Departments of Neurosurgery (ML, JB), Neurology (SP, JH), Neuroradiology (DB), and PET/Biomedical Cyclotron Unit (SG, LMB, AL, ES), Université Libre de Bruxelles, Erasme Hospital, Brussels, Belgium Neurosurgery 31; 792-797, 1992 ABSTRACT: WE DESCRIBE A technique that allows target definition for stereotactic brain biopsy using coordinates calculated on stereotactic positron emission tomographic (PET) images. In this study, PET images were obtained using [18F]-labeled fluorodeoxyglucose, a marker for glucose metabolism. The difference between PET-calculated and actual stereotactic coordinates of simulated targets is within PET spatial resolution. Combined computed tomography- and PET-guided stereotactic biopsies were performed in 11 patients with brain tumors. In this report, we describe two representative patients who underwent stereotactic brain biopsy using the present technique. Because of the complementary role of PET and computed tomography, their integration in multimodality planning might optimize the target selection for stereotactic brain biopsies. KEY WORDS: Brain neoplasm; Positron emission tomography; Stereotaxis Image-guided stereotactic brain biopsies using computed tomographic (CT) scanning or magnetic resonance imaging (MRI) are performed to identify the nature of intracranial lesions such as brain neoplasms (2,15,16). Because brain tumors like gliomas are histologically heterogeneous (19), pathological examination of the stereotactic samples do not always yield their actual grading and extension (5,10). Therefore, in an attempt to obtain more representative biopsies, we combined CT stereotactic planning with positron emission tomography (PET), a neuroimaging technique providing independent and complementary metabolic information. PET with [18F]-labeled fluorodeoxyglucose (PET-FDG) is used to study the brain glucose metabolism and, when performed in patients with brain tumors, was shown to improve the assessment of tumor heterogeneity, extension, and prognosis (for review, see Ref. 6). Even though PETgenerated images display poor anatomical definition, integration of PET data in multimodality stereotactic planning might be helpful for guiding stereotactic brain biopsies.
PATIENTS AND METHODS A one-day procedure was established for placement of the stereotactic frame and acquisition of images, including stereotactic PET-FDG, surgical planning, and biopsy. A commercial arc-quadrant stereotactic system was used in all procedures (ZD-Neurosurgical Localizing Unit, F. L. Fischer, Freiburg, Germany). Eleven patients had stereotactic brain biopsies with this procedure. All patients were given informed consent, and the procedure was in accordance with the ethical guidelines of our institution. PET-guided stereotactic biopsy The CT- and MRI-compatible carbon fiber head ring unit was attached to the patient's head under local anesthesia. The circular base ring of the unit defines the stereotactic coordinate system: the origin of the system is in the center of the ring plane and coordinates are defined orthogonally as x (left-right), y (posterior-anterior), and z (inferior-superior). CT scans were performed on a Siemens Somatom Plus (Siemens AG Medical Engineering, Erlangen, Germany). A clamp specifically designed to secure the head ring to the Siemens couch (F. L. Fischer, Freiburg, Germany) was used to provide perfect alignment of the base ring with the CT gantry during image acquisition. A localizing system consisting of four plastic plates was attached anteriorly, posteriorly, and bilaterally to the base ring. Three thin radiopaque fiducials embedded in each plate generated 12 reference marks on CT scan slices used for calculations, according to the technique reported by Sturm et al. (20). After a series of 2-mm thick contrast-enhanced CT scan slices was completed, the patient was transferred to the PET facility. Our PET/Cyclotron unit is equipped with a 15plane CTI/Siemens 933 tomograph (Knoxville, TN). During PET image acquisition, the stereotactic head ring is secured to the clamp used for CT, which also fits into the PET Siemens couch. This allows a fast and easy comparable positioning of the patient's head during CT and PET. Alignment of the base ring with the PET gantry is controlled using two crossed laser beams. To create a fiducial reference system compatible with PET, we used the MRI localizers with minor modifications. The localization plates for transverse MRI are the same as for CT, except that the fiducials are made of grooves containing tubing filled with a copper sulfate solution. The two lateral fiducials of the MRI localizing plates may be easily replaced by standard tubing of 1.2-mm internal diameter filled either with saline for the transmission scan, or with a [18F]fluoride solution (60-80 µCi/ml) for the emission scan. Transmission scan is performed using an eight-ring [18F]fluoride source to
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
AUTHOR(S): Levivier, Marc, M.D.; Goldman, Serge, M.D.; Bidaut, Luc M., Ph.D.; Luxen, André, Ph.D.; Stanus, Etienne, Ph.D.; Przedborski, Serge, M.D.; Balériaux, Danielle, M.D.; Hildebrand, Jerzy, M.D., Ph.D.; Brotchi, Jacques, M.D., Ph.D.
