(Radiology. 2000;214:591-595.)
© RSNA, 2000
Head and Neck Tumors: Fractionated Frameless Stereotactic Interstitial Brachytherapy-Initial Experience1
Reto J. Bale, MD,
Wolfgang Freysinger, PhD,
Andreas R. Gunkel, MD,
Michael Vogele, MD,
Arpad Sztankay, MD,
Thomas Auer, MD,
Paul Eichberger, MD,
Arno Martin, MD,
Thomas Auberger, MD,
Arne W. Scholtz, MD,
Werner Jaschke, MD,
Walter F. Thumfart, MD and
Peter Lukas, MD, PhD
1 From the Departments of Radiology (R.J.B., W.J.), Ear Nose and Throat (W.F., A.R.G., M.V., A.M., A.W.S., W.F.T.), and Radiotherapy and Oncology (A.S., T. Auer, P.E., T. Auberger, P.L.), University of Innsbruck, Anichstrasse 35, 6020 Innsbruck, Austria. From the 1998 RSNA scientific assembly. Received November 29, 1998; revision requested December 29; final revision received May 13, 1999; accepted May 20. Address reprint requests to R.J.B. (e-mail: reto.bale@uibk.ac.at).
 |
Abstract
|
|---|
The authors used a frameless stereotactic navigation system, the Vogele-Bale-Hohner head holder, and a targeting device to reproducibly position brachytherapy needles for fractionated interstitial brachytherapy in 12 patients with inoperable cancers of the head and neck. In all cases, deviations of the needle relative to the planned position were within 1–15 mm depending on the location of the tumor.
Index terms: Computed tomography (CT), three-dimensional, 10.12117, 20.12117 • Computers, examination control • Head and neck neoplasms, therapeutic radiology, 10.1267, 20.1267 • Stereotaxis, 10.1267, 20.1267
|
Introduction
|
|---|
Many patients experience progressive inoperable head and neck tumors (1). In these patients, palliative interstitial brachytherapy may be the treatment of choice (2,3). Interstitial brachytherapy allows the application of high radiation doses with accurate distribution into the tumor mass (2,4,5). The efficacy of fractionated brachytherapy delivered with needles depends on the precise and reproducible placement of the needle at the site of the desired target point.
Previous studies in a laboratory in our institution showed that exact placement of needles is possible with three-dimensional navigation systems (6–9). Herein, we present our initial experience with computer-assisted stereotactic interstitial brachytherapy in 12 patients with head or neck tumors.
|
Materials and Methods
|
|---|
Results of needle placement by means of three-dimensional navigation systems have been published previously (6–10). Therefore, we provide only a brief description of our technique including the recent modifications.
Frameless Stereotactic Navigation System
The frameless stereotactic navigation system (Viewing Wand; ISG Technologies, Mississauga, Ontario, Canada) allows real-time intraoperative visualization of the position of the probe relative to the preoperative computed tomographic (CT) and magnetic resonance (MR) imaging data set (11–13). We treated patients 1–4 (Table) with use of an articulated position-sensing arm (Faro Medical Technologies, Orlando, Fla), which was replaced for patients 5–12 with an infrared optical digitizing system (Polaris; Northern Digital, Waterloo, Ontario, Canada) (14). The software (ISG Technologies) was the same for all patients.
Head Holder
The Vogele-Bale-Hohner (VBH) head holder (VBH HeadFix; Medical Intelligence, Schwabmünchen, Germany) allows reproducible noninvasive immobilization by means of an individualized vacuum dental cast (8,9,13,15,16). Registration elements are mounted bilaterally on the mouthpiece. Fiducial markers on the registration rods provide accurate external reference points.
Patients 1–4 were treated with the original carbon mouthpiece (13,15), and patients 7–12 were treated by using the modified VBH mouthpiece. The latter is cast in one piece with two transverse and two anterior extensions that allow reproducible attachment of registration rods. Application of a recently developed vacuum spacer allows faster and easier production of the vacuum impression (Fig 1).
