumor dose is well above the lungtolerance, the dose to lung will have to be reduced tominimize the probability of lung complications. The

simplest procedure to reduce the dose to lung is to uselung shield blocks (El-Khatib and Valsourt, 1989;Miralbell et al., 1994). Our method for lung density

used CT data to determine the lung density of eachpatient for calculating the dose delivery. Dyk et al.(1981a, b) reported that the accuracy of 73% can be

achieved easily by this method.However, some institutions (Galvin et al., 1980) have

not shielded the lungs because of the problems of placingtissue compensators in this region: any error in the

position could result in low dose. The port films allowedthe reproducibility of the technique to be analyzed bymatching the port. Miralbell et al. (1994) suggested the

deviations within 5mm in the horizontal axis and within10mm in the vertical axis were reasonable. In addition,some centers not only use partial transmission lung

blocks but also give an additional electron boost to theshielded segments in order to deliver a uniform dose tothe whole bone marrow (Miralbell et al., 1994).

Treatment planning system provides the reference

information of dose distribution in the phantom or inpatients with compensator. If we want to obtain theaccurate organ dose, Planskoy et al. (1996) suggested that

the beam data, measured under TBI conditions, wereavailable in order to use the treatment planning system forTBI treatment planning. However, these were necessary (i)

to overcome the field size and SSD limitations imposed bythe TPS and (ii) to provide an adequately fine grid of datapoints for the large TBI fields.

5. Conclusion

The procedure described for TBI is both simple andpractical and does not require a specialized facility. Dose

Fig. 7. Dose distributions for the Rando phantom. The

maximum slice of lung dose with non-block shielding irradia-

tion. The mean dose is (a) 247.3 cGy for non-block shielding;

(b) 186.7 cGy for 0.5 cm block shielding; (c) 150 cGy for 1.5 cm

block shielding.

Fig. 8. Relative doses in a Rando phantom with compensator

and 0.5, 1.5 cm thickness block. Data are normalized to 100%

in the prescribed point.


measurement in phantom with compensator-rice indi-cates that 75% uniformity is attainable throughout the

body. Partial lung shielding will improve dose unifor-mity and it may be necessary to reduce the incidence ofradiation pneumonitis in high dose protocols. More-

over, very great comfort for the patient during the longtreatment time is necessary when low dose rates areemployed. A reasonably short setup and treatment timewould avoid overloading of the radiation oncology

department.

References

Breneman, J.C., Elson, H.R., Little, R., 1990. A technique for

delivery of total body irradiation for bone marrow

transplantation in adults and adolescents. Int. J. Radiat.

Oncol. Biol. 18, 1233–1236.

Dyk, V.J., 1983. Magna-field irradiation: physical considera-

tions. Int. J. Radiat. Oncol. Biol. 9, 1913–1918.

Dyk, V.J., Battista, J.J., Rider, W.D., 1981. The use of

computed tomography in lung dose calculations for upper

half-body irradiation. Br. J. Radiol. 15, 90–97.

Dyk, V.J., Keane, T.J., Kan, S., Rider, W.D., Fryer, C.J.H.,

1981b. Radiation pneumonitis following large single dose

irradiation: a re-evaluation based on absolute dose to lung.

Int. J. Radiat. Oncol. Biol. 7, 461–467.

El-Khatib, E.E., Valsourt, S., 1989. Calculation of lung

shielding for total body irradiation. Int. J. Radiat. Oncol.

Biol. 17, 1099–1102.

Galvin, J.M., D’Angio, G.J., Walsh, G., 1980. Use of tissue

compensators to improve the uniformity for total body

irradiation. Int. J. Radiat. Oncol. Biol. 6, 767–771.

Glasgow, G.P., Wang, S., Stanton, J., 1988. A total body

irradiation stand for bone marrow transplant patient. Int. J.

Radiat. Oncol. Biol. 16, 875–877.

Ho, A., Kishel, S., Proulz, G., 1998. Partial lung shield for TBI.

Med. Dosimetry 23, 299–301.

Khan, F.M., Williamson, J.F., Sewchand, W., Kim, T.H., 1980.

Basic data for dosage calculation and compensation. Int. J.


Lam, W.C., Lindskoug, B.A., Order, S.E., Grant, D.G., 1979.

The dosimetry of Co-60 total body irradiation. Int. J.


Miralbell, R., Rouzaud, M., Grob, E., Nouet, P., Bieri, S.,

Majno, S.B., 1994. Can a total body irradiation technique

be fast and reproducible? Int. J. Radiat. Oncol. Biol. 29,

1167–1173.

Planskoy, B., Bedford, A.M., Davis, F.M., Tapper, P.D.,

Loverock, L.T., 1996. Physical aspects of total-body

irradiation at the Middlesex hospital, UCL group of

hospitals, London 1988–1993: I Phantom measurements

and planning methods. Phys. Med. Biol. 41, 2307–2326.

Quast, U., 1985. Physical treatment planning of total-body

irradiation: patient translation and beam-zone method.

Med. Phys. 12, 567–574.

Quast, U., 1987. Total body irradiationFreview of treatment

techniques in Europe. Radiother. Oncol. 9, 91–106.

Ribas, M., Jornet, N., Eudaldo, T., Carabante, D., Duch,

M.A., Ginjaume, M., 1998. Midplane dose determination

during total body irradiation using in vivo dosimetry.

Radiother. Oncol. 335, 91–98.

Shank, B., 1983. Techniques of magna-field irradiation. Int. J.

Radiat. Oncol. Biol. 9, 1925.

Webster, E.W., 1960. Physical considerations in the design of

facilities for uniform whole-body irradiation in man.

Radiology 75, 19–32.


Documents

Dose compensation of the total body irradiation therapy