In the present report, we describe and illustrate the method developed to integrate PET-FDG images into a system of stereotactic coordinates. This technique is initially used to accurately define a biopsy target in the more metabolically active areas of the tumor. Then, the serial biopsy coordinates are transferred onto the stereotactic PET images for further correlation of tumor metabolism and histopathology.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Neurosurgery 1992-98 October 1992, Volume 31, Number 4 792 Positron Emission Tomography-Guided Stereotactic Brain Biopsy Technical Note
RESULTS Case reports Stereotactic PET was performed in 11 patients. Among them, 7 had hypermetabolic areas that were used to guide the biopsy. The technique described for PET-guided stereotactic biopsy is illustrated in the following two cases. Patient 1, a 52-year-old right-handed man, initially referred to our institution because of diffuse headaches and seizures. His neurological examination appeared normal. CT scan of the brain revealed an illdefined lesion in the left subcortical parietotemporal region. A stereotactic brain biopsy was scheduled. CT and PET were performed as described in the "Methods." PET revealed an area of increased metabolism in the left parietal lobe (Fig. 2A), which was used as a first biopsy target. A second target was selected in an hypometabolic area. Both targets were located in the tumor, controlled by projecting them on the stereotactic CT scan (Fig. 2B). Pathological examination revealed anaplastic astrocytoma in samples of both tracts. Patient 2, a 53-year-old right-handed man, who underwent an extensive resection of a right cerebellar low-grade astrocytoma 9 years previously. Adjuvant radiation therapy focused on the posterior fossa (5500 rads) was given. His neurological status improved and remained stable until recently. When he was admitted to our institution, CT and MRI scans of the brain showed a lesion located in the right part of the pons and in the right middle cerebellar peduncle. A stereotactic brain biopsy was scheduled. The stereotactic base ring was attached to the patient's skull in an inverted position. CT, vertebral DA, and PET were performed as described in "Methods." The PET study investigated the region of the brain located under the plane of the base ring. PET showed an area of increased metabolism set as the target (Fig. 3). Biopsy was performed using the transcerebellar route. Pathological examination revealed several malignant foci within a low-grade astrocytoma. Accuracy of technique Using the phantom containing the 11 radioactive targets (Fig. 4), the differences between PET and actual stereotactic coordinates were found to be (means ± SD): 1.03 ± 0.8 mm, 1.36 ± 0.59 mm, and 1.50 ± 0.91 mm for the x, y, and z axes, respectively. R values for correlation between PET-evaluated and actual coordinates were 0.999, 0.999, and 0.996 for the x, y, and z axes, respectively. DISCUSSION This report illustrates a method developed to integrate PET data into the procedure used for planning and performing stereotactic brain biopsies.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Accuracy of technique A hot-spot phantom simulating three-dimensional distribution of targets has been specifically designed for testing the accuracy of the PET system in calculating stereotactic coordinates (Fig. 1). The targets are made of 5-ml syringes filled with 0.5 ml of a [18F]fluoride solution; diameter and height of the cylindric targets are 8.4 mm and 9 mm, respectively. The stereotactic frame was fixed on the phantom and 11 targets were spirally placed on the phantom base in order to test the PET localization accuracy in the center as well as in the periphery of the stereotactic frame. Stereotactic PET was performed as described above. For each target, the plane with the highest radioactive count rate was selected and a pixel
located in the center of the target was interactively pointed at for calculation of coordinates. Mean and standard deviation of the differences between PET and actual x, y, and z stereotactic estimates of the 11 radioactive targets were calculated. Correlation between PET-evaluated and actual values was calculated for x, y, and z coordinates.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
correct the emission scan for attenuation. FDG is automatically synthesized (18), using the procedure described by Hamacher et al. (12). Radiochemical purity of the tracer is assessed by radio-thin layer chromatography (silica plates, acetonitrile/water: 95:5) and is higher than 99%. The patient is injected intravenously with 6 to 9 mCi FDG and maintained in the darkened room with no verbal stimulation. Images used for stereotactic calculation are acquired between 40 and 60 minutes after injection of the tracer. The PET camera allows simultaneous acquisition of 15 slices approximately 6.5 mm thick. Except for posterior fossa lesions, the most caudal slice is adjusted to the plane of the stereotactic ring. The eight planar landmarks generated by the radioactive fiducials allow rapid and accurate calculation of the zero point of the stereotactic system and of the coordinates of regions of interest using the PET processing software and customized personal computer spreadsheet software (4). Selective cerebral digital angiogram (DA) or plain anteroposterior and lateral skull x-rays are also performed, using the corresponding localizers. A personal computer and a digitizer are used to transfer coordinates from other image modalities onto the DA or the x-rays. The surgical planning begins with analysis of the PET images. If PET-FDG reveals a region of increased metabolism within the lesion boundaries, the plane that best displays the hypermetabolism is selected and a pixel located in the center of this zone is interactively pointed at on visual inspection. The coordinates of this pixel are then calculated and set as a target for biopsy. The target is then projected onto the corresponding stereotactic CT slice and, when performed, on anteroposterior and lateral DA, to control the reliability and the safety of the target and the trajectory. In patients in whom there is no obvious hyper-metabolic region on the stereotactic PET, surgery is planned using CT data only. The biopsy procedure was performed under general anesthesia with endotracheal intubation. Serial stereotactic biopsies were obtained with a sidecutting cannula, following the technique described by Kelly et al. (15). Postoperatively, biopsy sites in all patients were recorded and transferred onto the stereotactic PET for further correlation between tumor metabolism and histological results.
