Journal of Radiosurgery, Vol. 3, No. 4, 2000

Radiosurgery for Intracranial Arteriovenous Malformations in Children1 Beatriz E. Amendola, M.D., F.A.C.R.,2,4 Aizik Wolf, M.D.,2 Sammie R. Coy, Ph.D.,2 Steven DePrima, M.D.,2 and Marco A. Amendola, M.D.3

Background: The objective of this retrospective study is to determine the value of radiosurgery in the management of arteriovenous malformations (AVM) in the pediatric age group. Methods: From January 1994 through January 1999, thirty-one children with arteriovenous malformations (AVMs) were treated with radiosurgery. All patients were treated on an outpatient basis at the same institution by the same team. The Leksell Gamma Knife unit was used. Workup included angiography, MRI, and MRA. Follow-up ranged from 7 months to 67 months, with a median of 33 months. Minimum doses of radiation, depending on the size of the lesion, ranged from 20 Gy to 25 Gy. Treatment volumes for all the vascular malformations ranged from 0.6 cc to 17 cc with a mean volume of 4.7 cc. The mean number of isocenters was 4.8. Results: Total obliteration of AVM nidus was obtained in 22 of 31 (71%) patients, while 9 patients had partial obliteration. Stabilization of the benign lesions was obtained in all the patients treated. None had rebleeding after the procedure and, as of this writing, no patient required retreatment. Conclusion: Radiosurgery is an effective noninvasive and safe therapeutic modality for the management of vascular malformations independent of location, size, or grade. KEY WORDS: Benign intracranial lesions; radiosurgery; γ-Knife, AV malformations.

INTRODUCTION

ization and delivery of high-dose radiation to the volume of interest, while minimizing radiation to the surrounding normal tissue structures. This represents an excellent advantage of precision radiotherapy to be used in the pediatric population. In addition, radiosurgery is a noninvasive method, which allows most patients to be treated in a single outpatient visit without general anesthesia. It is effective not only for superficial, but also deep-seated lesions. Radiation therapy acts in the same manner when treating vascular malformations and benign tumors by obliteration of small blood vessels and decreasing circulation.

Radiosurgery is a well-established modality of treatment for the management of small vascular malformations and other benign intracranial lesions for adults (1–5). It is less used in children due to reluctance of the physicians to treat young patients with ionizing radiation. Therefore, intracranial lesions in the pediatric age group are usually managed by neurosurgical techniques. Recent advances in the hardware and software of treatment planning systems of different radiosurgical units, led by the γ-knife, have allowed for precise target local1

Paper presented at the International Society of Pediatric Oncology (SIOP), Montreal September 14–18, 1999. 2 Miami Neuroscience Center, 5000 University Drive, 298012 Coral Gables, Florida 33143. 3 Department of Radiology, University of Miami/Jackson Memorial Hospital, Miami, Florida 33124. 4 To whom all correspondence should be addressed at 270 Veleros Court, Coral Gables, Florida 33143.

MATERIALS AND METHODS This is a retrospective review of thirty-one children with intracranial AVMs treated at the Miami Neuroscience Center in Coral Gables, Florida from January 1994 to January 1999 inclusive. There were 14 females 159 1096-4053/00/1200-0159$18.00/0 © 2000 Plenum Publishing Corporation

160

Amendola, Wolf, Coy, DePrima, and Amendola

and 17 males with ages ranging from 7 years to 19 years with a median age of 12 years (Table I). The same team composed of a neurosurgeon, physicist, radiation oncologist, and neuro-oncology nurse treated all patients. All patients were treated with the Leksell Gamma Knife unit. The steps of the procedure have been previously described (6). Workup in all the patients included MRI and MRA and conventional catheter angiography in order to establish the diagnosis, as well as for treatment planning purposes. Patients were treated under local anesthesia and/or intravenous conscious sedation. Six of 31 children with AVMs had previously been treated with surgery and 4 with VP shunt. Intracranial bleeding, headaches, and seizures were the most common manifestations for these patients (Table II). The Spetzler–Martin grading (7) was used in all AVMs (Table IV). This grading system was originally developed to estimate the difficulty of surgical resection of AVMs. It assigns a score to each of three

Table I.

