Radiosurgery of cerebral arteriovenous malformations with the dynamic stereotactic irradiation

In!. _I. Radiatmn Oncology Biol. Phys.. Vol. Printed in the U.S.A. All rights reserved.

0360-3016/90 $3.00 + .oO Copyright 0 1990 Pergamon Press plc

19, pp. 775-782

0 Technical Innovations and Notes RADIOSURGERY OF CEREBRAL ARTERIOVENOUS MALFORMATIONS WITH THE DYNAMIC STEREOTACTIC IRRADIATION LUIS SOUHAMI, ERVIN

B. PODGORSAK,

M.D.,* ANDRE PH.D., AND

OLIVIER,

F.C.C.P.M.,*

M.D.,

MARINA

PH.D.,

F.R.C.S.(C),+

PLA, M.Sc.,

M.C.C.P.M.*

G. BRUCE PIKE, M.E.E.+

McGill University, Montrkal, Canada From December 1986 through December 1988, 33 patients with inoperable arteriovenous malformation (AVM) were treated in our center with the dynamic stereotactic radiosurgery, which uses a standard 10 MV isocentric

linear accelerator. There were 18 females and 15 males with a median age of 26 years (range: 9-69) and a median follow-up time of 16 months (range: 7-32). The arteriovenous malformation volumes treated ranged from 0.2 to 42 cm3. The prescribed doses at the isocenter varied from 50 to 55 Gy and were given as a single fraction in the majority of the patients (31/33). Late complications consisting of intracranial bleeding and/or hemiparesis were observed in three patients. To date, 21 patients underwent repeat angiographic studies at 1 year post-treatment. A complete obliteration of the lesion was achieved in 38% of these patients. For the patients whose arteriovenous malformation nidus was covered by a minimum dose of 25 Gy, the total obliteration rate was 61.5% (g/13), whereas none of the patients who had received less than 25 Gy at the edge of the nidus obtained a total obliteration. Our preliminary analysis at 1 year post-radiosurgery reveals results comparable to those previously reported for other radiosurgical techniques for the same follow-up period. Stereotactic radiosurgery, Radiosurgery, Arteriovenous malformation, Dynamic rotation, Radiosurgery with linear accelerator. INTRODUCXION

Since the pioneering work of Leksell in 195 1 ( 15), radiosurgery has become an attractive form of therapy for intracranial arteriovenous malformations (AVM) (29, 30) and other selected intracranial lesions (6, 11, 16). Initially, radiosurgery was performed with heavy charged particle beams from cyclotrons (9, 10, 13) and later by the Gamma unit* (32). The well proven efficacy of radiosurgery has in recent years led to the development of linear accelerator (linac) based radiosurgical techniques (1, 4, 7, 8, 18, 26), making radiosurgery potentially available to most major radiation oncology centers. Several centers have developed or are developing linacbased techniques. It is estimated that by early 1990, approximately 30-40 linac-based facilities will be operational in the U.S. (12). The linac-based radiosurgical techniques

Presented at the 3 1st Annual Meeting of the American Society for Therapeutic Radiology and Oncology, San Francisco, CA, 1-6 October, 1989. * Dept. of Radiation Oncology. + Dept. of Neurosurgery. Reprint requests to: L. Souhami, M.D., Department of Radiation Oncology, Montreal General Hospital, 1650 Cedar Ave., Montreal, QuCbec, H3G IA4 Canada.

reported so far are: (a) single plane rotation (8), (b) multiple non-coplanar converging arcs ( 1, 4, 7, 1S), and (c) dynamic rotation (25, 26). A comparison study among the high energy photon beam radiosurgical techniques has been published recently (27). Several important criteria should be fulfilled before any new radiosurgical technique is used in clinical practice, including high spatial and numerical accuracy of dose delivery to the target, steep dose fall-off outside the target volume, and knowledge of dose distribution within the target volume. Recently, Podgorsak et al. (28) have studied the adequacy of linacs in performing radiosurgery. They have shown that two of the existing linac-based radiosurgical techniques (multiple non-coplanar converging arcs and dynamic rotation) meet these criteria and can be considered an adequate and less expensive option to the Gamma unit and/or proton beam radiosurgery. In

Acknowledgements-The authors wish to thank Drs. G. Bertrand and R. Leblanc for helpful discussions and referral of some of the patients and Ms. Jennifer Manal for dedicated secretarial work. Accepted for publication 29 March 1990. * Leksell Gamma unit, Electa Instrument AB, Stockholm, Sweden.

