- Email: [email protected]
A new non-invasive and relocatable immobilization frame for fractionated stereotactic radiotherapy
Radiotherapy and Oncology 47 (1998) 313–317
A new non-invasive and relocatable immobilization frame for fractionated stereotactic radiotherapy Kyriaki Theodorou a ,*, Constantin Kappas a, Constantin Tsokas b a
b
Medical Physics Department, Medical School, University of Patras, 26500 Patras, Hellas, Greece Department of Fixed Posthodentics and Biomaterials, Dental School, University of Athens, Hellas, Greece Received 1 July 1997; revised version received 27 December 1997; accepted 4 March 1998
Abstract Purpose: A newly developed non-invasive immobilization frame for stereotactic radiotherapy is presented, which is intended to be used for both imaging (computed tomography (CT) and angiography) and radiotherapeutic procedures. Materials and methods: The frame is made of duraluminium so as to be stable and light and it has an elliptical shape. The immobilization is achieved using three stable locations on the patient’s head, i.e. the upper dentition, the nose and the back of the neck. The fixation on the three locations ensures complete immobilization in all directions. Results: The immobilization frame can be fitted as many times as is needed to most heads. In order to assess the accuracy of relocation, repeated fittings on two volunteers and on 22 patients undergoing stereotactic treatment were performed (more than 200 mountings in total), which showed maximum anterior-posterior, inferior-superior and lateral reproducibility in positioning of less than 1 mm in all cases. Conclusions and discussion: The in-house-constructed stereotactic frame is simple to use, easily made, non-invasive, relocatable and well tolerated by the patients, providing the possibility of multiple fractions. The major advantage of using such a non-invasive stereotactic frame is the flexibility in timing the different diagnostic procedures (CT and angiography) as well as providing the possibility to extend the use to large brain lesions (treatment without an additional collimator) where a high precision is also required. It also offers significant labour and cost saving over the invasive frames and the majority of the non-invasive frames. To date, 22 patients with ages varying between 12 and 70 years have been treated using this method. 1998 Elsevier Science Ireland Ltd. All rights reserved Keywords: Stereotactic radiotherapy; Non-invasive; Relocatable frame; Fractionation
1. Introduction Several stereotactic techniques have been developed using conventional linear accelerators and different solutions have been proposed for the head fixation in order to achieve correct and accurate co-ordinate translation from the imaging modalities to the therapeutic unit. A majority of authors present single session irradiation schemes using invasive stereotactic frames for the immobilization of the patient’s head [1,2,4,5,12]. In these cases there is no possibility of removing the frame between the diagnostic and the therapeutic procedures, so they have to be closely scheduled. Furthermore, even when delivering high dose irradiation in stereotactic conditions, a fractionation of the dose over several sessions may be preferable, as the tolerance of * Corresponding author.
normal brain tissues is increased with an increase in the number of fractions [3,8,11]. However, the main drawback for the fractionated stereotactic radiotherapy is the difficulty of accurately repeated matching of the target to the isocenter of the linear accelerator. Some authors have proposed the use of non-invasive head fixation for fractionated stereotactic irradiation with linear accelerators [3,6,8,9] and these are available commercially. The most well-known is the Gill–Thomas stereotactic frame [6,9], which is based on the fixation of the upper dentition and the back of the neck, while an elastic strap provides additional support. In this study, a newly developed non-invasive immobilization frame is presented which has been designed and constructed at the University of Patras. This frame is cheaper than the commercially available frames while still complying with the acceptable relocation accuracy limits and it is more comfortable for the
0167-8140/98/$19.00 1998 Elsevier Science Ireland Ltd. All rights reserved PII S0167-8140 (98 )0 0015-2
314
K. Theodorou et al. / Radiotherapy and Oncology 47 (1998) 313–317
system offers easy insertion and removal of the patient’s head into and out of the system. A support for the back of the head and neck is fixed at the posterior part of this elliptical frame (Fig. 1b). The caudal end of the head and neck support is made of duraluminium so as to be lightweight and highly stable, while the cranial part, which intercepts the CT beam, is made of Plexiglas. Plexiglas was selected to avoid artefacts during the imaging procedure. The anterior upper arc of the frame holds the systems for the upper dentition and the nose immobilization. Both systems, which control and measure the anterior-posterior and cranio-caudal direction movements, are provided with one micrometric system (resolution 0.01 mm) each (Fig. 1c). The upper dentition cast (Fig. 1d) is fixed at the end of a thin metal bar while the nose-shaped cast (Fig. 1e) is fixed on the edge of the Plexiglas support. Both the dental and the nose casts are made of acrylic resin. They easily take the desired shape before solidification and hold this shape with time after solidification. They are customized and constructed some days before the first radiotherapy session at the hospital’s dental department. 2.2. Patient immobilization
Fig. 1. The non-invasive stereotactic frame. (a) Joints for the frame separation into two arcs; (b) neck support; (c) micrometric systems for the upper dentition and nose immobilization; (d) upper dentition cast; (e) noseshaped cast.