View larger version (120K):
[in this window]
[in a new window]
|
Figure 1. Modified VBH mouthpiece: dental impression (1) with vacuum area (arrow); vacuum spacer (2); registration rods (3); anterior extensions (4); lateral extensions (5); plastic tube (6), which leads to vacuum pump (7) with manometer (8).
|
|
The impression tray of the modified VBH mouthpiece is covered with adhesive (Correct VPS Adhesive, Jeneric/Pentron, Wallingford, Conn, or Polyvinyl Siloxan Adhäsiv, ESPE, Seefeld, Germany), which is allowed to dry for 2–3 minutes. The vacuum spacer is inserted in the vacuum hole, and the impression tray is filled with dental material (Correct VPS Adhesive, Jeneric/Pentron, or Dimension Penta H, ESPE). The vacuum spacer is pushed slightly further into the impression material, and then the impression is taken. Hardening of the dental material takes about 5 minutes. The mouthpiece is gently removed from the mouth, and the vacuum spacer is peeled off. A vacuum area remains with a hole in the center that leads to the vacuum hole. A plastic tube is attached to the adapter on the lower side of the impression tray.
Targeting Device
The targeting device (8) consists of two components, each mounted on a hydraulic arm. The magnetic cup with the metal sphere allows the target point to be defined in stereotactic space. The actual targeting device provides exact guidance of the probe or the needle. If more than one needle is required, a template with parallel needle guides is used (6).
Procedure
For image acquisition, the patient is immobilized in the VBH head holder. The mouthpiece, with hydraulic arms fixated to a rigid unit, is removed and stored in this position. The CT data set is reconstructed in three dimensions and transferred to the frameless stereotactic navigation system, with which entrance and target points of the needle(s) are determined.
Simulation procedure.—Before the actual therapeutic procedure is initiated, planning is performed. Reference points on the registration rods are used to simulate the patient. With the probe of the frameless stereotactic navigation system, navigation with monitor control is possible in the virtual patient. Entry and target point(s), which have been previously defined on the monitor, can be determined in stereotactic space. This information is used to align the targeting device and to measure the insertion depth of the needle.
Therapy.—Exact repositioning of the patient is verified by means of the navigation system and repositioning control elements (8,16). The targeting device is repositioned and readjusted, if necessary. After a small skin incision, the hollow needle is advanced through the preset targeting device to the preplanned depth. If more than one hollow needle has to be inserted, the first needle is advanced with three-dimensional navigation by using a template with parallel holes spaced in fixed distances. The remaining needles are inserted according to the geometry of the tumor. The insertion depths of these needles are determined with the navigation system.
After CT (Figs 2, 3), the radiation therapy plan can be adjusted to the actual position of the needle. Dose distribution is calculated with software (ABACUS; Isotopen-Technik Dr Sauerwein, Haan, Germany ). In all cases, the 100% isodose covered the target contours (tumor volume) (Fig 4).
View larger version (155K):
[in this window]
[in a new window]
|
Figure 2a. Patient 2. (a) CT images obtained with the three-dimensional navigation system to determine the target point depict recurrence of a sinus maxillaris adenocarcinoma (arrowheads) growing into the right orbit. (b-d) Transverse CT scans depict monitoring of the position of the needle (arrow in b and arrowhead in c and d) in the first through third sessions, respectively.
|
|
View larger version (191K):
[in this window]
[in a new window]
|
Figure 2b. Patient 2. (a) CT images obtained with the three-dimensional navigation system to determine the target point depict recurrence of a sinus maxillaris adenocarcinoma (arrowheads) growing into the right orbit. (b-d) Transverse CT scans depict monitoring of the position of the needle (arrow in b and arrowhead in c and d) in the first through third sessions, respectively.
|
|
View larger version (143K):
[in this window]
[in a new window]
|
Figure 2c. Patient 2. (a) CT images obtained with the three-dimensional navigation system to determine the target point depict recurrence of a sinus maxillaris adenocarcinoma (arrowheads) growing into the right orbit. (b-d) Transverse CT scans depict monitoring of the position of the needle (arrow in b and arrowhead in c and d) in the first through third sessions, respectively.
|
|
View larger version (136K):
[in this window]
[in a new window]
|
Figure 2d. Patient 2. (a) CT images obtained with the three-dimensional navigation system to determine the target point depict recurrence of a sinus maxillaris adenocarcinoma (arrowheads) growing into the right orbit. (b-d) Transverse CT scans depict monitoring of the position of the needle (arrow in b and arrowhead in c and d) in the first through third sessions, respectively.
|
|
View larger version (152K):
[in this window]
[in a new window]
|
Figure 3a. Patient 3. (a) CT images obtained with the three-dimensional navigation system to determine the target point depict metastasis of a hepatocellular carcinoma (arrowheads) in the right orbit. (b-d) Transverse CT scans depict monitoring of the position of the needle (arrow) in the first through third sessions, respectively.