776

I. J. Radiation Oncology 0 Biology 0 Physics

this paper we report our initial results in the first 33 patients with AVM treated with the dynamic rotation at McGill University in Montreal. METHODS

AND MATERIALS

From December 1986 through December 1988,33 patients with AVM underwent stereotactic radiosurgery in our center. All patients were treated with the dynamic rotation, described in detail previously (2526). A 10 MV isocentric linac* with a remotely controlled couch rotation is used as the source of radiation. The gantry rotates from 30” to 330”, while simultaneously the couch rotates from 75” to -75” (Fig. 1). The circular beam converges at the target volume, but because of the simultaneous couch and gantry rotations, the entrance beam never coincides with an exit beam, yielding a uniform dose distribution in the target volume and a reasonably sharp dose fall-off

September 1990, Volume 19, Number 3

outside the target volume. Target localization, treatment set up, and patient immobilization during the treatment are accomplished with a stereotactic frame developed locally (2 1, 22), and compatible with computerized tomography scan (CT), magnetic resonance imaging (MRI), and digital subtraction angiography (DSA). All patients underwent MRI and DSA studies before radiosurgery to define the location and size of the AVM as well as its relationship to sensitive anatomical structures. MRI exams were performed on a 1.5 Tesla whole body system and consisted of short repetition time spin echo (T 1-weighted) acquisitions in the transverse, sagittal, and coronal orientations. The body coil was used in all studies since the stereotactic frame does not easily fit within a standard head coil. However, with slice thicknesses of 7.5 mm, good quality images were obtained with one signal measurement. We have achieved a localization accuracy off 1 mm using DSA and transverse MR images

Fig. 1. The dynamic stereotactic radiosurgical procedure starts with the couch at 75” and the gantry at 30”, as shown in Fig. IA. During treatment, the couch rotates 150” from +75’ to -75”, while the gantry simultaneously rotates 300” from 30” to 330”. Thus, each degree of couch rotation corresponds to two degrees of gantry rotation. Several sucessive positions through which the couch and gantry move during the complete radiosurgical procedure are shown, starting (A) with the gantry and couch angles of 30” and +75’, respectively, through (B) 90” and +45”, respectively, (C) 180” and O”, respectively, (D) 270” and -45”, respectively, and (E) stopping at 330” .and -75”, respectively.

* Clinac- 18, Varian Associates,

Palo Alto, CA.

Dynamic stereotactic radiosurgery 0 L. SOUHAMI efal.

(in plane); however, our sagittal and coronal MR images have been shown to be geometrically less reliable (23). DSA provides a good definition of the dynamic compartments of the AVM, namely the arterial feeders, the nidus, and the draining veins. MRI gives the picture of the bulk of the AVM without differentiating between arteries, capillaries, or veins. It delineates, however, the topographic information concerning vital cerebral structures neighboring the AVM (20). Following treatment, MRI was repeated every 3 months and DSA was obtained after 12 months or sooner if the MRI study demonstrated a major change in the lesion signal. A dedicated 3-dimensional treatment planning system developed at McGill was used for the patient dose distribution calculation. The system was verified experimentally and described in detail elsewhere (24). It was originally implemented on a large VAX computer and, more recently, a PC-based version has been used and integrated within a complete stereotactic image analysis system, capable of processing stereotactic CT, MR, and DSA images. Figure 2 shows lateral DSA as well as coronal and transverse MR diagnostic images and typical 3-dimensional isodose distributions superimposed onto these images. Specially made 10 cm thick lead collimators are used to obtain the small circular fields (5 mm to 30 mm diameter at SAD = 100 cm). The field diameter has been arbitrarily defined as the separation at 90% on the stationary beam dose profile measured at the depth of dose maximum in a water phantom. Once the patient set up is completed, it takes about 25-30 minutes to deliver a target dose of 50 Gy. Our patient group consisted of 18 females and 15 males with a median age of 26 years (range: 9-69). All patients were assessed by a radiosurgical team consisting of a neurosurgeon, a radiation oncologist, a neuro-radiologist, and a physicist. None of the patients were asymptomatic prior to radiosurgery. Intracranial bleeding had occurred in 22 patients and in 19 it was the presenting symptom. Seven patients presented with seizures, three with headaches without evidence of intracranial bleeding, and four with other neurological symptoms. Prior surgical intervention to resect the AVM was attempted in five patients. In another patient three embolization procedures were carried out without success. The median follow-up time postradiosurgery is 16 months (range: 7-30). The 21 patients for whom follow up DSA is reported in this paper received doses of 50 or 55 Gy prescribed at the isocenter (100%). The choice between 50 or 55 Gy essentially depended on the size and/or location of the malformation. A single fraction was given to the majority of the patients (92%) with only two patients receiving fractionated treatment of 2 fractions each. The field diameters ranged from 8 mm to 25 mm. The 21 patients fall into three categories according to the dose delivered to the edge of the AVM nidus: (a) in seven patients the nidus was covered by the 90% isodose contour (45 or 50 Gy), (b) in six it was covered by the 50% contour (25 or