patients. The frame is used for both imaging (CT and angiography) and therapeutic procedures. A number of patients with primary and metastatic brain tumours have been treated in a fractionated stereotactic irradiation scheme with good preliminary results in tumour control.
The first fixation of the patient’s head to the stereotactic frame is done in the treatment room by positioning the patient’s head on the head and neck support and separating the two parts of the frame. Then the two parts are locked together and the maxilla and nasion casts are adjusted to the frame in such a way that the head is at the correct inclination with respect to the plane of source rotation (Fig. 2). The adjustment of the two casts to the immobilization frame is done by the dentist. Neither local anaesthesia nor sedation is needed. The readings of the micrometric systems are recorded at the position of complete immobilization. After each diagnostic or treatment procedure the stereotactic frame is
2. Materials and methods 2.1. Technical description of the frame The frame is non-invasive and the immobilization is achieved using three stable locations on the patient’s head, i.e. the upper dentition, the nose and the back of the neck (Fig. 1). The easy insertion of the patient’s head into the frame and the good conformation of the casts to the upper dentition and the nasion immobilize the head completely, minimizing movements in all directions. The frame is made of duraluminium so as to be stable but not heavy and it has an elliptical shape. The anterior-posterior free dimension is 26 cm and the lateral dimension is 23 cm. The frame is divided into two parts connected to each other with two lateral joints, one permanent and one detachable; the posterior immobile arc is attached to the support system and the anterior mobile arc is detached at one of its ends and is rotated around its other end (Fig. 1a). This
Fig. 2. Immobilization of the patient’s head in the Linac.
315
K. Theodorou et al. / Radiotherapy and Oncology 47 (1998) 313–317
removed from the patient’s head. The reproducibility of the treatment set-up is achieved by following the same procedure and adjusting the micrometric systems to the prerecorded readings.
Table 2
AVM position
Mean ± 2SD (mm)
Maximum (mm)
2.3. Localization procedure
Frontal Thalamic
0.6 ± 0.4 0.5 ± 0.2
0.8 0.6
Both the CT scanner and the angiography unit have been provided with specially designed holders for the reception of the stereotactic frame. The holder design ensures the duplication of the horizontal positioning of the stereotactic frame during the imaging procedure. The mounting of the frame to each one of the imaging modalities takes less than 2 min. The patient is then laid on the CT table and immobilized according to the pre-recorded readings of the micrometric systems. This procedure takes less than 2 min and neither local anaesthesia nor sedation is needed. Two different localization boxes, one for each diagnostic unit, are used in conjunction with the immobilization frame during the imaging procedure. The whole imaging procedure at the CT scanner usually takes 30–45 min depending on the case.
a
2.4. Treatment At each irradiation session, the immobilization frame is attached to the floor-stand head support holder in the linac. An extensive quality control follows for the correct position of the frame and the correct positioning of the target to the linac’s isocenter. Afterwards, the patient enters the therapeutic unit and is positioned in the stereotactic frame at the position of complete immobilization by adjusting the micrometric systems to the pre-recorded readings. The localization box is again used in conjunction with the frame for the verification of the target positioning. After confirmation of the patient’s proper position, the localization box is removed and the first irradiating arc is performed. The above control procedure is repeated for each one of the non-coplanar arcs and for each treatment session. The preparation of the equipment takes about 20 min and the patient immobilization takes only 2 min. The total treatment time for six non-coplanar arcs is between 50 min and 1 h.