|
|
View larger version (153K):
[in this window]
[in a new window]
|
Figure 3b. Patient 3. (a) CT images obtained with the three-dimensional navigation system to determine the target point depict metastasis of a hepatocellular carcinoma (arrowheads) in the right orbit. (b-d) Transverse CT scans depict monitoring of the position of the needle (arrow) in the first through third sessions, respectively.
|
|
View larger version (180K):
[in this window]
[in a new window]
|
Figure 3c. Patient 3. (a) CT images obtained with the three-dimensional navigation system to determine the target point depict metastasis of a hepatocellular carcinoma (arrowheads) in the right orbit. (b-d) Transverse CT scans depict monitoring of the position of the needle (arrow) in the first through third sessions, respectively.
|
|
View larger version (153K):
[in this window]
[in a new window]
|
Figure 3d. Patient 3. (a) CT images obtained with the three-dimensional navigation system to determine the target point depict metastasis of a hepatocellular carcinoma (arrowheads) in the right orbit. (b-d) Transverse CT scans depict monitoring of the position of the needle (arrow) in the first through third sessions, respectively.
|
|
View larger version (161K):
[in this window]
[in a new window]
|
Figure 4. Patient 2. Clinical target volume (arrowheads indicate tumor) covered by 100% isodose. Note that the patient's anatomic right side is displayed on the viewer's right.
|
|
Brachytherapy is introduced (Gammamed 12i; Isotopen-Technik Dr Sauerwein) with use of iridium 192.
Patients
The Table provides information for the 12 patients treated with this method. The decision for computer-assisted brachytherapy as the method of choice was made by means of an interdisciplinary conference. All patients had previously been treated with a combination of standard external-beam radiation therapy and chemotherapy. All patients treated with this method from February 1996 through October 1998 were included in this study. The Karnofsky index of their disease was 50–80. Informed consent was obtained from all patients after the nature of the procedure had been fully explained. The study was approved by our institutional review board.
Owing to the approximate rotational symmetry of the tumors in nine of the patients, only one hollow needle was applied per radiation session. The asymmetric shape of the tumor necessitated insertion of five needles in one patient and three needles in two patients (Table).
The time during which the patients were treated depended on the number of fractions. One fraction was given every week, which resulted in a treatment time of 2–6 weeks (mean, 3.8 weeks). The dose and number of fractions was determined on the basis of the individual's previous treatment and our long-term experience with high-dose-rate interstitial brachytherapy since 1986. The individual doses ranged from 10 to 30 Gy (Table). The tumor volumes treated were in the range of 9–51 cm3 (mean, 23 cm3 ± 13[SD]).
The follow-up time for surviving patients varied between 3 and 15 months (mean, 7.3 months ± 4.1). The approximate deviations were determined of the needle tip position depicted at CT from that planned with the navigation system, and the clinical outcome was evaluated.
|
Results
|
|---|
The number of hollow needles per session, the number of brachytherapy sessions, the total radiation doses, the mean accuracy of targeting, the follow-up time, and the remission rates are listed in the Table.
Patients 1, 2 (Fig 2), 9, and 12 were repositioned precisely in each session, and the needle was placed within the range of 1–2 mm. For patients 3 (Fig 3) and 11, exact repositioning was not achievable with the VBH head holder owing to an edentulous upper palate, which resulted in a needle deviation of approximately 3 mm.
Modifications of the procedure were required for patients 4–8 and 10. The targeting device was adjusted according to the actual patient setup on the basis of the trajectory images obtained with the three-dimensional navigation system (6,7). In patient 4, insertion of the mouthpiece was not possible owing to a rigid jaw. In this patient, fixation of the head was achieved by using surgical plastic tape. Anatomic landmarks were used to register the patient, and the registration was optimized by means of surface matching. In patient 5, repositioning of the targeting device was impeded by the clavicle and the sternum, which were not part of the three-dimensional data set. Registration was achieved by using the reference points on the registration rods, anatomic landmarks, and surface matching. In patients 6–8 and 10, accurate repositioning of the neck was impossible owing to a variety of reasons, including short neck (patient 6), extreme kyphosis (patient 7), and reduced compliance related to primary disease (patients 8, 10). Therefore, the entire procedure had to be performed with general anesthesia. CT was performed prior to each session, and the patients were transported to the brachytherapy theater on a portable couch while they remained in the same position. The targeting device was adjusted without previous simulation.