Number Sex/age 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 a

M/7 M/8 F/9 F/9 M/9 M/9 M/9 F/10 F/11 M/11 F/11 M/12 M/12 M/12 F/12 M/12 M/13 M/13 F/14 M/15 M/15 F/17 M/17 F/17 F/18 F/18 M/18 F/18 F/19 F/19 M/19

Dose 25 Gy 23 21 20 23 20 23 25 25 25 25 20 22 22 23 23 25 20 22 23 23 23 22 23 23 25 23 24 25 23 21

Vol/(cc) Grade 1.2 1.4 4.8 12.3 12.8 17.0 2.7 2.2 0.6 0.6 0.6 8.7 0.9 6.3 5.2 2.3 1.3 10.7 5.7 1.7 6.2 5.4 3.6 0.65 1.5 1.6 3.4 3.5 1.7 5.2 13.8

III III II III III IV III I II III II IV I IV II II IV IV III II IV III III III III III II II III III III

Obliteration Previous rate (%) Rxa 100 100 100 80 95 100 100 100 100 100 100 80 100 100 90 100 100 95 100 100 100 75 100 100 90 100 100 80 90 100 100

VP, Ventriculoperitoneal shunt; S, surgery, craniotomy.

VP

S

S

S;VP

S S;VP

Table II. Vascular Malformations Presentation AVM Intracranial bleeding Headache Seizures Dizziness Visual symptoms Precocious puberty Focal deficit

18 23 10 5 0 1 7

features: (1) nidus size (<3 cm = 1), (3 cm to 6 cm = 2), (>6 cm = 3), (2) pattern of venous drainage, either superficial (with all the drainage from the AVM being through the cortical venous system) = 0, or with a deep venous system component = 1, and (3) location of the AVM in an eloquent = 1 or in a noneloquent = 0 area of the brain. The scores are then added up so that AVMs can be classified into one of five grades. Grade I malformations are small, superficial, and located in noneloquent cortex. Grade V lesions are large, deep, and situated in neurologically critical areas. Grade VI are inoperable lesions corresponding to extremely large, diffuse AVMs that are dispersed through critical neurologically eloquent areas, or malformations with a diffuse nidus that encompasses critical structures, such as the hypothalamus or brain stem. In our series, 68% of the patients (21 of 31) fell into the Spetzler–Martin Grade III or IV category. Regarding location in the brain, 19 AVMs were lobar, 7 were thalamic, and 5 were in the basal ganglia. AVM volumes were less than 3 cc in 15 patients, between 3 cc and 6 cc in 8, and larger than 6 cc in 8 patients. Minimum doses ranged from 20 Gy to 25 Gy to the 50% isodose line. Treatment volumes for the AVMs ranged from 0.6 cc to 17 cc with a mean volume of 4.7 cc. Volumes of treatment included the nidus of the AVM. Mean number of isocenters was 4.8 (Table III). Patients underwent angiograms, MRIs, and MRAs to determine volume of treatment (Fig. 1). A single dose of precisely targeted irradiation was delivered with the γ-knife. Patients are being followed every 6 months with MRI and then with angiograms if the AVM appears obliterated by MRI until the 100% obliteration rate is documented. Table III. γ-Knife Technique Vascular malformations (AVM)

S

Minimum dose range (Gy) Mean volume (cc) Mean number of isocenters

20–25 (AVM) 4.7 (AVM) 4.8 (AVM)

Radiosurgery of Pediatric Intracranial AVMs Table IV. AVM Characteristics Spetzler–Martin Grade I II III IV Total:

No. Patients 2 8 15 6 31

RESULTS The endpoints of the study were to determine obliteration rate for the AVMs and stabilization of the lesions in the other children. Twenty-two of 31 patients with AVMs (71%) had successful obliteration of the nidus on follow-up angiograms within 2 years (Fig. 1). Of the remaining 9 children, 2 patients had 95%, 3 had 90%, 3 had 80%, and 1 patient had 75% obliteration respectively. They continue to be followed. Two of the patients had neurological deficit prior to γ-knife treatment

161 from intracranial bleeding, but none experienced worsening after radiosurgery. None of the patients experienced rebleeding after the procedure. Twenty-eight of 31 patients are doing well, attending school or college, and have normal family life. One patient was lost to follow-up 3 years post-treatment. Two patients are stable with neurological deficit that was present before γ-knife. One of the patients is a 9-year-old mentally retarded boy with a large AVM (12.8 cc, Grade III) of the basal ganglia that rebled after original surgery 6 years prior to γ-knife. Another patient is an 18-year-old girl with a basal ganglia AVM that had her first episode of intracranial bleeding in 1990, resulting in hemiparesis. She suffered a recurrence in 1996 and was treated with γ-knife; her deficit has not progressed with a 90% occlusion rate on follow-up angiogram. Chi-square analysis and Fisher’s Exact test (SAS software package, SAS Institute, Cary, NC) were used to determine adverse factors in the obliteration rate. Neither the grade, size, or location of the AVMs were found to adversely contribute to occlusion of the vessels.