3. Results The same immobilization frame can be mounted as many times as is needed to most heads. In order to assess the Table 1 Relocation accuracy measured in 15 patientsa Direction of displacement
Mean ± 2SD (mm)
Maximum (mm)
Y-axis Z-axis
0.5 ± 0.4 0.3 ± 0.2
0.7 0.4
a
Eighty-five repositionings in total.
3D relocation accuracy measured in two AVM casesa
Two repositionings for each case.
accuracy of relocation, repeated mountings of the immobilization frame were studied in two volunteers. The volunteers had four tattoo skin-marks in different locations on the head along the path of the central beam axis when the gantry was rotated. They were positioned in the frame several times and each time the deviation of the skin-mark from the central beam axis was measured for each appropriate gantry angle. The measurements were repeated for different table angles. The maximal deviation between two mountings of the frame on the head, extrapolated to central brain structures, was shown to be less than 1 mm. The simulator was used as an additional verification procedure for the precise relocation of the frame. Before each treatment session of real patients, lateral X-ray films were taken. The whole stereotactic apparatus was transferred to the simulator before each session and one lateral projection of the patient’s head was taken in the position of complete immobilization using the same beam set-up as for the irradiation. It was not possible to also take anterior-posterior projections as a floor-stand system is used. The relocation was measured for 15 different patients (85 repositionings) by comparing the Y and Z stereotactic co-ordinates of the target centre with the calculated co-ordinates at the treatment planning session. The results are presented in Table 1. The maximum anterior-posterior and inferior-superior displacements was less than 1 mm in all cases. This procedure was shown to be very useful and has been adopted for all patients. Systematically, after the treatment planning procedure and the calculation of the target co-ordinates, verification of the correct position of the frame assembly is done prior to the first treatment session. This procedure is repeated again one or more times during the treatment course, depending on the total number of sessions, and systematically before the last session. In two AVM cases, it was possible to check the 3D relocation accuracy of the frame by measuring the 3D displacement of the target centre between two repeated mountings using two orthogonal projected angiographic images. The relocation was measured by comparing the X, Y and Z stereotactic co-ordinates of each mounting of the frame with the calculated co-ordinates in the treatment planning session. The results are presented in Table 2.
4. Discussion The in-house-constructed stereotactic frame described
316
K. Theodorou et al. / Radiotherapy and Oncology 47 (1998) 313–317
above is simple to use, easily constructed, non-invasive, relocatable and well tolerated by the patients. Furthermore, the accuracy of relocation provides the possibility of planning the patient using multiple imaging modalities and allows the selection of multiple fraction treatment schemes. The use of cellulose acetate cast systems for non-invasive fractionation presents two major disadvantages, i.e. low relocation accuracy compared with the invasive and some of the non-invasive frames and inconvenience to the patient as some find it claustrophobic. One other non-invasive head fixation method used in the past was presented by Greitz et al. [7], where a plastic helmet and a base-ring were mounted to the head of the patient. However, the total relocation error was reported to be 2–3 mm in the transverse plane and 3–4 mm in the vertical plane and also the method necessitates time for fabrication. Finally, the non-invasive technique of Hariz et al. [8] requires an aware and fully co-operative patient so if this is not the case a general anaesthesia may be required. The immobilization technique presented here has some similarities with the Gill–Thomas stereotactic frame [5,10]. However, it has three key differences as follows: (a) the elastic strap has been replaced by the nose cast which makes the immobilization more convenient to the patient; (b) the addition of the two micrometric systems controls the correct patient immobilization and relocation and assures the accuracy of the method; and (c) it is not necessary for the patient to wear the frame while travelling between any of the diagnostic or therapeutic procedures. For the Gill– Thomas stereotactic frame, measurements of the relocation accuracy reported for four consecutive patients showed an overall mean AP displacement of 1 mm (range 0.7–1.2 mm) and an overall mean lateral displacement of 1 mm (range 0.4–1.6 mm) [6]. From the presented results, a small improvement has been achieved in terms of relocation accuracy. The immobilization frame developed and presented here is relatively cheap to use per patient and it does not take long to customize the frame for each patient. The first fixation of the head takes approximately 30 min and is performed by an experienced dentist, whereas all the other fixations during imaging and therapeutic procedures take only 2 min. The cost of the impression material and the production of the casts is usually less than US$100 (US$67 for the construction of the acrylic dental and nasion casts and around US$30 for the solid silicon used as impression material and the cold acrylic used for the fixation of the casts to the frame).
