Pergamon

Medical Dosimetry, Vol. 20, No. 1. pp. l-5, 1995 Copyright 0 1995 American Association of Medical Dosimetrists

Printed in the USA. All rights reserved 09X3947/95 $9.50 + .OO

0958-3947( 94)00049-2

THE INCORPORATION OF PARTIAL SHIELDING OF THE SPINAL CORD IN A TISSUE DEFICIT COMPENSATOR IN

RADIOTHERAPY OF THE THORAX

DOUGLAS JONES, ’ DONALD CHRISTOPHERSON, ’ DAVID JUDD, ’ LAURA ESAGUI, 2

MARK D. HAFERMANN, M.D., 2 and JOHN W. RIEKE, M.D.2 ‘Northwest Medical Physics Center, 21031 67th Ave. W., Lynnwood, WA 98036-7306, USA; and

‘Radiation Oncology Section, Virginia Mason Medical Center, 1100 Ninth Ave., Seattle, WA 98101, USA

Abstract-A method to determine the shape of a patient by placing radiopaque wires and chains on the skin and taking two isocentric X-ray films is described. The wire locations are reconstructed by X-ray stereo photogrammetry, and a beam’s eye view of the wire frame structure can be obtained with reference to the original setup of the “stereo-pair” films. An algorithm for paving between the wires with triangular plates is described which allows the calculation of the tissue deficit distance and compensator thickness. The depth and distance to points on the spinal cord are calculated, and the dose rate is calculated using a standard irregular field computation program. The limit for spinal cord tolerance is specified in terms of the maximum daily dose based on an equivalent dose formula. The additional thickness of compensator, required for the posterior field compensator to satisfy the tolerance limit, is calculated. The technique readily accommodates the kyphotic and scoliotic spine and has been in routine clinical use for seven years.

Key Words: Compensators, Lung cancer, Myelopathy.

INTRODUCTION

Radiotherapy of lung cancer is a common procedure that is generally accomplished by directing beams ver- tically down and up onto a supine patient. Initial fields will treat both the primary site and involved lymphatics in the mediastinum and supraclavicular fossa. The sep- aration of the patient varies considerably, resulting in a large, typically > 15% variation in the dose rate at midplane through the treatment volume. Various au- thors have devised schemes to modify the dose rate, usually from the anterior field only, using tissue deficit compensators; the method described by Purdy et al.’ is typical of this approach.

The spinal cord is invariably within the irradiated volume of the initial fields, and the prescription usually exceeds what is customarily considered spinal cord tolerance. Thus, spinal cord blocking must be intro- duced during the course, and we previously derived a basis for this.’ An alternative to this is to insert a transmission block at the outset, reducing the fractional dose to the cord to a satisfactory level. Orton et a1.3 have discussed the merits of this approach, which are considered to be a radiobiological and a quality assur- ance advantage. Bagne et als4 have presented a method to produce transmission blocks covering the spinal cord in the treatment of lung cancer.

The technique described in this article combines these two ideas, wherein we produce anterior and pos- terior tissue deficit compensators and the posterior compensator is modified to include a transmission

block to limit the cord dose to an acceptable level. The importance of the inclusion of a compensator for the posterior field and evaluation of dose to the spinal cord as part of the treatment plan is exemplified in this article by Luka and Marks,5 who describe an instance of radiation myelopathy.

MATERIALS AND METHODS

The reconstruction of a surface shape from the images of metal wires on X-rays has been described by us,’ and application of this idea to produce tissue deficit compensators in radiotherapy of the intact breast has been described previously.7 While the patient is in a sitting position, thin metal wires are taped to the back, and these run in a head-to-toe direction. Normally five are used separated by about 5 cm, and the placement is not critical. The patient then lies down on the simulator couch, the treatment field is localized using standard techniques, and field simulation films are produced. Then metal chains of various link patterns are placed on the anterior skin, and films are taken at standard gantry angles of from 0” to 30”. The angles chosen are not critical, although we have found that separations of greater than 20” are required to minimize inaccura- cies in this stereo X-ray photogrammetry process. A typical pair of films is shown in Fig. 1. The images of the wires and chains are digitized onto a computer, from which the reconstruction is made. A lateral film is also produced from which the location of points along the spinal cord is determined and correlated to

Medical Dosimetry Volume 20, Number 1, 1995

Fig. 1. A typical pair of isocentric films produced at standard gantry angles of 0 & 30”, illustrating the appearance of the chains on the front and back of the patient.

nAolIfE: u1sooo5 couol IN 0.0 a4 cum lustn 0.0 cn CuJmw o.oo( cam IblTATIoN 180 DES SMRY iBorEs ccu.mTm 180 ES 510 - 100.0 FIELD LEN6ltc 29.5 YIDllt 21.5 SuFcE TO TRAY OIST . n.4 w OEPiH * 10.5 PSI - 09.5 EOIJIV ML5 OF CC@ - 7.7 CUPENSAWS IN YI Bass ~111 be on STWCB rids of tray

MXlN CLINIC PATIENR SHV JAY. Pi&TIN? 145B27 Rm1c1Nt la Pimnm 03-Jk-91 FuNst cd co!NENmA/PLlm

..--.. _. -.._ -

Fig. 2. A plot showing the reconstruction of the chains in a beam’s-eye-view of the anterior field and the array of brass chips required for tissue deficit compensation.