711

Fig. 2. Lateral DSA images and transverse and coronal MRI scans showing a right supra sylvian region AVM fed by branch of the right middle cerebral artery (A, C, E), and the radiosurgery treatment plans (B, D, F) for such AVM. Isodose distributions calculated with the McGill stereotactic treatment planning system. The 90% isodose contour covers the spherical target volume with a diameter of 5 mm. The isodose contours displayed are 90%, 50%, and 10%.

27.5 Gy), and (c) in the remaining eight patients the periphery of the lesion was covered by lower than 50% isodose contours (dose < 25 Gy). RESULTS Up to the time of this analysis, 2 1 patients had angiographies repeated at about 12 months following radiosurgery. Table 1 lists the clinical characteristics, the treatment parameters, the complications and percent AVM obliteration obtained for these patients, whereas Tables 2 and 3 summarize the treatment results. In 38% of the patients (8/21) the AVM was completely obliterated. A further five patients (24%) had a partial obliteration, between 50% and 99%, whereas eight other patients (38%) had an obliteration between 0 and 49%. Of the 13 patients

no.

35/F 25/F 39/M 18/F 26/M 20/F 21/F 24/F 22/F 50/M 56/M 22/F 11/F 27/F 25/F 32/M 16/M 28/F 44/M 38/M 9/F

Age/sex

* As of July 1989.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21

FT.

midbrain caudate nucleus splenial thalamus occipital parieto-occipital splenial temporal fronto-parietal parietal central thalamus sylvian internal capsule sylvian central occipital frontal frontal parieto-occipital splenial

Location 4.5 0.5 3.7 0.4 0.3 0.2 6.3 3.5 0.3 8.2 11.5 2.6 0.5 16.3 0.9 13.0 42.0 8.2 22.4 5.6 0.4

Volume (cm31 10 10 15 10 10 10 20 15 20 20 20 10 10 25 8 10 20 17.5 15 15 5

Field diameter (mm) 27.5 X 2 55 50 55 55 55 55 55 50 55 55 50 50 27.5 X 2 50 55 55 50 55 50 50

<50 90 50 90 90 90 <50 50 90 50 50 <50 90 <50 90 <50 <50 50 <50 50 <50

Isodose contour at edge (W) ~25 50 25 50 50 50 ~25 27.5 45 27.5 27.5 <25 45 <25 45 ~25 <25 25 ~25 25 <25

Dose at edge (GY) 50 100 75 20 100 90 25 90 100 100 100 20 100 60 0 0 0 100 0 100 0

Obliteration (%I

parameters, and results at 1 year postradiosurgery

Dose at isocenter (GY)

Table 1. Patient characteristics, treatment

none none none bleeding none none none none none none hemiparesis none none hemiparesis/bleeding none none none none none none none

Complications

30 29 27 27 26 24 23 22 22 22 21 18 17 16 16 15 15 14 13 13 11

(month:

Pf-XiOd

Follow- /

Dynamic stereotactic radiosurgery 0 L. Table 2. Treatment

results at l-year follow-up Obliteration

of

AVM

Number of patients

100%

50-99%

O-49%

21

8/21 (38%)