5. Conclusions The proposed non-invasive immobilization frame presents some advantages over many commercial systems. Experience from 22 patients and more than 200 positionings
showed that a general or partial anaesthesia was never necessary for the patient to endure the immobilization and generally all patients tolerated the frame well without complaint. Furthermore, the insertion of the patient’s head into the frame is very easy and the immobilization procedure is very fast. The major advantage of using such a non-invasive stereotactic frame is the flexibility in timing the different diagnostic procedures (CT and angiography) as well as the freedom to decide when and for how many sessions the stereotactic irradiation will be performed. In addition, the frame can be used not only for stereotactic cases with small brain lesions, but also as an immobilization device for conformal radiotherapy of brain lesions using stationary fields and/or multiple converging irradiation arcs. To date, 22 patients with ages varying between 12 and 70 years have been treated using this method with or without the additional circular collimators and with a multiple fraction scheme.
Acknowledgements The authors express their sincere thanks to A. Saliagas and S. Alvanos for the design and construction of many mechanical parts of the frame and Dr A. Mazal and Dr J.C. Rosenwald of the Institute Curie for their support and advice. This project was supported by the ‘Bodosakis’ Foundation.
References [1] AAPM (American Association of Physics in Medicine) Report No. 54. Stereotactic Radiotherapy. AAPM, 1995. [2] Colombo, F. Stereotactic radiosurgery utilizing a linear accelerator. Appl. Neurophysiol. 48: 133–145, 1985. [3] Delannes, M., Daly, N.J., Bonnet, J., Sabatier, J. and Tremoulet, M. Fractionated radiotherapy of small inoperable lesions of the brain using a non-invasive stereotactic frame. Int. J. Radiat. Oncol. Biol. Phys. 21: 749–755, 1991. [4] Engler, M.J., Curran, B.H., Tsai, J., et al. Fine tuning of linear accelerators accessories for stereotactic radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 28: 1001–1008, 1994. [5] Gaboriaud, G., Mazal, A., Pontvert, D., et al. Radiothe´rapie ste´re´rotaxique a` l’Institute Curie. Bull Cancer/Radiother. 78: 311– 319, 1991. [6] Graham, J.D., Warrington, A.P., Gill, S.S. and Brada, M. A noninvasive, relocatable frame for fractionated radiotherapy and multiple imaging. Radiother. Oncol. 21: 60–62, 1991. [7] Greitz, T., Lax, I., Bergstro¨m, M., et al. Stereotactic radiation therapy of intracranial lesions. Methodological aspects. Acta Radiol. Oncol. 25: 81–89, 1986. [8] Hariz, M.I., Henriksson, R., Lofroth, P., Laitinen, L. and Saberborg, N.E. A noninvasive method for fractionated stereotactic irradiation of brain tumors with linear accelerator. Radiother. Oncol. 17: 57–72, 1990. [9] Kooy, H.M., Dunbar, S.F., Tarbell, N.J., et al. Adaption and verification of the relocatable Gill–Thomas–Cosman frame in stereo-
K. Theodorou et al. / Radiotherapy and Oncology 47 (1998) 313–317 tactic radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 30: 685– 691, 1994. [10] Leksell, L. The stereotactic method and radiosurgery of the brain. Acta. Chir. Scand. 102: 316–319, 1951. [11] Podgorsak, E.B., Souhami, L., Caron, J., et al. A technique for frac-
317
tionated stereotactic radiotherapy in the treatment of intracranial tumors. Int. J. Radiat. Oncol. Biol. Phys. 27: 1225–1230, 1993. [12] Webb, S. The Physics of Three-Dimensional Radiation Therapy. Medical Science Series, Institute of Physics, London, 1993.