Partial shielding of spinal cord 0 D. JONES er al.

RF mm - 10.5 SSD- 09.4 EuJxvoD6wco9- 7

Fig. 3. A beam’s-eye-view of the posterior field. The location of the spinal cord is shown as a series of points superimposed on the brass array required for tissue deficit compensation.

the location of the chains. A computer program, COM- PEN, provides a beam’s eye view of the chains, points along the cord. and calculates the depth and distance to the source of each point. It also provides plots of the array of brass chips required for tissue deficit com- pensators for both the anterior and posterior fields. The anterior compensator plot is shown in Fig. 2, and Fig. 3 shows the posterior. The significant thickness of brass toward the superior end of the field (labeled “head”) should be noted. The brass chip size for the anterior compensator is typically 9 mm on the side when 12 mm are used in the posterior compensator with 6-mm chips covering the spinal cord region, where that reso- lution is required. From this information and an outline of the field, the dose rate at each point is calculated.

The prescription specifies the number of fractions and cord tolerance level, from which the maximum daily cord dose is calculated using formulas previous described by US.~ The contribution of the anterior and posterior field to the dose along the spinal cord is evaluated. The attenuation of the dose from the poste- rior field, to limit the spinal cord dose, is readily calcu- lated and the attenuator thickness required to achieve this is computed. This thickness is added to the brass thickness required for tissue deficit compensator thick- ness along the spinal cord.

The thickness of the additional attenuator will de- pend on the prescription, and occasionally only 1 or 2 mm are added (this is difficult to see and verify on a port film). Our practice is to take port films including the compensators, but we also expose a film in the absence of the patient at the same source to film distance as the port film. The location of the thickened section of the compensator is marked, overlaid on the port film, and used to verify that the attenuator is aligned accu- rately to the spinal cord.

The efficacy of the method was evaluated by mea- surements of the dose distribution in the coronal plane of the curvaceous phantom shown in Fig. 4. The phantom consists of a plastic (ABS) shell 3 mm thick which is filled with water. A 0. l-cc ionization chamber is moved remotely to any point within the phantom. A 25 x 25 cm field generated by a 4-MV linac was directed up, and the phantom was set at 80 cm TSD with the central axis at midline between the breasts.

RESULTS

The isodose distribution in a coronal plane 6 cm deep with and without the compensator in the beam is shown in Fig. 5. A 40% range in dose over the area studied is reduced to 12% with the compensator. The

4 Medical Dosimetry Volume 20, Number 1. 1995

Fig. 4. Two views of the phantom used to evaluate the efficiency of the tissue deficit compensators.

Partial shielding of spinal cord 0 D. JONES er a/. 5

Fig. 5. Isodose distribution in a coronal plane at 6 ems depth. The solid lines show the uncompensated beam and the dose ranges from maximum 119 to minimum 79 over the 22 X 10 cm area studied. The dotted lines show

the &close distribution with a compensator when the maximum was 102 and minimum 90 in the plane.

use of a simple exponential calculation with a single value for linear attenuation coefficient was also evalu- ated with the phantom described, and it was shown that errors of less than 2% in dose delivery result from this approach even when the brass thickness over the cord is thickened by 6 mm.

CONCLUSION

The advantage of using tissue deficit compensa- tors and the transmission blocking of the cord is funda- mentally a quality assurance issue, and the setup is exactly the same on the first and last days of treatment. The problems associated with “block-back” schemes to equalize the total dose in all parts of the irradiated volume are well known. The possibility that the inser- tion of a spinal cord block would be overlooked is a problem. In principle, such blocks should be inserted at different times at various points along the cord to reach the tolerance level, and this is not practical.

REFERENCES

1. Purdy, J. A.; Keys, D. J.; Zivnvska, F. A compensator filter for chest portals. Inr. J. Radiar. Oncol. Biol. Phys. 2:1213-1215; 1977.

2. Jones, D.; Hafermann, M. D.; Richardson, R. G. A basis for blocking the spinal cord in opposed field irradiation. Int. J. Ra- diar. Oncol. Biol. Phys. 11:627-630; 1985.

3. Orton, C. G.; Herskovic, A. M.; Ezzell, G. A.; Spick, J. T.; Vitaslis, T. Transmission blocks: Clinical and biological ratio- naies. Inr. J. Radiar. Oncol. Biol. Phys. 11:2155-2158; 1985.

4. Bagne, F. R.; Bronn, D. G.; Fadell, L. A. The use of a novel dose-conformational transmission blocking system in the treat- ment of lung cancer (abstract). Inr. X Radiar. Oncol. Biol. Phys. 15126; 1988.

5. Luka, S.; Marks, J. E. Beam attenuators and the risk of unrecog- nized large-fraction irradiation of critical tissues. Med. Dosim. 19:15-21; 1994.

6. Christopherson, D. A.; Jones, D. A general approach to the location of radiopaque objects, with applications in radiotherapy. Phys. Med. Biol. 12:1581-1986; 1984.

7. Jones, D. A technique for the precise, uniform irradiation of breast cancer (abstract). Inr. J. Radiar. Oncol. Biol. Phys. 9(Suppl. 1):155; 1983.