The treatment was beneficial in all patients on the basis of improvement in the initial signs (eg, tenderness, pain, and symptoms caused by tumor compression) that resulted in at least a transient improvement in the quality of life. Follow-up CT studies revealed partial tumor remission in 10 patients and complete tumor regression in two.
|
Discussion
|
|---|
The key steps of computer-assisted frameless stereotactic brachytherapy in the head and neck region are registration of the patient's data set and accurate repositioning of the patient's head or neck. The accuracy of repositioning with the VBH mouthpiece itself was less than 1 mm (15). The error of repositioning the patient's head in the VBH head holder was 1.02 mm (16). To achieve high accuracy, at least two or three healthy teeth on the upper palate are required.
Our first clinical experiences demonstrate that this method allows accurate needle insertion to a predefined target in certain circumstances. In another study performed by our group, the trigeminal ganglion was successfully targeted on the first attempt in five cadavers and nine patients. Because the foramen ovale has a diameter of about 3 x 5 mm and the needle was aimed through its center, a deviation greater than 1.5–2.5 mm would result in the progress of the needle being halted by bone (8).
For intracranial tumors, invasive stereotactic frames screwed into the calvaria have been used for accurate insertion of iodine seeds (17,18). Stereotactic frames cannot be easily relocated and are therefore not suitable for fractionated procedures. In contrast, the VBH head holder allows noninvasive, reproducible, and precise immobilization of the head, which permits temporal and spatial separation of image acquisition, planning, and execution of therapy. This allows a reduction in the duration and complexity of the entire procedure. This method is efficient in the head area owing to the lack of tissue displacement.
In the neck area, mobility and soft-tissue displacement may limit the use of our technique. We tried to compensate for these problems by not repositioning the patient between imaging and therapy. With the patient's head fixated in the head holder, planning and execution of therapy were conducted in one session for every fraction delivered, by leaving the patient in the same position as evaluated during CT. However, immobilization of the patient's neck with the VBH head holder has to be optimized. In some patients, the radiation plan had to be adapted to the actual position of the needle. Great deviations occurred between planned and achieved needle position owing to the flexibility of the neck, soft-tissue displacement, and lack of clearly defined anatomic landmarks for patient-to-image registration in this area.
Owing to the rotational symmetry of the tumors in most patients, only one hollow needle had to be inserted. In the three patients (patients 5–7) in whom more than one needle was used, accurate execution of the planned trajectory could not be accomplished. In these three patients, the deviations of needle positions occurred with all needles including those that had been advanced by using the frameless stereotactic navigation system. This fact indicates that the deviations were not a result of the number of needles but of the reasons discussed previously, especially the lack of rigid immobilization of the soft tissue in the lower neck area. Improved immobilization may increase the accuracy of needle placement.
A new vacuum-based immobilization device for extremities (19), which has recently been developed in collaboration with our institution, may be adapted to the requirements of neck fixation. Nevertheless, more patients and data are needed to allow a definitive statement about the efficacy of treatment of large tumors with asymmetric shapes, especially in the lower neck.
Although experience is still limited, especially with respect to the number of patients, clinical data, and treatment of large asymmetric tumors, computer-assisted targeting seems to be an accurate procedure in the head region and therefore suitable for brachytherapy of tumors in this area. We believe use of the noninvasive VBH head holder is a prerequisite for accurate and reproducible needle positioning. Our approach to brachytherapy allows prospective planning on the basis of a single CT-MR data set that is established and reproducibly executed in each fraction of therapy. Further modifications in this technique are necessary to treat tumors in the lower neck region owing to the soft tissue and possible displacement of the point of interest in this area.
|
Acknowledgments
|
|---|
R.J.B. thanks Reinhart Sweeney, MD, and Gerald Wetscher, MD, for assistance in manuscript preparation.
|
Footnotes
|
|---|
Abbreviation: VBH = Vogele-Bale-Hohner
Author contributions: Guarantors of integrity of entire study, P.L., W.F.T., W.F.; study concepts and design, P.L., R.J.B., M.V., W.F.; definition of intellectual content, P.L., R.J.B., M.V., W.F.; literature research, R.J.B., W.F., A.M., A.S.; clinical studies, all authors; data acquisition and analysis, W.F., A.S., A.M., R.J.B.; manuscript preparation, R.J.B.; manuscript editing, R.J.B., W.F., A.S.; manuscript review, W.F., A.S., P.L., W.J.