A new non-invasive and relocatable immobilization frame for fractionated stereotactic radiotherapy Kyriaki Theodorou a ,*, Constantin Kappas a, Constantin Tsokas b a
b
Medical Physics Department, Medical School, University of Patras, 26500 Patras, Hellas, Greece Department of Fixed Posthodentics and Biomaterials, Dental School, University of Athens, Hellas, Greece Received 1 July 1997; revised version received 27 December 1997; accepted 4 March 1998
Abstract Purpose: A newly developed non-invasive immobilization frame for stereotactic radiotherapy is presented, which is intended to be used for both imaging (computed tomography (CT) and angiography) and radiotherapeutic procedures. Materials and methods: The frame is made of duraluminium so as to be stable and light and it has an elliptical shape. The immobilization is achieved using three stable locations on the patient’s head, i.e. the upper dentition, the nose and the back of the neck. The fixation on the three locations ensures complete immobilization in all directions. Results: The immobilization frame can be fitted as many times as is needed to most heads. In order to assess the accuracy of relocation, repeated fittings on two volunteers and on 22 patients undergoing stereotactic treatment were performed (more than 200 mountings in total), which showed maximum anterior-posterior, inferior-superior and lateral reproducibility in positioning of less than 1 mm in all cases. Conclusions and discussion: The in-house-constructed stereotactic frame is simple to use, easily made, non-invasive, relocatable and well tolerated by the patients, providing the possibility of multiple fractions. The major advantage of using such a non-invasive stereotactic frame is the flexibility in timing the different diagnostic procedures (CT and angiography) as well as providing the possibility to extend the use to large brain lesions (treatment without an additional collimator) where a high precision is also required. It also offers significant labour and cost saving over the invasive frames and the majority of the non-invasive frames. To date, 22 patients with ages varying between 12 and 70 years have been treated using this method. 1998 Elsevier Science Ireland Ltd. All rights reserved Keywords: Stereotactic radiotherapy; Non-invasive; Relocatable frame; Fractionation
1. Introduction Several stereotactic techniques have been developed using conventional linear accelerators and different solutions have been proposed for the head fixation in order to achieve correct and accurate co-ordinate translation from the imaging modalities to the therapeutic unit. A majority of authors present single session irradiation schemes using invasive stereotactic frames for the immobilization of the patient’s head [1,2,4,5,12]. In these cases there is no possibility of removing the frame between the diagnostic and the therapeutic procedures, so they have to be closely scheduled. Furthermore, even when delivering high dose irradiation in stereotactic conditions, a fractionation of the dose over several sessions may be preferable, as the tolerance of * Corresponding author.
normal brain tissues is increased with an increase in the number of fractions [3,8,11]. However, the main drawback for the fractionated stereotactic radiotherapy is the difficulty of accurately repeated matching of the target to the isocenter of the linear accelerator. Some authors have proposed the use of non-invasive head fixation for fractionated stereotactic irradiation with linear accelerators [3,6,8,9] and these are available commercially. The most well-known is the Gill–Thomas stereotactic frame [6,9], which is based on the fixation of the upper dentition and the back of the neck, while an elastic strap provides additional support. In this study, a newly developed non-invasive immobilization frame is presented which has been designed and constructed at the University of Patras. This frame is cheaper than the commercially available frames while still complying with the acceptable relocation accuracy limits and it is more comfortable for the
0167-8140/98/$19.00 1998 Elsevier Science Ireland Ltd. All rights reserved PII S0167-8140 (98 )0 0015-2
314
K. Theodorou et al. / Radiotherapy and Oncology 47 (1998) 313–317
system offers easy insertion and removal of the patient’s head into and out of the system. A support for the back of the head and neck is fixed at the posterior part of this elliptical frame (Fig. 1b). The caudal end of the head and neck support is made of duraluminium so as to be lightweight and highly stable, while the cranial part, which intercepts the CT beam, is made of Plexiglas. Plexiglas was selected to avoid artefacts during the imaging procedure. The anterior upper arc of the frame holds the systems for the upper dentition and the nose immobilization. Both systems, which control and measure the anterior-posterior and cranio-caudal direction movements, are provided with one micrometric system (resolution 0.01 mm) each (Fig. 1c). The upper dentition cast (Fig. 1d) is fixed at the end of a thin metal bar while the nose-shaped cast (Fig. 1e) is fixed on the edge of the Plexiglas support. Both the dental and the nose casts are made of acrylic resin. They easily take the desired shape before solidification and hold this shape with time after solidification. They are customized and constructed some days before the first radiotherapy session at the hospital’s dental department. 2.2. Patient immobilization
Fig. 1. The non-invasive stereotactic frame. (a) Joints for the frame separation into two arcs; (b) neck support; (c) micrometric systems for the upper dentition and nose immobilization; (d) upper dentition cast; (e) noseshaped cast.