Research Grant 3.4565.88 from the National Fund for Scientific Research, Belgium. Received, December 3, 1991. Accepted, April 17, 1992. Reprint requests: Marc Levivier, M.D., Department of Neurosurgery, U.L.B.-Erasme Hospital, 808, Route de Lennick, B-1070 Brussels, Belgium.
1.
2.
3.
4.
5.
6.
7.
8.
9. 10.
ACKNOWLEDGMENTS The authors thank Dr. Robert E. Burke, Department of Neurology, College of Physicians and Surgeons of Columbia University, New York, for critical reading of the manuscript. This project was supported by Research Grant 9.4503.87 from the Belgian National Lottery and
11.
Alavi J, Alavi A, Chawluk J, Kushner M, Powe J, Hickey W, Reivich M: Positron emission tomography in patients with glioma. A predictor of prognosis. Cancer 62:10741078, 1988. Apuzzo MLJ, Chandrasoma PT, Cohen D, Zee C-S, Zelman V: Computed imaging stereotaxy: Experience and perspective related to 500 procedures applied to brain masses. Neurosurgery 20:930-937, 1987. Bergström M, Collins P, Ehrin E, Ericson K, Eriksson L, Greitz T, Halldin C, von Holst H, Langström B, Lilja A, Lundqvist H, Nagren K: Discrepancies in brain tumor extent as shown by computed tomography and positron emission tomography using [68Ga]EDTA, [11C]glucose, and [11C]methionine. J Comput Assist Tomogr 7:1062-1066, 1983. Bidaut L, Levivier M, Hildebrand J, Luxen A, Goldman S: Multimodality tumor evaluation through stereotactic correlation between XCT and PET. Proceedings of the 2nd European Workshop on FDG in Oncology, Heidelberg, Germany, December 9-11, 1991. Chandrasoma PT, Smith MM, Apuzzo MLJ: Stereotactic biopsy in the diagnosis of brain masses: Comparison of results of biopsy and resected surgical specimen. Neurosurgery 24:160-165, 1989. Coleman RE, Hoffman JM, Hanson MW, Sostman HD, Schold SC: Clinical application of PET for the evaluation of brain tumors. J Nucl Med 32:616-622, 1991. Derlon JM, Bourdet C, Bustany P, Chatel M, Theron J, Darcel F, Syrota A: [11C]Lmethionine uptake in gliomas. Neurosurgery 25:720-728, 1989. Di Chiro G: Positron emission tomography using [18F] fluorodeoxyglucose in brain tumors. A powerful diagnostic and prognostic tool. Invest Radiol 22:360-371, 1986. Di Chiro G: Which PET radiopharmaceutical for brain tumors? J Nucl Med 32:1346-1348, 1991. Feiden W, Steude U, Bise K, Gündisch O: Accuracy of stereotactic brain tumor biopsy: Comparison of the histologic findings in biopsy cylinders and resected tumor tissue. Neurosurg Rev 14:51-56, 1991. Glantz MG, Hoffman JM, Coleman RE, Friedman AH, Hanson MW, Burger PC, Herndon JE II, Meisler WJ, Schold SC Jr:
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
REFERENCES: (1-21)
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
PET acquisition in stereotactic conditions is easily performed, does not exceed more than 90 minutes, and does not cause any major additional discomfort to the patient. In our hands, data acquisition and target selection with PET appears as straightforward as with CT scanning, including that for posterior fossa lesions, as seen with Patient 2. The accuracy of target calculation is within PET resolution as demonstrated using the phantom and we always found a close correspondence of PET- and CTlocated targets. In the present protocol, PET sections are made of 2.591 × 2.591 mm pixels and are 6.75 mm thick; the precision of the target definition corresponds to a volume of 45 µl, which is about 5 times the volume of the biopsy specimen. Kelly (14) reported that 1.5 to 5 mm thick slices are adequate for CT-guided stereotactic brain biopsies, depending on the size of the lesion. PET cameras of the new generation will acquire sections of 3 mm thick, which should allow more accurate PET-guided stereotactic brain biopsies. PET is an imaging technique that provides functional data that are independent from CT or MRI anatomical images. The type of information obtained with PET mostly depends on the radiotracer selected. In this study, we have performed stereotactic PET with FDG because of the large amount of information available on the use of this marker for tracing brain tumors (9). PET-FDG is helpful in assessing the degree of malignancy and the prognosis of brain neoplasms independently of the histological grading (1,8) . It can also help in differentiating between the effects of treatments and in assessing tumor persistence, progression, and recurrence (6,11). Therefore, the present technique has been developed to take advantage of PET-FDG data in the stereotactic sampling of intracerebral tumors. In addition, we may anticipate that this method will allow us to accurately compare metabolic information obtained with PETFDG with data provided by other imaging modalities and neuropathology. This might eventually help in a better understanding of information provided by PET and answer the question of whether or not PET imaging adds to the diagnostic yield of stereotactic biopsies when glial tumors are suspected. This technique is also applicable to PET with other radiotracers, such as L-[11C]methionine (17), 2-[18F]fluoro-L-tyrosine, and other labeled amino acids that may be more suitable for the delineation of brain tumors or the study of their specific metabolism (3,7,13,21) . At the present stage, our data do not establish the role for PET-guided stereotactic brain biopsies in clinical routine. Studies comparing stereotactic biopsies done with and without PET guidance are needed to evaluate whether this imaging modality adds to the diagnostic yield.