5/21 (24%)

8/21 (38%)

in whom the angiographically evident nidus received a minimum dose of 25 Gy, eight (6 1.5%) achieved a complete obliteration at 1 year, and in three other patients an almost complete obliteration (~90%) was seen. On the other hand, when the AVM nidus received less than 25 Gy, the result was an incomplete obliteration or no change at all in the AVM (8 patients). Thus, it appears that the whole AVM nidus should be encompassed by at least 25 Gy to obtain an optimal therapeutic effect. These results, of course, must be viewed cautiously because of the small number of patients studied. Only three patients (#I, #3, and #4) had a repeated angiography at 24 months posttreatment. None of these three patients have yet achieved a full obliteration, although all have progressed to at least a 75% reduction in the size of their original lesion. Figure 3 illustrates the pre- and post-treatment angiograms of a 50-year-old male who presented with severe headache, dizziness, and blurred vision. He had previously bled from a large left parietal AVM (Fig. 3a-b). He received 55 Gy in a single fraction with a 20 mm collimator. Six months after the treatment, the AVM was completely obliterated (Fig. 3C-D). Late complications were seen in three patients. An 18year-old female hemorrhaged 7 months after treatment of an AVM in the anterior thalamus with a dose of 55 Gy delivered with a 10 mm collimator. She had a rapid and complete neurological recovery. Repeated angiography 24 months post-treatment showed a 75% obliteration. The patient has had no further bleeding and remains symptom-free. Two other patients developed a hemiparesis. A 56-yearold male with a large AVM in the central region developed hemiparesis 9 months after receiving 55 Gy with a 20 mm collimator. CT scan revealed extensive vasogenic edema. The patient was started on oral steroid and showed a gradual improvement in his neurological deficit. Angiographic studies demonstrated that his AVM was fully obliterated. A 27-year-old female who presented with a

SOUHAMI

779

et al.

seizure resulting from a large AVM in the left internal capsule was treated with a cone size of 25 mm and a dose of 55 Gy delivered in 2 fractions, 2 weeks apart. Six months post-treatment she experienced a progressive right sided weakness and did not respond to steroid therapy. Imaging studies done at the time of her neurological deficit showed a 60% obliteration of the AVM. There was no evidence of intracranial bleeding; however, extensive edema was demonstrated. Her neurological picture remained stable, but 3 months later she sustained a massive intracranial hemorrhage and expired soon after that. Thrombosed vessels within the AVM with some recanalization were seen at autopsy. There was no evidence of necrosis in the target volume or elsewhere in the brain,

DISCUSSION Radiosurgery can now be considered a valid option for treatment of AVM’s that are inaccessible or otherwise unsuitable for surgery. The proven effectiveness of this method stimulated the recent development of new techniques based on isocentric linacs (1, 4, 7, 8, 18, 26). The adequacy of linac-based radiosurgical techniques in fulfilling the standards on mechanical precision and reproducibility has been questioned (14). In terms of the numerical dose delivery to the target, dose fall-off outside the target volume, isodose distributions, and doses of radiation to sensitive organs, the linac-based multiple converging arcs and the dynamic rotation technique are similar to the Gamma unit (27, 28). Our preliminary treatment results for patients with AVM at 1 year post-treatment are encouraging and comparable to results reported for treatment with the Gamma unit for the same time period (29,30). Similar obliteration rates have also been reported by other centers using the linac-based multiple converging arcs techniques (2, 4). The optimal irradiation dose needed to provide the maximal therapeutic effect continues to be the subject of discussion. Steiner (30), in his series of more than 300 patients treated with the Gamma unit, used single doses varying between 30 and 125 Gy, usually prescribed at the center of the cluster of pathologic vessels. Although not entirely clear from Steiner’s publications, it appears that the edge of the lesion was enclosed completely by the 50% isodose surface. He has suggested that a minimum dose

Table 3. Obliteration rate for 2 1 patients as a function of AVM nidus coverage by various isodose contours and dose prescriptions at the edge of the nidus Obliteration Nidus coverage 90% contour (45 or 50 Gy) 50% contour (25 or 27.5 Gy) ~50% contour (~25 Gy) Note: Dose at isocenter

was 50 or 55 Gy.