patients. The frame is used for both imaging (CT and angiography) and therapeutic procedures. A number of patients with primary and metastatic brain tumours have been treated in a fractionated stereotactic irradiation scheme with good preliminary results in tumour control.
The first fixation of the patient’s head to the stereotactic frame is done in the treatment room by positioning the patient’s head on the head and neck support and separating the two parts of the frame. Then the two parts are locked together and the maxilla and nasion casts are adjusted to the frame in such a way that the head is at the correct inclination with respect to the plane of source rotation (Fig. 2). The adjustment of the two casts to the immobilization frame is done by the dentist. Neither local anaesthesia nor sedation is needed. The readings of the micrometric systems are recorded at the position of complete immobilization. After each diagnostic or treatment procedure the stereotactic frame is
2. Materials and methods 2.1. Technical description of the frame The frame is non-invasive and the immobilization is achieved using three stable locations on the patient’s head, i.e. the upper dentition, the nose and the back of the neck (Fig. 1). The easy insertion of the patient’s head into the frame and the good conformation of the casts to the upper dentition and the nasion immobilize the head completely, minimizing movements in all directions. The frame is made of duraluminium so as to be stable but not heavy and it has an elliptical shape. The anterior-posterior free dimension is 26 cm and the lateral dimension is 23 cm. The frame is divided into two parts connected to each other with two lateral joints, one permanent and one detachable; the posterior immobile arc is attached to the support system and the anterior mobile arc is detached at one of its ends and is rotated around its other end (Fig. 1a). This
Fig. 2. Immobilization of the patient’s head in the Linac.
315
K. Theodorou et al. / Radiotherapy and Oncology 47 (1998) 313–317
removed from the patient’s head. The reproducibility of the treatment set-up is achieved by following the same procedure and adjusting the micrometric systems to the prerecorded readings.
Table 2
AVM position
Mean ± 2SD (mm)
Maximum (mm)
2.3. Localization procedure
Frontal Thalamic
0.6 ± 0.4 0.5 ± 0.2
0.8 0.6
Both the CT scanner and the angiography unit have been provided with specially designed holders for the reception of the stereotactic frame. The holder design ensures the duplication of the horizontal positioning of the stereotactic frame during the imaging procedure. The mounting of the frame to each one of the imaging modalities takes less than 2 min. The patient is then laid on the CT table and immobilized according to the pre-recorded readings of the micrometric systems. This procedure takes less than 2 min and neither local anaesthesia nor sedation is needed. Two different localization boxes, one for each diagnostic unit, are used in conjunction with the immobilization frame during the imaging procedure. The whole imaging procedure at the CT scanner usually takes 30–45 min depending on the case.
a
2.4. Treatment At each irradiation session, the immobilization frame is attached to the floor-stand head support holder in the linac. An extensive quality control follows for the correct position of the frame and the correct positioning of the target to the linac’s isocenter. Afterwards, the patient enters the therapeutic unit and is positioned in the stereotactic frame at the position of complete immobilization by adjusting the micrometric systems to the pre-recorded readings. The localization box is again used in conjunction with the frame for the verification of the target positioning. After confirmation of the patient’s proper position, the localization box is removed and the first irradiating arc is performed. The above control procedure is repeated for each one of the non-coplanar arcs and for each treatment session. The preparation of the equipment takes about 20 min and the patient immobilization takes only 2 min. The total treatment time for six non-coplanar arcs is between 50 min and 1 h.