14. 15.
16.
17.
18.
19. 20.
21.
COMMENTS Computed tomographic (CT)-guided stereotactic biopsy is anatomically very accurate. Unfortunately, there is often poor correlation between CT depiction of the lesions and the precise intralesional location of
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
13.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
12.
Identification of early recurrence of primary central nervous system tumors by [18F]fluorodeoxyglucose positron emission tomography. Ann Neurol 29:347-355, 1991. Hamacher K, Coenen HH, Stöcklin G: Efficient stereospecific synthesis of no-carrieradded 2- [18F]fluoro-2-deoxy-D-glucose using amino polyether supported nucleophilic substitution. J Nucl Med 27:235-238, 1986. Hübner KF, Purvis JT, Mahaley SM, Robertson Jr. JT, Rogers S, Gibbs WD, King P, Partain CL: Brain tumors imaging by positron emission computed tomography using 11C-labelled amino acids. J Comput Assist Tomogr 6:544-550, 1982. Kelly PJ: Tumor Stereotaxis. Philadelphia, WB Saunders, 1991. Kelly PJ, Daumas-Duport C, Kispert DB, Kall BA, Scheithauer BW, Illig JJ: Imaging-based stereotactic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 66:865-874, 1987. Lunsford LD: Diagnosis and treatment of mass lesions using the Leksell stereotactic system, in Lunsford LD (ed): Modern Stereotactic Neurosurgery. Boston, Martinus Nijhoff Publishing, 1988, pp 145-168. Mosskin M, von Holst H, Bergström M, Collins VP, Eriksson L, Johnström P, Norén G: Positron emission tomography with 11Cmethionine and computed tomography of intracranial tumors compared with histopathological examination of multiple biopsies. Acta Radiol 28:673-681, 1987. Padgett HC, Schmidt DG, Luxen A, Bida GT, Satyamurthy N, Barrio JR: Computercontrolled radiochemical synthesis: A chemistry process control unit for the automated production of radiochemicals. Appl Radiat Isot 40:433-445, 1989. Russell DS, Rubinstein LJ: Pathology of Tumours of the Nervous System. London, Edward Arnold, 1989, ed 5, pp 83-161. Sturm V, Pastyr O, Schlegel W, Scharfenberg H, Zabel H-J, Netzeband G, Schabbert S, Berberich W: Stereotactic computer tomography with a modified RiechertMundinger device as the basis for integrated stereotactic neuroradiological investigations. Acta Neurochir (Wien) 68:11-17, 1983. Wienhard K, Herholz K, Coenen HH, Rudolf J, Kling P, Stöcklin G, Heiss WD: Increased amino acid transport into brain tumors measured by PET of L-(2-18F)fluorotyrosine. J Nucl Med 32:1338-1346, 1991.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
Figure 1. PET data acquisition using the phantom. The radioactive targets are made of 5-ml syringes filled with 0.5 ml of the [18F]fluoride solution. The phantom is attached to the base ring of the stereotactic system, which is fixed to the clamp specifically designed to secure the head ring to the Siemens couch. The four localizers containing tubing filled with the [18F]fluoride solution are also fixed to the base ring.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
Figure 2. Stereotactic PET scan (A) and contrastenhanced CT scan (B) in Patient 1. The center of the hypermetabolic area on PET-FDG was visually defined and used as a target for biopsy (cross). Coordinates were calculated from the PET data and secondarily projected onto the CT scan (×). Note that the target selected on PET is within the limits of the lesion ill-defined on CT scan. The white circles were superimposed on the radioactive dots generated by the fiducials, which cannot be visualized with the color scale used in the present PET image.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
Figure 3. Stereotactic PET scan (A) and contrastenhanced CT scan (B) in Patient 2. The center of the hypermetabolic area on PET-FDG was visually defined and used as a target for biopsy (cross). Coordinates were calculated from the PET data and secondarily projected onto CT scan (×).