Number of patients

100%

so-99%

O-49%

I 6

417 (57%) 416 (67%)

l/7 (14%) 2/6 (33%)

2/7 (29%) O/6

8

018

2/8 (25%)

6/8 (75%)

780

I. J. Radiation Oncology 0 Biology 0 Physics

September 1990, Volume 19, Number 3

Fig. 3. Cerebral angiograms (frontal and lateral views) of a 50-year-old male with an AVM (volume 8.2 cm3) in the left parietal region fed by posterior branches of the middle cerebral artery. A and B, before irradiation; C and D, 6 months after radiosurgery, show complete obliteration of the malformation.

of 20 to 25 Gy delivered to the periphery of the AVM gives a good chance for a full obliteration (3 1). Colombo et al. (3, 5) have treated 97 patients with AVM using the multiple, non-coplanar arc irradiation technique. Doses varied from 18.7 to 40 Gy prescribed at the isodose contour which would coincide with the periphery of the target (90% to 60%). They suggested that for AVMs up to 15 mm in diameter doses of 22 to 30 Gy should be delivered at the edge of the lesion. They made no comments on the effectiveness of one dose level over the other. Kjellberg et al. (9, 10) reported on 439 patients treated with proton beam therapy. They were able to correlate an increased number of complications with the total dose and field size. They concluded that for larger AVM’s, which require larger fields, the prescribed dose should be reduced to avoid radiation-induced complications. Although they were able to estimate the dose for a 1% risk of cerebral necrosis for proton beams with diameters from 7 mm to 50 mm, the lowest, most effective dose to obliterate completely different AVM sizes was not determined. Recently, Levy et al. (17) reported on a group of children and adolescents with AVM treated with stereotactic heavycharged-particle Bragg peak radiosurgery, using the helium

ion beam. These authors initially used maximum central (100%) doses biologically equivalent to 34.5 Gy. For larger AVM’s or those near sensitive regions, the prescribed dose ranged from 26.9 to 30.8 Gy. The periphery of the AVM was included in the 90% isodose contour. Subsequently, the prescribed dose was lowered, in a stepwise fashion, to 19.2 Gy. They have not observed significant differences in the obliteration rate among the different groups and currently are testing doses of 11.5 to 19.2 Gy with the aim of determining the lowest yet still effective dose. In our own series, we did not observe any difference in the obliteration rate between doses of 25 or 50 Gy, provided the delivered dose encompassed the AVM nidus in its entirety. It seems, therefore, that 25 Gy given at the periphery of the AVM is sufficient to obliterate the malformation completely. The question of the maximum required target dose to achieve the dose of 25 Gy at the edge of the nidus, however, remains unresolved. To us it now appears better to prescribe the 25 Gy at the edge of the nidus with the 90% isodose contour (maximum target dose = 27.5 Gy) rather than with the 50% isodose contour (maximum target dose = 50 Gy). This implies the use of larger cones but makes the dose distribution inside the

Dynamic stereotactic radiosurgery 0 L.

nidus much more uniform and gives a better dose fall-off outside the target volume. Moreover, the treatment time is cut in half, which is of considerable benefit to the patient and staff. Our present treatment policy is to cover the whole nidus with a dose of 25 Gy prescribed at the 90% isodose surface. The definition of the target volume appears to be of crucial importance as well. Steiner (29) grouped his patients in four categories depending upon the volume irradiated: (a) 166 patients in whom the irradiation field completely covered the cluster of pathologic vessels (group A), (b) 19 cases with only partial coverage of the pathologic vessels (group B), (c) three patients where all feedings arteries were included (group C), and (d) four cases where only some of the multiple feeders were irradiated (group D). The complete obliteration rates at 2 years were as follows: 86.5%, O%, 33%, and 0% for groups A, B, C, and D, respectively. Groups B, C, and D are somewhat small for meaningful conclusions, but similar observations have been made by the investigators from the Lawrence Berkeley Laboratory (17). Thus, it appears that a partial cov-

SOUHAMI

et nl.