3. Results The same immobilization frame can be mounted as many times as is needed to most heads. In order to assess the Table 1 Relocation accuracy measured in 15 patientsa Direction of displacement
Mean ± 2SD (mm)
Maximum (mm)
Y-axis Z-axis
0.5 ± 0.4 0.3 ± 0.2
0.7 0.4
a
Eighty-five repositionings in total.
3D relocation accuracy measured in two AVM casesa
Two repositionings for each case.
accuracy of relocation, repeated mountings of the immobilization frame were studied in two volunteers. The volunteers had four tattoo skin-marks in different locations on the head along the path of the central beam axis when the gantry was rotated. They were positioned in the frame several times and each time the deviation of the skin-mark from the central beam axis was measured for each appropriate gantry angle. The measurements were repeated for different table angles. The maximal deviation between two mountings of the frame on the head, extrapolated to central brain structures, was shown to be less than 1 mm. The simulator was used as an additional verification procedure for the precise relocation of the frame. Before each treatment session of real patients, lateral X-ray films were taken. The whole stereotactic apparatus was transferred to the simulator before each session and one lateral projection of the patient’s head was taken in the position of complete immobilization using the same beam set-up as for the irradiation. It was not possible to also take anterior-posterior projections as a floor-stand system is used. The relocation was measured for 15 different patients (85 repositionings) by comparing the Y and Z stereotactic co-ordinates of the target centre with the calculated co-ordinates at the treatment planning session. The results are presented in Table 1. The maximum anterior-posterior and inferior-superior displacements was less than 1 mm in all cases. This procedure was shown to be very useful and has been adopted for all patients. Systematically, after the treatment planning procedure and the calculation of the target co-ordinates, verification of the correct position of the frame assembly is done prior to the first treatment session. This procedure is repeated again one or more times during the treatment course, depending on the total number of sessions, and systematically before the last session. In two AVM cases, it was possible to check the 3D relocation accuracy of the frame by measuring the 3D displacement of the target centre between two repeated mountings using two orthogonal projected angiographic images. The relocation was measured by comparing the X, Y and Z stereotactic co-ordinates of each mounting of the frame with the calculated co-ordinates in the treatment planning session. The results are presented in Table 2.
4. Discussion The in-house-constructed stereotactic frame described
316
K. Theodorou et al. / Radiotherapy and Oncology 47 (1998) 313–317
above is simple to use, easily constructed, non-invasive, relocatable and well tolerated by the patients. Furthermore, the accuracy of relocation provides the possibility of planning the patient using multiple imaging modalities and allows the selection of multiple fraction treatment schemes. The use of cellulose acetate cast systems for non-invasive fractionation presents two major disadvantages, i.e. low relocation accuracy compared with the invasive and some of the non-invasive frames and inconvenience to the patient as some find it claustrophobic. One other non-invasive head fixation method used in the past was presented by Greitz et al. [7], where a plastic helmet and a base-ring were mounted to the head of the patient. However, the total relocation error was reported to be 2–3 mm in the transverse plane and 3–4 mm in the vertical plane and also the method necessitates time for fabrication. Finally, the non-invasive technique of Hariz et al. [8] requires an aware and fully co-operative patient so if this is not the case a general anaesthesia may be required. The immobilization technique presented here has some similarities with the Gill–Thomas stereotactic frame [5,10]. However, it has three key differences as follows: (a) the elastic strap has been replaced by the nose cast which makes the immobilization more convenient to the patient; (b) the addition of the two micrometric systems controls the correct patient immobilization and relocation and assures the accuracy of the method; and (c) it is not necessary for the patient to wear the frame while travelling between any of the diagnostic or therapeutic procedures. For the Gill– Thomas stereotactic frame, measurements of the relocation accuracy reported for four consecutive patients showed an overall mean AP displacement of 1 mm (range 0.7–1.2 mm) and an overall mean lateral displacement of 1 mm (range 0.4–1.6 mm) [6]. From the presented results, a small improvement has been achieved in terms of relocation accuracy. The immobilization frame developed and presented here is relatively cheap to use per patient and it does not take long to customize the frame for each patient. The first fixation of the head takes approximately 30 min and is performed by an experienced dentist, whereas all the other fixations during imaging and therapeutic procedures take only 2 min. The cost of the impression material and the production of the casts is usually less than US$100 (US$67 for the construction of the acrylic dental and nasion casts and around US$30 for the solid silicon used as impression material and the cold acrylic used for the fixation of the casts to the frame).