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
Figure 4. Scattergrams showing the correlation between actual (abscissa) and PET-calculated (ordinate) values for x, y, and z coordinates. Note the absence of distortion for high values of x and y corresponding to the periphery of the stereotactic frame. The inset shows the mean ± standard deviation for the absolute values of the differences between PET and actual stereotactic coordinates (mm). These values are within our PET scan resolution.
PATIENTS AND METHODS A one-day procedure was established for placement of the stereotactic frame and acquisition of images, including stereotactic PET-FDG, surgical planning, and biopsy. A commercial arc-quadrant stereotactic system was used in all procedures (ZD-Neurosurgical Localizing Unit, F. L. Fischer, Freiburg, Germany). Eleven patients had stereotactic brain biopsies with this procedure. All patients were given informed consent, and the procedure was in accordance with the ethical guidelines of our institution. PET-guided stereotactic biopsy The CT- and MRI-compatible carbon fiber head ring unit was attached to the patient's head under local anesthesia. The circular base ring of the unit defines the stereotactic coordinate system: the origin of the system is in the center of the ring plane and coordinates are defined orthogonally as x (left-right), y (posterior-anterior), and z (inferior-superior). CT scans were performed on a Siemens Somatom Plus (Siemens AG Medical Engineering, Erlangen, Germany). A clamp specifically designed to secure the head ring to the Siemens couch (F. L. Fischer, Freiburg, Germany) was used to provide perfect alignment of the base ring with the CT gantry during image acquisition. A localizing system consisting of four plastic plates was attached anteriorly, posteriorly, and bilaterally to the base ring. Three thin radiopaque fiducials embedded in each plate generated 12 reference marks on CT scan slices used for calculations, according to the technique reported by Sturm et al. (20). After a series of 2-mm thick contrast-enhanced CT scan slices was completed, the patient was transferred to the PET facility. Our PET/Cyclotron unit is equipped with a 15plane CTI/Siemens 933 tomograph (Knoxville, TN). During PET image acquisition, the stereotactic head ring is secured to the clamp used for CT, which also fits into the PET Siemens couch. This allows a fast and easy comparable positioning of the patient's head during CT and PET. Alignment of the base ring with the PET gantry is controlled using two crossed laser beams. To create a fiducial reference system compatible with PET, we used the MRI localizers with minor modifications. The localization plates for transverse MRI are the same as for CT, except that the fiducials are made of grooves containing tubing filled with a copper sulfate solution. The two lateral fiducials of the MRI localizing plates may be easily replaced by standard tubing of 1.2-mm internal diameter filled either with saline for the transmission scan, or with a [18F]fluoride solution (60-80 µCi/ml) for the emission scan. Transmission scan is performed using an eight-ring [18F]fluoride source to
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
AUTHOR(S): Levivier, Marc, M.D.; Goldman, Serge, M.D.; Bidaut, Luc M., Ph.D.; Luxen, André, Ph.D.; Stanus, Etienne, Ph.D.; Przedborski, Serge, M.D.; Balériaux, Danielle, M.D.; Hildebrand, Jerzy, M.D., Ph.D.; Brotchi, Jacques, M.D., Ph.D.