Fig. 1. AVM in a 9 year old girl treated with γ-knife. On the left side an MRA shows the AVM (within box) on the day of treatment with γ-knife. Inset shows the 25% and 50% isodose lines. A minimum dose of 21 Gy was delivered to a 4.8 cc target volume. On the right side, a follow-up MRA obtained 9 months later documents total obliteration of the AVM.

162 DISCUSSION Stereotactic radiosurgery is the precise delivery of a single fraction of high-dose ionizing radiation to an image-defined target. With the development of CT and especially MRI, aiding in precise three-dimensional treatment planning, radiosurgery is being increasingly used for a variety of brain lesions, including pediatric benign brain tumors and vascular malformations (1–6). AVMs were one of the first indications for radiosurgery because they can be visualized with cerebral angiography and they are the most common clinically symptomatic cerebrovascular malformations. The lesions represent direct dysplastic vascular communications between arterial and venous channels without an intervening normal capillary network. The abnormal vessels at the site of the arteriovenous communication or nidus replace the normal arterioles and capillaries with a low-resistance, high-flow vascular bed. This results in excessive blood through the nidus and delivery of increased blood volume under high velocity and pressure into the cerebral venous system (7, 8). The long-term prognosis for patients with untreated AVMs is grim. The most dangerous complication is cerebral hemorrhage, which occurs as first presentation in about one half of the patients. The ongoing risk of intracranial bleed from an AVM is approximately 2% to 4% per year (9–12). The risk of death associated with initial AVM rupture is about 10%, which increases with each subsequent hemorrhage. A rebleeding rate of approximately 6% for the first year follows each episode of hemorrhage. After the first year, the rate decreases to that of unruptured AVMs—2% to 4% each year. Although only one quarter of all AVMs hemorrhage within the first 15 years of life, children are in greater danger than adults, since they have before them the period of life of highest risk for AVM rupture during the ages of 15 years to 40 years (7, 8). The goal of treatment is to eliminate the risk of bleeding with the minimal therapy-related complications. Results obtained by microsurgery with complete excision and interventional neuroradiology with endovascular embolization have improved steadily during the last 20 years (13–15). Radiosurgery can also afford a high rate of success in terms of obliteration rate. The best results for AVM obliteration occur when the whole nidus is irradiated and not just the feeding vessels. In our patient population, we have used high-resolution MRI/MRA, as well as catheter angiography, to delineate the nidus of the lesions and optimize the treatment planning protocol. In general, patients with small AVMs located in noncritical brain regions (Spetzler–Martin Grades I or II) are advised to

Amendola, Wolf, Coy, DePrima, and Amendola undergo surgical excision or radiosurgery (2). Radiosurgery is especially useful for AVMs in deeper brain sites (the basal ganglia, internal capsule, thalamus or brain stem) and for critical lobar areas (sensorimotor or visual cortex) (10–15). In our series, the great majority of patients had lesions seated deep in the brain and 68% of them (21 of 31 patients) were graded III or IV according to the Spetzler–Martin classification. Appropriate selection of patients using an experienced multidisciplinary team with accurate imaging is essential. Radiosurgery is thought to produce thrombosis of the AVM, by inducing a pathological process in the nidus with endothelial cell proliferation leading to gradual thickening of the vessels, eventually resulting in luminal closure (10, 11). From the radiobiological standpoint, fractionated irradiation is unlikely to cause complete AVM obliteration because its doses are not high enough in comparison to radiosurgery. Therefore, AVMs are typically treated with radiosurgery using a highly focused, single high-dose fraction delivered either by modified linear accelerators (LINAC), cyclotron, or γ-knife devices. The latent period (1–3 years) for angiographic obliteration, during which the patient remains at risk for hemorrhage, is the major drawback of radiosurgery as compared to neurosurgical treatment. Recent reports have shown that the annual risk of AVM bleeding is either unchanged during the latent interval (16, 17) or decreased during the first 6 months after radiosurgery (18). It is attractive to speculate that such protective effect has manifested itself in our group of patients, since we have not encountered any episode of rebleeding. However, our numbers are relatively small and our follow-up period not long enough to make a definitive statement at this time. Using a LINAC system, Loeffler et al. reported a 2-year complete obliteration rate of 73% with a dose of 15 Gy to 25 Gy generally applied to the 80% to 90% isodose line surrounding the nidus (19). Friedman and Bova (20) reported an incidence of complete thrombosis of 81% (21 of 25 eligible patients) with doses of between 20 Gy and 25 Gy. In a series of 180 patients, Colombo et al. (21) reported a 1-year thrombosis rate of 46% and a 2-year rate of 80%. In their group of patients treated with LINAC, Engenhart et al. (22) documented, with angiography, a complete overall obliteration rate of 72% above a threshold dose of 18 Gy. After 3 years, the obliteration rates were 83% with volumes of less than 4.2 cc, 75% with volumes of up to 33.5 cc, and 50% with volumes of up to 113 cc. In their series, the rebleeding rate was 2% during the first 2 years after therapy. Using γ-knife radiosurgery, Steiner (3) reported AVM obliteration in 79% of cases after 2 years. With a