5. Conclusions The proposed non-invasive immobilization frame presents some advantages over many commercial systems. Experience from 22 patients and more than 200 positionings
showed that a general or partial anaesthesia was never necessary for the patient to endure the immobilization and generally all patients tolerated the frame well without complaint. Furthermore, the insertion of the patient’s head into the frame is very easy and the immobilization procedure is very fast. The major advantage of using such a non-invasive stereotactic frame is the flexibility in timing the different diagnostic procedures (CT and angiography) as well as the freedom to decide when and for how many sessions the stereotactic irradiation will be performed. In addition, the frame can be used not only for stereotactic cases with small brain lesions, but also as an immobilization device for conformal radiotherapy of brain lesions using stationary fields and/or multiple converging irradiation arcs. To date, 22 patients with ages varying between 12 and 70 years have been treated using this method with or without the additional circular collimators and with a multiple fraction scheme.
Acknowledgements The authors express their sincere thanks to A. Saliagas and S. Alvanos for the design and construction of many mechanical parts of the frame and Dr A. Mazal and Dr J.C. Rosenwald of the Institute Curie for their support and advice. This project was supported by the ‘Bodosakis’ Foundation.
References [1] AAPM (American Association of Physics in Medicine) Report No. 54. Stereotactic Radiotherapy. AAPM, 1995. [2] Colombo, F. Stereotactic radiosurgery utilizing a linear accelerator. Appl. Neurophysiol. 48: 133–145, 1985. [3] Delannes, M., Daly, N.J., Bonnet, J., Sabatier, J. and Tremoulet, M. Fractionated radiotherapy of small inoperable lesions of the brain using a non-invasive stereotactic frame. Int. J. Radiat. Oncol. Biol. Phys. 21: 749–755, 1991. [4] Engler, M.J., Curran, B.H., Tsai, J., et al. Fine tuning of linear accelerators accessories for stereotactic radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 28: 1001–1008, 1994. [5] Gaboriaud, G., Mazal, A., Pontvert, D., et al. Radiothe´rapie ste´re´rotaxique a` l’Institute Curie. Bull Cancer/Radiother. 78: 311– 319, 1991. [6] Graham, J.D., Warrington, A.P., Gill, S.S. and Brada, M. A noninvasive, relocatable frame for fractionated radiotherapy and multiple imaging. Radiother. Oncol. 21: 60–62, 1991. [7] Greitz, T., Lax, I., Bergstro¨m, M., et al. Stereotactic radiation therapy of intracranial lesions. Methodological aspects. Acta Radiol. Oncol. 25: 81–89, 1986. [8] Hariz, M.I., Henriksson, R., Lofroth, P., Laitinen, L. and Saberborg, N.E. A noninvasive method for fractionated stereotactic irradiation of brain tumors with linear accelerator. Radiother. Oncol. 17: 57–72, 1990. [9] Kooy, H.M., Dunbar, S.F., Tarbell, N.J., et al. Adaption and verification of the relocatable Gill–Thomas–Cosman frame in stereo-
K. Theodorou et al. / Radiotherapy and Oncology 47 (1998) 313–317 tactic radiotherapy. Int. J. Radiat. Oncol. Biol. Phys. 30: 685– 691, 1994. [10] Leksell, L. The stereotactic method and radiosurgery of the brain. Acta. Chir. Scand. 102: 316–319, 1951. [11] Podgorsak, E.B., Souhami, L., Caron, J., et al. A technique for frac-
317
tionated stereotactic radiotherapy in the treatment of intracranial tumors. Int. J. Radiat. Oncol. Biol. Phys. 27: 1225–1230, 1993. [12] Webb, S. The Physics of Three-Dimensional Radiation Therapy. Medical Science Series, Institute of Physics, London, 1993.