In the present report, we describe and illustrate the method developed to integrate PET-FDG images into a system of stereotactic coordinates. This technique is initially used to accurately define a biopsy target in the more metabolically active areas of the tumor. Then, the serial biopsy coordinates are transferred onto the stereotactic PET images for further correlation of tumor metabolism and histopathology.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Neurosurgery 1992-98 October 1992, Volume 31, Number 4 792 Positron Emission Tomography-Guided Stereotactic Brain Biopsy Technical Note
RESULTS Case reports Stereotactic PET was performed in 11 patients. Among them, 7 had hypermetabolic areas that were used to guide the biopsy. The technique described for PET-guided stereotactic biopsy is illustrated in the following two cases. Patient 1, a 52-year-old right-handed man, initially referred to our institution because of diffuse headaches and seizures. His neurological examination appeared normal. CT scan of the brain revealed an illdefined lesion in the left subcortical parietotemporal region. A stereotactic brain biopsy was scheduled. CT and PET were performed as described in the "Methods." PET revealed an area of increased metabolism in the left parietal lobe (Fig. 2A), which was used as a first biopsy target. A second target was selected in an hypometabolic area. Both targets were located in the tumor, controlled by projecting them on the stereotactic CT scan (Fig. 2B). Pathological examination revealed anaplastic astrocytoma in samples of both tracts. Patient 2, a 53-year-old right-handed man, who underwent an extensive resection of a right cerebellar low-grade astrocytoma 9 years previously. Adjuvant radiation therapy focused on the posterior fossa (5500 rads) was given. His neurological status improved and remained stable until recently. When he was admitted to our institution, CT and MRI scans of the brain showed a lesion located in the right part of the pons and in the right middle cerebellar peduncle. A stereotactic brain biopsy was scheduled. The stereotactic base ring was attached to the patient's skull in an inverted position. CT, vertebral DA, and PET were performed as described in "Methods." The PET study investigated the region of the brain located under the plane of the base ring. PET showed an area of increased metabolism set as the target (Fig. 3). Biopsy was performed using the transcerebellar route. Pathological examination revealed several malignant foci within a low-grade astrocytoma. Accuracy of technique Using the phantom containing the 11 radioactive targets (Fig. 4), the differences between PET and actual stereotactic coordinates were found to be (means ± SD): 1.03 ± 0.8 mm, 1.36 ± 0.59 mm, and 1.50 ± 0.91 mm for the x, y, and z axes, respectively. R values for correlation between PET-evaluated and actual coordinates were 0.999, 0.999, and 0.996 for the x, y, and z axes, respectively. DISCUSSION This report illustrates a method developed to integrate PET data into the procedure used for planning and performing stereotactic brain biopsies.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Accuracy of technique A hot-spot phantom simulating three-dimensional distribution of targets has been specifically designed for testing the accuracy of the PET system in calculating stereotactic coordinates (Fig. 1). The targets are made of 5-ml syringes filled with 0.5 ml of a [18F]fluoride solution; diameter and height of the cylindric targets are 8.4 mm and 9 mm, respectively. The stereotactic frame was fixed on the phantom and 11 targets were spirally placed on the phantom base in order to test the PET localization accuracy in the center as well as in the periphery of the stereotactic frame. Stereotactic PET was performed as described above. For each target, the plane with the highest radioactive count rate was selected and a pixel
located in the center of the target was interactively pointed at for calculation of coordinates. Mean and standard deviation of the differences between PET and actual x, y, and z stereotactic estimates of the 11 radioactive targets were calculated. Correlation between PET-evaluated and actual values was calculated for x, y, and z coordinates.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
correct the emission scan for attenuation. FDG is automatically synthesized (18), using the procedure described by Hamacher et al. (12). Radiochemical purity of the tracer is assessed by radio-thin layer chromatography (silica plates, acetonitrile/water: 95:5) and is higher than 99%. The patient is injected intravenously with 6 to 9 mCi FDG and maintained in the darkened room with no verbal stimulation. Images used for stereotactic calculation are acquired between 40 and 60 minutes after injection of the tracer. The PET camera allows simultaneous acquisition of 15 slices approximately 6.5 mm thick. Except for posterior fossa lesions, the most caudal slice is adjusted to the plane of the stereotactic ring. The eight planar landmarks generated by the radioactive fiducials allow rapid and accurate calculation of the zero point of the stereotactic system and of the coordinates of regions of interest using the PET processing software and customized personal computer spreadsheet software (4). Selective cerebral digital angiogram (DA) or plain anteroposterior and lateral skull x-rays are also performed, using the corresponding localizers. A personal computer and a digitizer are used to transfer coordinates from other image modalities onto the DA or the x-rays. The surgical planning begins with analysis of the PET images. If PET-FDG reveals a region of increased metabolism within the lesion boundaries, the plane that best displays the hypermetabolism is selected and a pixel located in the center of this zone is interactively pointed at on visual inspection. The coordinates of this pixel are then calculated and set as a target for biopsy. The target is then projected onto the corresponding stereotactic CT slice and, when performed, on anteroposterior and lateral DA, to control the reliability and the safety of the target and the trajectory. In patients in whom there is no obvious hyper-metabolic region on the stereotactic PET, surgery is planned using CT data only. The biopsy procedure was performed under general anesthesia with endotracheal intubation. Serial stereotactic biopsies were obtained with a sidecutting cannula, following the technique described by Kelly et al. (15). Postoperatively, biopsy sites in all patients were recorded and transferred onto the stereotactic PET for further correlation between tumor metabolism and histological results.
Research Grant 3.4565.88 from the National Fund for Scientific Research, Belgium. Received, December 3, 1991. Accepted, April 17, 1992. Reprint requests: Marc Levivier, M.D., Department of Neurosurgery, U.L.B.-Erasme Hospital, 808, Route de Lennick, B-1070 Brussels, Belgium.
1.
2.
3.
4.
5.
6.
7.
8.
9. 10.