|
References
|
|---|
-
Yasumoto M, Shibuya H, Hoshina M, Takeda M, Matsumoto S, Suzuki S. External and interstitial radiotherapy in the treatment of oropharyngeal squamous cell carcinoma. Br J Radiol 1995; 68:630-635.[Abstract]
-
Alesch F, Hawliczek R, Koos WT. Interstitial irradiation of brain metastases. Acta Neurochir Suppl Wien 1995; 63:29-34.[Medline]
-
Ostertag CB. Brachytherapy: interstitial implant radiosurgery. Acta Neurochir Suppl Wien 1993; 58:79-84.[Medline]
-
Beier J, Wust P, Huth O, et al. Planning of catheter implantation for interstitial radiation and hyperthermia therapy. In: Lemke HU, Vannier MW, Inamura K, eds. Computer assisted radiology and surgery. Amsterdam, the Netherlands: Elsevier Science, 1996; 643-647.
-
Kovacs G, Hebbinghaus D, Dennert P, Kohr P, Wilhelm R, Kimming B. Conformal treatment planning for interstitial brachytherapy. Strahlenther Onkol 1996; 172:469-474.[Medline]
-
Auer T, Hensler E, Eichberger P, et al. 3D navigation in der interstitiellen stereotaktischen brachytherapie. Strahlenther Onkol 1998; 174:82-87.
-
Lukas P, Hensler E, Auer T, et al. Stereotactic interstitial brachytherapy with 3D navigation in head and neck tumors (abstr). Radiology 1996; 201(P):233.[Abstract/Free Full Text]
-
Bale RJ, Vogele M, Martin A, et al. The VBH head holder improves frameless stereotactic brachytherapy of head and neck tumors. Comput Aided Surg 1997; 2:286-291.[Medline]
-
Bale RJ, Martin A, Vogele M, et al. Erste erfahrungen mit computerunterstützter stereotaktischer interstitieller brachytherapie. Strahlenther Onkol 1998; 174:473-477.[Medline]
-
Mösges R, Korves B, Ammon J, Kremer B. Computerunterstützte positionierung für das nachladeverfahren mit iridium-192. HNO 1991; 39:429-432.[Medline]
-
Freysinger W, Gunkel AR, Bale RJ, et al. 3D navigation in ENT surgery. Ann Otol Rhinol Laryngol 1998; 107:953-958.[Medline]
-
Zinreich SJ, Tebo SA, Long DM, et al. Frameless stereotaxic integration of CT imaging data: accuracy and initial applications. Radiology 1993; 188:735-742.[Abstract/Free Full Text]
-
Bale RJ, Vogele M, Freysinger W, et al. Minimally invasive head holder to improve the performance of frameless stereotactic surgery. Laryngoscope 1997; 107:373-377.[Medline]
-
Glossop N, Hu R. Assessment of vertebral body motion during spine surgery. Spine 1997; 22:903-909.[Medline]
-
Martin A, Bale RJ, Vogele M, Gunkel AR, Thumfart WF, Freysinger W. Vogele-Bale-Hohner mouthpiece: registration device for frameless stereotactic surgery. Radiology 1998; 208:261-265.[Free Full Text]
-
Sweeney R, Bale RJ, Vogele M, et al. Repositioning accuracy: comparison of a noninvasive head holder with thermoplastic mask for fractionated radiotherapy and a case report. Int J Radiat Oncol Biol Phys 1998; 41:475-483.[Medline]
-
Gutin PH, Phillips TL, Wara WM, et al. Brachytherapy of recurrent malignant brain tumors with removable high-activity iodine-125 sources. J Neurosurg 1984; 60:61-68.[Medline]
-
Hughes SW, Sofat A, Kitchen ND, et al. Computer planning of stereotactic iodine-125 seed brachytherapy for recurrent malignant gliomas. Br J Radiol 1995; 68:175-181.[Abstract]
-
Bale RJ, Vogele M, Rieger M, Buchberger W, Lukas P, Jaschke W. A new vacuum-based device for extremity immobilization. AJR Am J Roentgenol 1999; 172:1093-1094.[Free Full Text]
TOP
Abstract
Introduction