Radiosurgery of Pediatric Intracranial AVMs similar unit, Yamamoto et al. (23), achieved a 64% thrombosis rate after 2 years and a 73% rate after 5 years in Japanese patients. In the Mayo Clinic series of Pollock et al. (13), 75% of patients had AVMs of Grade III Spetzler–Martin or greater. Of their 97 patients who underwent follow-up angiography 2 years or more after a single radiosurgical procedure, 72 patients (74%) had complete obliteration of their AVM. Our series have shown comparable results in a pediatric age population with an angiographically documented complete obliteration rate of 71% (22 of 31 patients) during the first 2 years after radiosurgery. The remaining 9 patients had partial obliteration and continue to be followed with MRI/MRA. They will undergo angiography once their MRA studies suggest total obliteration of the AVM. Regarding γ-knife-related complications, we have not yet encountered any mortality or significant morbidity; however, our follow-up time is relatively short. All our patients received prophylactic steroid and anticonvulsive medication immediately prior and during the radiosurgical procedure, which may explain these excellent results. It must be emphasized that radiosurgery can be safely used even in large arteriovenous malformations. Review of the literature demonstrates the value of radiosurgery for large AVMs with complication rates between 1%–10% (21–24); however, the data of Miyawaki (25), which is of special interest, demonstrated excellent local control and obliteration rate with minimal postradiosurgical complications. Risk of central nervous system parenchymal injury or second malignancy caused by radiation therapy has been estimated to be less than 1 or 2% (2). Since 1980, morbidity and mortality rates for patients who have undergone surgical resection of arteriovenous malformations (AVMs) of the brain vary markedly among series. Morbidity rates ranging from 7.8% to 30% and mortality rates from 0% to 12.5% had been reported in the literature (26). Factors that may explain these different results include the recent changes in operative techniques and postoperative care, as well as differences in patient selection. Recently, Morgan et al. (26) reported delayed neurological deficits in 5% of patients (10/200) after recovering from anesthesia and surgery. The delayed deficit was due to hemorrhage in 4 of the 10 patients that had undergone resection of an AVM. An additional 4 patients developed a delayed deficit as a result of vasospasm. They found the size of the lesion was a factor in the complication rate, with lesions 4 cm in diameter contributing to their increase. These results underscore the relatively noninvasive nature and the lack of acute complications related to the radiosurgical management of AVMs. Long-term follow-up will be needed to

163 assess for delayed complications and late effects of radiosurgery in this particular age group.

CONCLUSIONS In our group of 31 children with AVMs, γ-knife radiosurgery achieved total obliteration of the lesions in 22 patients (71%) and partial obliteration in the remaining 9, with stabilization in all 4 patients with cavernous angiomas. These results were obtained with minimal morbidity with all patients being treated as outpatients. We recommend that radiosurgery be used as an effective, noninvasive primary therapeutic modality for the management of vascular malformations in children independent of location, size, or grade.