ACKNOWLEDGMENTS The authors thank Dr. Robert E. Burke, Department of Neurology, College of Physicians and Surgeons of Columbia University, New York, for critical reading of the manuscript. This project was supported by Research Grant 9.4503.87 from the Belgian National Lottery and
11.
Alavi J, Alavi A, Chawluk J, Kushner M, Powe J, Hickey W, Reivich M: Positron emission tomography in patients with glioma. A predictor of prognosis. Cancer 62:10741078, 1988. Apuzzo MLJ, Chandrasoma PT, Cohen D, Zee C-S, Zelman V: Computed imaging stereotaxy: Experience and perspective related to 500 procedures applied to brain masses. Neurosurgery 20:930-937, 1987. Bergström M, Collins P, Ehrin E, Ericson K, Eriksson L, Greitz T, Halldin C, von Holst H, Langström B, Lilja A, Lundqvist H, Nagren K: Discrepancies in brain tumor extent as shown by computed tomography and positron emission tomography using [68Ga]EDTA, [11C]glucose, and [11C]methionine. J Comput Assist Tomogr 7:1062-1066, 1983. Bidaut L, Levivier M, Hildebrand J, Luxen A, Goldman S: Multimodality tumor evaluation through stereotactic correlation between XCT and PET. Proceedings of the 2nd European Workshop on FDG in Oncology, Heidelberg, Germany, December 9-11, 1991. Chandrasoma PT, Smith MM, Apuzzo MLJ: Stereotactic biopsy in the diagnosis of brain masses: Comparison of results of biopsy and resected surgical specimen. Neurosurgery 24:160-165, 1989. Coleman RE, Hoffman JM, Hanson MW, Sostman HD, Schold SC: Clinical application of PET for the evaluation of brain tumors. J Nucl Med 32:616-622, 1991. Derlon JM, Bourdet C, Bustany P, Chatel M, Theron J, Darcel F, Syrota A: [11C]Lmethionine uptake in gliomas. Neurosurgery 25:720-728, 1989. Di Chiro G: Positron emission tomography using [18F] fluorodeoxyglucose in brain tumors. A powerful diagnostic and prognostic tool. Invest Radiol 22:360-371, 1986. Di Chiro G: Which PET radiopharmaceutical for brain tumors? J Nucl Med 32:1346-1348, 1991. Feiden W, Steude U, Bise K, Gündisch O: Accuracy of stereotactic brain tumor biopsy: Comparison of the histologic findings in biopsy cylinders and resected tumor tissue. Neurosurg Rev 14:51-56, 1991. Glantz MG, Hoffman JM, Coleman RE, Friedman AH, Hanson MW, Burger PC, Herndon JE II, Meisler WJ, Schold SC Jr:
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
REFERENCES: (1-21)
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
PET acquisition in stereotactic conditions is easily performed, does not exceed more than 90 minutes, and does not cause any major additional discomfort to the patient. In our hands, data acquisition and target selection with PET appears as straightforward as with CT scanning, including that for posterior fossa lesions, as seen with Patient 2. The accuracy of target calculation is within PET resolution as demonstrated using the phantom and we always found a close correspondence of PET- and CTlocated targets. In the present protocol, PET sections are made of 2.591 × 2.591 mm pixels and are 6.75 mm thick; the precision of the target definition corresponds to a volume of 45 µl, which is about 5 times the volume of the biopsy specimen. Kelly (14) reported that 1.5 to 5 mm thick slices are adequate for CT-guided stereotactic brain biopsies, depending on the size of the lesion. PET cameras of the new generation will acquire sections of 3 mm thick, which should allow more accurate PET-guided stereotactic brain biopsies. PET is an imaging technique that provides functional data that are independent from CT or MRI anatomical images. The type of information obtained with PET mostly depends on the radiotracer selected. In this study, we have performed stereotactic PET with FDG because of the large amount of information available on the use of this marker for tracing brain tumors (9). PET-FDG is helpful in assessing the degree of malignancy and the prognosis of brain neoplasms independently of the histological grading (1,8) . It can also help in differentiating between the effects of treatments and in assessing tumor persistence, progression, and recurrence (6,11). Therefore, the present technique has been developed to take advantage of PET-FDG data in the stereotactic sampling of intracerebral tumors. In addition, we may anticipate that this method will allow us to accurately compare metabolic information obtained with PETFDG with data provided by other imaging modalities and neuropathology. This might eventually help in a better understanding of information provided by PET and answer the question of whether or not PET imaging adds to the diagnostic yield of stereotactic biopsies when glial tumors are suspected. This technique is also applicable to PET with other radiotracers, such as L-[11C]methionine (17), 2-[18F]fluoro-L-tyrosine, and other labeled amino acids that may be more suitable for the delineation of brain tumors or the study of their specific metabolism (3,7,13,21) . At the present stage, our data do not establish the role for PET-guided stereotactic brain biopsies in clinical routine. Studies comparing stereotactic biopsies done with and without PET guidance are needed to evaluate whether this imaging modality adds to the diagnostic yield.