781

erage of the AVM will result in less than optimal obliteration and that the entire cluster of pathologic vessels corresponding to the nidus should be included in the irradiated target volume. The complications encountered in this study are comparable to other reported series. Radiologic abnormalities (vasogenic edema, mass effect) with progressive neurologic deficit occurred in two of our patients (6%). The pathophysiological mechanisms involved in the development of these complications following radiosurgery are not entirely clear, but probably are a consequence of obliteration of small vessels within or near the malformation leading to necrosis of dependent tissue ( 19). Several questions remain unanswered in radiosurgery, and further experience with longer follow-up is required. A number of parameters need to be properly established to facilitate meaningful comparisons among different centers and techniques. These include imaging evaluation of the AVM, dose prescription, target volume definition. criteria to assess response to treatment, size and location of the malformation, and previous therapy.

REFERENCES 1. Betti, 0. 0.; Derechinsky, V. E. Hyperselective encephalic irradiation with linear accelerator. Acta Neurochir. 33(Suppl.):385-390; 1984. 2. Betti, 0.0.; Munari, C.; Rosler, R. Stereotactic radiosurgery with the linear accelerator: treatment of arteriovenous malformations. Neurosurgery 24:3 1 l-32 1; 1989. 3. Colombo, F.; Benedetti, A.; Casentini, L.; Zanusso, M.; Pozza, F. Linear accelerator radiosurgery for arteriovenous malformations. Appl. Neurophysiol. 50:257-26 1; 1987. 4. Colombo, F.; Benedetii, A.; Pozza, F.; Avanzo, R. C.; Marchetti, C.; Chierego, G.; Zanardo, A. External stereotactic irradiation by linear accelerator. Neurosurgery 16: 154- 160; 1985. 5. Colombo, F.; Benedetti, A.; Pozza, F.; Marchetti, C.; Chierego, G. Linear accelerator radiosurgery of cerebral arteriovenous malformations. Neurosurgery 24:833-840; 1989. 6. Degerblad, M.; Rahn, T.; Bergstrand, G.; Thor&n, M. Longterm results of stereotactic radiosurgery to the pituitary gland in Gushing’s disease. Acta Endocrinol. 112:3 1O-3 14; 1986. 7. Hartmann, G. H.; Schlegel, W.; Sturm, V.; Kober, B.; Pastyr, 0.; Lorenz, W. J. Cerebral radiation surgery using moving field irradiation at a linear accelerator facility. Int. J. Radiat. Oncol. Biol. Phys. 11:1185-l 192; 1985. 8. Houdek, P. V.; Fayos, J. V.; Van Buren, J. M.; Ginsberg, M. S. Stereotactic radiotherapy technique for small intracranial lesions. Med. Phys. I2:469-472; 1985. 9. Kjellberg, R. N.; Davis, K. R.; Lyons, S. L.; Butler, W.; Adams, R. D. Bragg peak proton beam therapy for arteriovenous malformation of the brain. Clin. Neurosurg. 3 1:248290; 1983. 10. Kjellberg, R. N.; Hanamura, T.; Davis, K. R.; Lyons, S. L.; Adams, R. D. Bragg-peak proton-beam therapy of arteriovenous malformation of the brain. N. Engl. J. Med. 309: 269-274; 1983. 1 I. Kjellberg, R. N.; Shintani, A.; Frantz, A. G.; Kliman, B. Proton beam therapy in acromegaly. N. Engl. J. Med. 278: 689-695; 1968. 12. Larson, D. A. Review of linac-based stereotactic techniques in the U.S. In: Proceedings of the International Workshop

13.

14. 15. 16. 17.

18.

19.

20.

2 1.

22.

23.

24.