REFERENCES 1. Corn BW, Curran WJ Jr, Shrieve DC, Loeffler JS: Stereotactic radiosurgery and radiotherapy: New developments and new directions. Semin Oncol 29:1707–1714, 1997 2. Tsao MN, Wara WM, Larson DA: Radiation therapy for benign central nervous system disease. Semin Rad Oncol 9:120–133, 1999 3. Steiner L: Radiosurgery in cerebral arteriovenous malformations. In Cerebrovascular surgery. JM Fein, ES Flamm, (eds). New York, Springer-Verlag, 1985, pp. 1161–1251 4. Maesawa S, Kondiziolka D, Lunsford LD: Stereotactic radiosurgery for management of deep brain cavernous malformations. Neurosurg Clin NA 10:503–511, 1999 5. Elisevich K, Ratkewicz A, Lownie S: Radiosurgery of deep arteriovenous malformations of the brain. Neurosurg Quart 9:33–48, 1999 6. Coy SR, Houdek PV: Radiosurgery treatment planning. Semin Rad Oncol 5:213–219, 1995 7. Spetzler RF, Martin NA: A proposed grading system for arteriovenous malformations. J Neurosurg 65:476–483, 1986 8. Osborn AG: Intracranial Vascular Malformations. In Diagnostic neuroradiology. DG Osborn (ed). St. Louis, Mosby, 1994, pp. 284–329 9. Morgan MK, Johnston IH, Hallinan JM, et al: Complications of surgery for arteriovenous-malformations of the brain. J Neurosurg 78:176–182, 1993 10. Ogilvy CS: Radiation therapy for arteriovenous malformations: A review. Neurosurgery 26:725–735, 1990 11. Schneider BF, Eberhard DA, Steiner LE: Histopathology of arteriovenous malformations after gamma knife radiosurgery. J Neurosurg 87:352–357, 1997 12. Ondra SL, Troupp H, George ED, Schwab K: The natural history of symptomatic arteriovenous malformations of the brain: A 24-year follow-up assessment. J Neurosurg 73:387–391, 1990 13. Pollock BE, Gorman DA, Schomberg PJ, Kline RW: The Mayo Clinic Gamma Knife experience: Indications and initial results. Mayo Clin Proc 74:5–13, 1999 14. Heros RC, Korosue K, Diebold PM: Surgical excision of cerebral arteriovenous malformations: Late results. Neurosurgery 26:570– 578, 1990 15. Sasaki T, Kurita H, Saito I, et al: Arteriovenous malformations in the basal ganglia and thalamus: Management and results in 101 cases. J Neurosurg 88:285–292, 1998

164 16. Friedman WA, Blatt D, Bova FJ, et al: The risk of hemorrhage after radiosurgery for arteriovenous malformations. J Neurosurg 84:912–919, 1996 17. Pollock BE, Flickinger JC, Lunsford LD, et al: Hemorrhage risk after stereotactic radiosurgery of cerebral arteriovenous malformations. Neurosurgery 38:652–659, 1996 18. Karlsson B, Linquist C, Steiner L: Effect of Gamma Knife surgery on the risk of rupture prior to AVM obliteration. Minim Invasive Neurosurg 39:21–27, 1996 19. Loeffler JS, Alexander E III, Siddon RL, et al: Stereotactic radiosurgery for intracranial arteriovenous malformations using a standard linear accelerator. Int J Radiat Oncol Biol Phys 17:673–677, 1989 20. Friedman WA, Bova FJ: Linear accelerator radiosurgery for arteriovenous malformations. J Neurosurg 77:832–841, 1992 21. Colombo F, Pozza F, Chierego E, et al: Linear accelerator radiosurgery of cerebral arteriovenous malformations: An update. Neurosurgery 34:14–21, 1994

Amendola, Wolf, Coy, DePrima, and Amendola 22. Engenhart R, Wowra B, Debus J, et al: The role of high-dose, single-fraction irradiation in small and large intracranial arteriovenous malformations. Int J Radiat Onc Biol Phys 30:521–529, 1994 23. Yamamoto M, Jimbo M, Kobayashi M, et al: Long-term results of radiosurgery for arteriovenous malformation: neurodiagnostic imaging and histological studies of angiographically confirmed nidus obliteration. Surg Neurol 37:219–230, 1992 24. Karlsson B, Lindquist C, Steiner L: Prediction of obliteration after Gamma Knife surgery for cerebral arteriovenous malformations. Neurosurgery 40:425–431, 1997 25. Miyawaki L, Dowd C, Wara W, et al: Five year results of LINAC radiosurgery for arteriovenous malformations: Outcome for large AVMS. Int J Radiat Onc Biol Phys 44:1089–1106, 1999 26. Morgan MK, Sekhon L, Finfer S, Grinnell V: Delayed neurological deterioration following resection of arteriovenous malformations of the brain. J Neurosurg 90:695–701, 1999