14. 15.
16.
17.
18.
19. 20.
21.
COMMENTS Computed tomographic (CT)-guided stereotactic biopsy is anatomically very accurate. Unfortunately, there is often poor correlation between CT depiction of the lesions and the precise intralesional location of
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
13.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
12.
Identification of early recurrence of primary central nervous system tumors by [18F]fluorodeoxyglucose positron emission tomography. Ann Neurol 29:347-355, 1991. Hamacher K, Coenen HH, Stöcklin G: Efficient stereospecific synthesis of no-carrieradded 2- [18F]fluoro-2-deoxy-D-glucose using amino polyether supported nucleophilic substitution. J Nucl Med 27:235-238, 1986. Hübner KF, Purvis JT, Mahaley SM, Robertson Jr. JT, Rogers S, Gibbs WD, King P, Partain CL: Brain tumors imaging by positron emission computed tomography using 11C-labelled amino acids. J Comput Assist Tomogr 6:544-550, 1982. Kelly PJ: Tumor Stereotaxis. Philadelphia, WB Saunders, 1991. Kelly PJ, Daumas-Duport C, Kispert DB, Kall BA, Scheithauer BW, Illig JJ: Imaging-based stereotactic serial biopsies in untreated intracranial glial neoplasms. J Neurosurg 66:865-874, 1987. Lunsford LD: Diagnosis and treatment of mass lesions using the Leksell stereotactic system, in Lunsford LD (ed): Modern Stereotactic Neurosurgery. Boston, Martinus Nijhoff Publishing, 1988, pp 145-168. Mosskin M, von Holst H, Bergström M, Collins VP, Eriksson L, Johnström P, Norén G: Positron emission tomography with 11Cmethionine and computed tomography of intracranial tumors compared with histopathological examination of multiple biopsies. Acta Radiol 28:673-681, 1987. Padgett HC, Schmidt DG, Luxen A, Bida GT, Satyamurthy N, Barrio JR: Computercontrolled radiochemical synthesis: A chemistry process control unit for the automated production of radiochemicals. Appl Radiat Isot 40:433-445, 1989. Russell DS, Rubinstein LJ: Pathology of Tumours of the Nervous System. London, Edward Arnold, 1989, ed 5, pp 83-161. Sturm V, Pastyr O, Schlegel W, Scharfenberg H, Zabel H-J, Netzeband G, Schabbert S, Berberich W: Stereotactic computer tomography with a modified RiechertMundinger device as the basis for integrated stereotactic neuroradiological investigations. Acta Neurochir (Wien) 68:11-17, 1983. Wienhard K, Herholz K, Coenen HH, Rudolf J, Kling P, Stöcklin G, Heiss WD: Increased amino acid transport into brain tumors measured by PET of L-(2-18F)fluorotyrosine. J Nucl Med 32:1338-1346, 1991.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
Figure 1. PET data acquisition using the phantom. The radioactive targets are made of 5-ml syringes filled with 0.5 ml of the [18F]fluoride solution. The phantom is attached to the base ring of the stereotactic system, which is fixed to the clamp specifically designed to secure the head ring to the Siemens couch. The four localizers containing tubing filled with the [18F]fluoride solution are also fixed to the base ring.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
Figure 2. Stereotactic PET scan (A) and contrastenhanced CT scan (B) in Patient 1. The center of the hypermetabolic area on PET-FDG was visually defined and used as a target for biopsy (cross). Coordinates were calculated from the PET data and secondarily projected onto the CT scan (×). Note that the target selected on PET is within the limits of the lesion ill-defined on CT scan. The white circles were superimposed on the radioactive dots generated by the fiducials, which cannot be visualized with the color scale used in the present PET image.
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
Figure 3. Stereotactic PET scan (A) and contrastenhanced CT scan (B) in Patient 2. The center of the hypermetabolic area on PET-FDG was visually defined and used as a target for biopsy (cross). Coordinates were calculated from the PET data and secondarily projected onto CT scan (×).
Redistribution of this article permitted only in accordance with the publisher’s copyright provisions.
Downloaded from https://academic.oup.com/neurosurgery/article-abstract/31/4/792/2752109 by Universitaetsbibliothek Muenchen user on 02 February 2019
Figure 4. Scattergrams showing the correlation between actual (abscissa) and PET-calculated (ordinate) values for x, y, and z coordinates. Note the absence of distortion for high values of x and y corresponding to the periphery of the stereotactic frame. The inset shows the mean ± standard deviation for the absolute values of the differences between PET and actual stereotactic coordinates (mm). These values are within our PET scan resolution.