on Proton and Narrow Photon Beam Therapy, Oulu, Finland. 1989:83-85. Larsson, B.; Leksell, L.; Rexed, B.; Sourander, P.; Mair, W.; Anderson, B. The high energy proton beam as a neurosurgical tool. Nature 182: 1222- 1223; 1958. Leksell, D. G. Stereotactic radiosurgery. Present status and future trends. Neurolog. Res. 9:60-68; 1987. Leksell, L. The stereotactic method and radiosurgery of the brain. Acta Chir. Stand. 102:3 16-3 19, 195 1. Leksell, L. Stereotactic radiosurgery. J. Neurol. Neurosurg. Psych. 43:797-803; 1983. Levy, R. P.; Fabrikant, J. I.; Frankel, K. A.; Phillips, M. H.; Lyman, J. T. Stereotactic heavy-charged-particle Bragg peak radiosurgery for the treatment of intracranial arteriovenous malformations in children and adolescence. Neurosurgery 24:841-852; 1989. Lutz, W.; Winston, K. R.; Maleki, N. A system for stereotactic radiosurgery with a linear accelerator. Int. J. Radiat. Oncol. Biol. Phys. 14:373-381; 1988. Martins, A. N.; Johnston, J.; Henry, J. M.; Stoffel, T. J.; Chiro, G. Delayed radiation necrosis of the brain. J. Neurosurger. 47~336-345; 1977. Olivier, A.; de Lotbiniere, A.; Peters, T.; Pike, B.; Ethier. R.; Melanson, D.; Bertrand, G.; Podgorsak, E. Combined use of digital subtraction angiography and MRI for radiosurgery and stereoencephalography. Appl. Neurophysiol. 50: 92-99; 1987. Olivier, A.; Peters, T. M.; Bertrand, G. Stereotaxic systems and apparatus for use with MRI, CT and DSA. Appl. Neurophysiol. 48:94-96; 1986. Peters, T. M.; Clark, J. A.; Olivier, A.; Mawko, G.; Marchand, E.; Dieumegarde, M.; Muresan, L. V.; Ethier, R. Integrated stereotactic imaging with CT, MRI and DSA. Radiology 161:821-826; 1987. Peters, T. M.; Clark, J. A.; Pike, G. B.; Henri, C.; Collins, L.; Leksell, D.; Jeppsson, 0. Stereotactic neurosurgery planning on a PC based workstation. J. Digital Imaging 2(2): 75-81; 1989. Pike, B.; Podgorsak, E. B.: Peters, T. M.; Pla, C. Dose dis-

782

25.

26.

27.

28.

I. J. Radiation Oncology 0 Biology0 Physics tributions in dynamic stereotactic radiosurgery. Med. Phys. 14:780-789; 1987. Podgorsak, E. B.; Olivier, A.; Pla, M.; Hazel, J.; de Lotbimere, A.; Pike, B. Physical aspects of dynamic stereotactic radiosurgery. Appl. Neurophysiol. 50:263-268; 1987. Podgorsak, E. B.; Olivier, A.; Pla, M.; Lefebvre, P. Y.; Hazel, J. Dynamic stereotactic radiosurgery. Int. J. Radiat. Oncol. Biol. Phys. 14:115-125; 1988. Podgorsak, E. B.; Pike, G. B.; Olivier, A.; Pla, M.; Souhami, L. Radiosurgery with high energy photons: a comparison among techniques. Int. J. Radiat. Oncol. Biol. Phys. 16: 857-865; 1989. Podgorsak, E. B.; Pike, G. B.; Pla, M.; Olivier, A.; Souhami, L. Radiosurgery with photon beams: physical aspects and adequacy of linear accelerators. Radioth. Oncol. 17:349358; 1990.

September 1990, Volume 19, Number 3 29. Steiner, L. Treatment of arteriovenous malformations by radiosurgery. In: Wilson, C. B., Stein, B. M., eds. Intracranial arteriovenous malformations. Baltimore: Williams and Wilkins; 1984:295-3 13. 30. Steiner, L. Radiosurgery in cerebral arteriovenous malformations. In: Fein, J., Flamm, E., eds. Cerebrovascular surgery, Vol 4. New York: Springer-Verlag; 1986: 116 I- 12 15. 3 1. Steiner, L. Stereotactic radiosurgery with the cobalt 60 Gamma unit in the surgical treatment of intracranial tumors and arteriovenous malformations. In: Schmidek, H. H., Sweet, W. H., eds. Operative neurosurgical techniques, Vol. I. Philadelphia: W. B. Saunders; 1988:5 15-529. 32. Steiner, L.; Leksell, L.; Greitz, T.; Forster, D. M. C.; Backlund, E-O. Stereotaxic radiosurgery for cerebral arteriovenous malformations. Report of a case. Acta Chir. Stand. 138:459-464; 1972.